<?xml version="1.0" encoding="utf-8"?>
<feed xmlns="http://www.w3.org/2005/Atom"><title>VamimiAI guides</title><subtitle>How-tos for people who build automations for clients</subtitle><link href="https://www.vamimi.com/blog/feed.xml" rel="self"/><link href="https://www.vamimi.com/blog"/><id>https://www.vamimi.com/blog</id><updated>2026-11-16T00:00:00Z</updated><entry><title>User Flow Diagram: What It Is and How to Draw One</title><link href="https://www.vamimi.com/blog/user-flow-diagram-guide"/><id>https://www.vamimi.com/blog/user-flow-diagram-guide</id><published>2026-11-16T00:00:00Z</published><updated>2026-11-16T00:00:00Z</updated><author><name>VamimiAI team</name></author><summary>What a user flow diagram is, how it differs from a task flow and a user journey, and how to draw one in five steps, with a sign-up example.</summary><content type="html">&lt;p&gt;A user flow diagram shows the steps one person takes to reach one goal in a product: where they enter, each screen or action in order, each decision, and where the flow ends. To draw one, choose a single goal and a single kind of user, list the steps of the successful path, then add a branch for every point where something can go differently.&lt;/p&gt;
&lt;p&gt;Teams draw user flows before designing screens, because a missing step costs minutes to fix in a diagram and days to fix in a build.&lt;/p&gt;
&lt;h2 id=&quot;user-flow-task-flow-and-user-journey&quot;&gt;User flow, task flow and user journey&lt;/h2&gt;
&lt;p&gt;The three names are often mixed up.&lt;/p&gt;
&lt;div class=&quot;table&quot;&gt;&lt;table&gt;&lt;thead&gt;&lt;tr&gt;&lt;th&gt;&lt;/th&gt;&lt;th&gt;Shows&lt;/th&gt;&lt;th&gt;Branches&lt;/th&gt;&lt;th&gt;Typical use&lt;/th&gt;&lt;/tr&gt;&lt;/thead&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td&gt;Task flow&lt;/td&gt;&lt;td&gt;The steps of one task, one path&lt;/td&gt;&lt;td&gt;None&lt;/td&gt;&lt;td&gt;Agreeing the basic sequence&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;User flow&lt;/td&gt;&lt;td&gt;One goal, with entry points, decisions and error paths&lt;/td&gt;&lt;td&gt;Yes&lt;/td&gt;&lt;td&gt;Planning screens and logic&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;User journey map&lt;/td&gt;&lt;td&gt;A whole experience across channels and time, with thoughts and feelings&lt;/td&gt;&lt;td&gt;Not usually&lt;/td&gt;&lt;td&gt;Understanding the customer&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt;&lt;/div&gt;
&lt;p&gt;A user flow is the one engineering and design both build from. If your diagram has no decisions, it is a task flow; that is a fine start, and the steps below turn it into a user flow.&lt;/p&gt;
&lt;h2 id=&quot;step-1-choose-one-goal-and-one-user&quot;&gt;Step 1: Choose one goal and one user&lt;/h2&gt;
&lt;p&gt;Write the goal as something the person wants: &amp;quot;create an account&amp;quot;, &amp;quot;buy one item&amp;quot;, &amp;quot;invite a teammate&amp;quot;. Then say who they are and what they already have: a new visitor with no account is a different flow from a returning customer.&lt;/p&gt;
&lt;p&gt;One goal, one diagram. A flow that tries to show sign-up, checkout and settings together cannot be read by anyone.&lt;/p&gt;
&lt;h2 id=&quot;step-2-mark-the-entry-points-and-the-end&quot;&gt;Step 2: Mark the entry points and the end&lt;/h2&gt;
&lt;p&gt;People do not all arrive at the home page. List where this flow can start: an advertisement, an emailed invitation, a link from inside the product. Each is an entry point.&lt;/p&gt;
&lt;p&gt;Then define the end as a state, not a screen: &amp;quot;account exists and the person is signed in&amp;quot;. Knowing the end tells you when a branch is finished.&lt;/p&gt;
&lt;h2 id=&quot;step-3-draw-the-successful-path&quot;&gt;Step 3: Draw the successful path&lt;/h2&gt;
&lt;p&gt;Lay out the steps in a line, left to right. Use three kinds of shape and stay consistent:&lt;/p&gt;
&lt;ul&gt;&lt;li&gt;&lt;strong&gt;A screen or page&lt;/strong&gt; the person sees.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;An action&lt;/strong&gt; the person takes.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;A decision or check&lt;/strong&gt;, drawn as a diamond: a choice the person makes, or something the system verifies.&lt;/li&gt;&lt;/ul&gt;
&lt;p&gt;Label each with a few words. &amp;quot;Sign-up form&amp;quot;, &amp;quot;Submit details&amp;quot;, &amp;quot;Details valid?&amp;quot;.&lt;/p&gt;
&lt;h2 id=&quot;step-4-add-every-branch&quot;&gt;Step 4: Add every branch&lt;/h2&gt;
&lt;p&gt;This is where a user flow earns its keep. Go back to each decision and each action and ask what else can happen:&lt;/p&gt;
&lt;ul&gt;&lt;li&gt;The details are not valid.&lt;/li&gt;&lt;li&gt;The email is already registered.&lt;/li&gt;&lt;li&gt;The verification email is never opened.&lt;/li&gt;&lt;li&gt;The payment is declined.&lt;/li&gt;&lt;li&gt;The person closes the tab and comes back tomorrow.&lt;/li&gt;&lt;/ul&gt;
&lt;p&gt;Draw a line for each, and follow it until it rejoins the main path or reaches its own end. A check with only its &amp;quot;Yes&amp;quot; line drawn is the most common fault in a user flow, and it is the one that becomes a bug.&lt;/p&gt;
&lt;h2 id=&quot;step-5-review-it-with-design-and-engineering&quot;&gt;Step 5: Review it with design and engineering&lt;/h2&gt;
&lt;p&gt;Walk through the flow together, one line at a time. Designers will find screens that are missing; engineers will find states the diagram did not consider (&amp;quot;what if the link has expired?&amp;quot;). Add what they find. The flow is finished when nobody can name a case it does not cover.&lt;/p&gt;
&lt;h2 id=&quot;an-example-sign-up&quot;&gt;An example: sign-up&lt;/h2&gt;
&lt;ol&gt;&lt;li&gt;Landing page, then the sign-up form.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Details valid?&lt;/strong&gt; No: show what to fix and return to the form. Yes: send a verification email.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Link opened?&lt;/strong&gt; No: offer to resend. Yes: set up the workspace.&lt;/li&gt;&lt;li&gt;End: the dashboard, signed in.&lt;/li&gt;&lt;/ol&gt;
&lt;p&gt;VamimiAI&amp;#x27;s &lt;a href=&quot;/user-flow-diagram&quot;&gt;user flow diagram maker&lt;/a&gt; page shows this flow drawn, with both error paths. You can describe your own flow and ask for the error paths, or attach the requirements document and ask for the flow of one goal.&lt;/p&gt;
&lt;aside class=&quot;try-card&quot;&gt;&lt;p&gt;Try it in VamimiAI&lt;/p&gt;&lt;a href=&quot;/app?q=User%20flow%20diagram%20of%20signing%20up%20for%20a%20web%20app%3A%20landing%20page%2C%20sign-up%20form%2C%20email%20verification%2C%20workspace%20setup%2C%20dashboard%2C%20with%20the%20error%20paths&quot;&gt;&lt;q&gt;User flow diagram of signing up for a web app: landing page, sign-up form, email verification, workspace setup, dashboard, with the error paths&lt;/q&gt;&lt;span&gt;Try this prompt →&lt;/span&gt;&lt;/a&gt;&lt;small&gt;The prompt goes into the message box for you to edit; nothing is sent until you press Send.&lt;/small&gt;&lt;/aside&gt;
&lt;h2 id=&quot;mistakes-to-avoid&quot;&gt;Mistakes to avoid&lt;/h2&gt;
&lt;ul&gt;&lt;li&gt;&lt;strong&gt;Only the successful path.&lt;/strong&gt; Without branches it is a task flow, and the hard cases are still undecided.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Several goals in one diagram.&lt;/strong&gt; Split it. Link flows by ending one where the next begins.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Screens with no way out.&lt;/strong&gt; Every screen needs a line leaving it, even if it is &amp;quot;close&amp;quot;.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Too much interface detail.&lt;/strong&gt; Button colours and field labels belong in the design. The flow shows what happens, not what it looks like.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Never updated.&lt;/strong&gt; When the product changes, change the flow, or the next person plans from a picture of last year&amp;#x27;s product.&lt;/li&gt;&lt;/ul&gt;
&lt;p&gt;A user flow is a flowchart with a product-specific purpose, so any &lt;a href=&quot;/ai-flowchart-generator&quot;&gt;flowchart tool&lt;/a&gt; can draw one. When the question changes from &amp;quot;what does the person do?&amp;quot; to &amp;quot;which system calls which?&amp;quot;, a sequence diagram is the better picture.&lt;/p&gt;
&lt;h2 id=&quot;questions&quot;&gt;Questions&lt;/h2&gt;
&lt;h3 id=&quot;what-is-a-user-flow-diagram&quot;&gt;What is a user flow diagram?&lt;/h3&gt;
&lt;p&gt;A user flow diagram is a picture of the path a person takes through a product to reach one goal. It shows the entry point, each screen and action in order, each decision, the error paths and the end state.&lt;/p&gt;
&lt;h3 id=&quot;what-is-the-difference-between-a-user-flow-and-a-user-journey&quot;&gt;What is the difference between a user flow and a user journey?&lt;/h3&gt;
&lt;p&gt;A user flow covers one goal inside one product and shows its steps and decisions. A user journey map covers a longer experience across several channels and records what the person does, thinks and feels at each stage.&lt;/p&gt;
&lt;h3 id=&quot;what-should-a-user-flow-diagram-include&quot;&gt;What should a user flow diagram include?&lt;/h3&gt;
&lt;p&gt;The entry points, the steps of the successful path, every decision with a line for each outcome, the error paths, and a clearly defined end. Keep it to one goal for one kind of user.&lt;/p&gt;
&lt;h3 id=&quot;when-should-i-create-a-user-flow&quot;&gt;When should I create a user flow?&lt;/h3&gt;
&lt;p&gt;Before designing the screens, and again whenever the logic changes. It is the cheapest place to find a missing step or an unhandled error.&lt;/p&gt;
&lt;h3 id=&quot;how-detailed-should-a-user-flow-be&quot;&gt;How detailed should a user flow be?&lt;/h3&gt;
&lt;p&gt;Detailed enough that a designer knows which screens are needed and an engineer knows which cases to handle. Leave out visual detail such as layout and wording.&lt;/p&gt;</content></entry><entry><title>Troubleshooting Flowchart: How to Build One for Support</title><link href="https://www.vamimi.com/blog/troubleshooting-flowchart"/><id>https://www.vamimi.com/blog/troubleshooting-flowchart</id><published>2026-11-16T00:00:00Z</published><updated>2026-11-16T00:00:00Z</updated><author><name>VamimiAI team</name></author><summary>How to build a troubleshooting flowchart or decision tree for a support team, with an example, the order to ask questions in and when to escalate.</summary><content type="html">&lt;p&gt;A troubleshooting flowchart is a decision tree for one problem: it starts from the symptom the customer reports, asks one question at a time, and ends every path in a fix or an escalation. To build one, pick a single common problem, list its causes from most to least likely, turn each check into a question with a line for every answer, and test it on real tickets.&lt;/p&gt;
&lt;p&gt;A good one lets a new agent reach the same answer as your most experienced one.&lt;/p&gt;
&lt;h2 id=&quot;start-with-one-problem&quot;&gt;Start with one problem&lt;/h2&gt;
&lt;p&gt;Do not draw &amp;quot;technical support&amp;quot;. Draw &amp;quot;customer cannot log in&amp;quot;, &amp;quot;order has not arrived&amp;quot; or &amp;quot;printer will not print&amp;quot;. Choose by volume: the three or four problems that fill most of the queue are worth a flowchart each.&lt;/p&gt;
&lt;p&gt;Write the starting point in the customer&amp;#x27;s words. Agents match what they hear to the top of the chart, so &amp;quot;I can&amp;#x27;t get into my account&amp;quot; belongs there, not &amp;quot;authentication failure&amp;quot;.&lt;/p&gt;
&lt;h2 id=&quot;order-the-checks-likely-and-cheap-first&quot;&gt;Order the checks: likely and cheap first&lt;/h2&gt;
&lt;p&gt;List everything that can cause the problem. Then sort by two things: how often it is the cause, and how quick it is to check. A cause that explains half the cases and takes ten seconds to confirm goes first. A rare cause that needs an engineer goes last.&lt;/p&gt;
&lt;p&gt;This order matters more than anything else in the chart. With the common cause first, most contacts leave the tree after one or two questions.&lt;/p&gt;
&lt;h2 id=&quot;write-each-check-as-a-question-an-agent-can-answer&quot;&gt;Write each check as a question an agent can answer&lt;/h2&gt;
&lt;p&gt;Each decision should be one fact the agent can find out in seconds:&lt;/p&gt;
&lt;ul&gt;&lt;li&gt;&amp;quot;Account found?&amp;quot; works. &amp;quot;Is there a problem with the account?&amp;quot; does not.&lt;/li&gt;&lt;li&gt;&amp;quot;Is the power light on?&amp;quot; works. &amp;quot;Is the hardware working?&amp;quot; does not.&lt;/li&gt;&lt;/ul&gt;
&lt;p&gt;Avoid a question that needs judgement, and never put two questions in one. If the answers are not yes and no, name them on the lines: &amp;quot;Billing&amp;quot;, &amp;quot;Outage&amp;quot;, &amp;quot;How-to&amp;quot;.&lt;/p&gt;
&lt;h2 id=&quot;give-every-branch-an-end&quot;&gt;Give every branch an end&lt;/h2&gt;
&lt;p&gt;There are only two acceptable ways for a path to finish:&lt;/p&gt;
&lt;ul&gt;&lt;li&gt;&lt;strong&gt;A fix&lt;/strong&gt;, with the action written out: &amp;quot;Send password reset link&amp;quot;, not &amp;quot;resolve&amp;quot;.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;An escalation&lt;/strong&gt;, saying when and to whom: &amp;quot;Not solved after reset: pass to tier 2 with the account email and the error shown&amp;quot;.&lt;/li&gt;&lt;/ul&gt;
&lt;p&gt;A branch that simply stops is where agents start improvising. Follow every line and check that it reaches one of these two.&lt;/p&gt;
&lt;h2 id=&quot;an-example-customer-cannot-log-in&quot;&gt;An example: customer cannot log in&lt;/h2&gt;
&lt;ol&gt;&lt;li&gt;&lt;strong&gt;Account found?&lt;/strong&gt; No: help them create an account, then close. Yes: continue.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Account locked?&lt;/strong&gt; Yes: verify identity and unlock. No: send a password reset link.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Can they log in now?&lt;/strong&gt; Yes: close the ticket. No: escalate to tier 2.&lt;/li&gt;&lt;/ol&gt;
&lt;p&gt;Three questions, four ends, and nothing left to guess. VamimiAI&amp;#x27;s &lt;a href=&quot;/decision-tree-maker&quot;&gt;decision tree maker&lt;/a&gt; page shows this example drawn, and the prompt below starts one of your own.&lt;/p&gt;
&lt;aside class=&quot;try-card&quot;&gt;&lt;p&gt;Try it in VamimiAI&lt;/p&gt;&lt;a href=&quot;/app?q=Troubleshooting%20flowchart%20for%20a%20customer%20who%20cannot%20log%20in%3A%20account%20found%2C%20account%20locked%2C%20password%20reset%2C%20and%20escalation%20to%20tier%202%20if%20it%20still%20fails&quot;&gt;&lt;q&gt;Troubleshooting flowchart for a customer who cannot log in: account found, account locked, password reset, and escalation to tier 2 if it still fails&lt;/q&gt;&lt;span&gt;Try this prompt →&lt;/span&gt;&lt;/a&gt;&lt;small&gt;The prompt goes into the message box for you to edit; nothing is sent until you press Send.&lt;/small&gt;&lt;/aside&gt;
&lt;h2 id=&quot;escalation-rules-that-get-followed&quot;&gt;Escalation rules that get followed&lt;/h2&gt;
&lt;p&gt;Escalation is the part teams leave vague, and vague rules produce both kinds of mistake: tickets passed on too early, and customers kept too long at the first tier.&lt;/p&gt;
&lt;ul&gt;&lt;li&gt;&lt;strong&gt;State the condition.&lt;/strong&gt; &amp;quot;After two failed attempts&amp;quot;, &amp;quot;customer reports data loss&amp;quot;, &amp;quot;customer asks for a manager&amp;quot;.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Name the receiver.&lt;/strong&gt; A team or a role, not &amp;quot;escalate&amp;quot;.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;List what to hand over.&lt;/strong&gt; The account, the steps already tried and the exact error, so the next person does not ask the customer again.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Draw the hand-over as its own end&lt;/strong&gt;, visibly different from &amp;quot;solved&amp;quot;.&lt;/li&gt;&lt;/ul&gt;
&lt;h2 id=&quot;test-it-on-real-tickets&quot;&gt;Test it on real tickets&lt;/h2&gt;
&lt;p&gt;Take twenty closed tickets for this problem and walk each one through the chart.&lt;/p&gt;
&lt;ul&gt;&lt;li&gt;If a ticket&amp;#x27;s real cause is not in the chart, add a branch.&lt;/li&gt;&lt;li&gt;If most tickets pass through the same four questions before reaching the common fix, reorder.&lt;/li&gt;&lt;li&gt;If agents solved it a different way from the chart, find out why; one of the two is wrong.&lt;/li&gt;&lt;/ul&gt;
&lt;p&gt;Then give the chart to someone new and watch them use it without help. Every place they hesitate is a question to rewrite.&lt;/p&gt;
&lt;h2 id=&quot;keep-it-where-agents-work&quot;&gt;Keep it where agents work&lt;/h2&gt;
&lt;p&gt;A flowchart in a shared drive is not used during a live contact. Put it in the help centre article or the knowledge base page for that problem, as an image at the top. Give it an owner and a review date, and change it when the product changes.&lt;/p&gt;
&lt;p&gt;A flowchart is a picture to follow, not a script that advances by itself. If your team needs agents to click through one step at a time, that is a guided-workflow tool; draw and agree the tree first either way.&lt;/p&gt;
&lt;p&gt;For triage across many problem types, a wider &lt;a href=&quot;/process-flowchart-maker&quot;&gt;process flowchart&lt;/a&gt; with a lane for each tier shows who handles what.&lt;/p&gt;
&lt;h2 id=&quot;questions&quot;&gt;Questions&lt;/h2&gt;
&lt;h3 id=&quot;what-is-a-troubleshooting-flowchart&quot;&gt;What is a troubleshooting flowchart?&lt;/h3&gt;
&lt;p&gt;A troubleshooting flowchart is a diagram that leads from a reported symptom, through a series of checks, to a fix or an escalation. Each check is a question, and each answer has its own line to the next step.&lt;/p&gt;
&lt;h3 id=&quot;what-is-the-difference-between-a-troubleshooting-flowchart-and-a-decision-tree&quot;&gt;What is the difference between a troubleshooting flowchart and a decision tree?&lt;/h3&gt;
&lt;p&gt;In support work they are the same thing. &amp;quot;Decision tree&amp;quot; stresses that the chart branches at each question; &amp;quot;troubleshooting flowchart&amp;quot; stresses what it is used for.&lt;/p&gt;
&lt;h3 id=&quot;how-many-questions-should-a-troubleshooting-flowchart-have&quot;&gt;How many questions should a troubleshooting flowchart have?&lt;/h3&gt;
&lt;p&gt;As few as resolve the problem. If a common case needs more than four or five questions, reorder the checks so the likeliest cause comes first, or split the chart into two problems.&lt;/p&gt;
&lt;h3 id=&quot;when-should-a-ticket-be-escalated&quot;&gt;When should a ticket be escalated?&lt;/h3&gt;
&lt;p&gt;When a stated condition is met: the listed fixes did not work, the problem is outside the first tier&amp;#x27;s access, or the customer is at risk of losing data or money. Write the condition and the receiving team on the chart.&lt;/p&gt;
&lt;h3 id=&quot;can-i-build-a-troubleshooting-flowchart-from-an-existing-guide&quot;&gt;Can I build a troubleshooting flowchart from an existing guide?&lt;/h3&gt;
&lt;p&gt;Yes. In VamimiAI, attach the guide as a PDF, Word or text file and ask for the decision tree of the steps it describes, then check each branch against the guide.&lt;/p&gt;</content></entry><entry><title>SOP Flowchart: How to Turn a Procedure into a Flowchart</title><link href="https://www.vamimi.com/blog/sop-flowchart"/><id>https://www.vamimi.com/blog/sop-flowchart</id><published>2026-11-16T00:00:00Z</published><updated>2026-11-16T00:00:00Z</updated><author><name>VamimiAI team</name></author><summary>How to turn a standard operating procedure into an SOP flowchart, when a flowchart beats written steps, and how to keep the two in step.</summary><content type="html">&lt;p&gt;An SOP flowchart is a standard operating procedure drawn as a diagram: each instruction becomes a step, each &amp;quot;if&amp;quot; becomes a decision with a line for every answer, and each way the procedure can finish becomes an end. To make one, mark the actions and the conditions in the written SOP, draw them in order, and check that every decision has all of its exits.&lt;/p&gt;
&lt;p&gt;A flowchart does not replace the written procedure. It sits at the top of it and tells the reader where they are.&lt;/p&gt;
&lt;h2 id=&quot;when-an-sop-needs-a-flowchart&quot;&gt;When an SOP needs a flowchart&lt;/h2&gt;
&lt;p&gt;Written steps work well for a procedure that runs in a straight line: do this, then this, then this. They work badly as soon as the procedure branches. A reader in the middle of step 7 who meets &amp;quot;if the reading is out of range, see section 4.2, unless the unit was serviced this month&amp;quot; has to hold the whole document in their head.&lt;/p&gt;
&lt;p&gt;Draw a flowchart when the SOP has:&lt;/p&gt;
&lt;ul&gt;&lt;li&gt;&lt;strong&gt;Decisions.&lt;/strong&gt; Two or more places where the next step depends on a result.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Hand-offs.&lt;/strong&gt; Work that passes between people, shifts or departments.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Loops.&lt;/strong&gt; Steps that repeat until something passes.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Exceptions.&lt;/strong&gt; A &amp;quot;what to do if it goes wrong&amp;quot; section that is longer than the main steps.&lt;/li&gt;&lt;/ul&gt;
&lt;p&gt;A five-step procedure with no conditions does not need one. A numbered list is clearer.&lt;/p&gt;
&lt;h2 id=&quot;step-1-mark-the-actions-and-the-conditions&quot;&gt;Step 1: Mark the actions and the conditions&lt;/h2&gt;
&lt;p&gt;Read the SOP with two highlighters. With the first, mark every action: a verb that someone carries out (&amp;quot;weigh the sample&amp;quot;, &amp;quot;record the result&amp;quot;, &amp;quot;notify the supervisor&amp;quot;). With the second, mark every condition: any sentence with &amp;quot;if&amp;quot;, &amp;quot;unless&amp;quot;, &amp;quot;when&amp;quot;, &amp;quot;in case of&amp;quot;, &amp;quot;otherwise&amp;quot; or &amp;quot;until&amp;quot;.&lt;/p&gt;
&lt;p&gt;The actions become the steps of the flowchart. The conditions become its decisions.&lt;/p&gt;
&lt;h2 id=&quot;step-2-rewrite-each-condition-as-a-question&quot;&gt;Step 2: Rewrite each condition as a question&lt;/h2&gt;
&lt;p&gt;A decision in a flowchart is a question with short answers. Turn &amp;quot;if the temperature exceeds 8 °C, discard the batch&amp;quot; into &amp;quot;Temperature above 8 °C?&amp;quot; with two lines: &amp;quot;Yes&amp;quot; to &amp;quot;Discard the batch&amp;quot;, &amp;quot;No&amp;quot; to the next step.&lt;/p&gt;
&lt;p&gt;Now look for the answer the SOP does not mention. Procedures are usually written by someone describing what to do when a check fails, or when it passes, but not both. Every missing answer is a gap in the procedure itself, and finding those gaps is the most useful thing the flowchart does.&lt;/p&gt;
&lt;h2 id=&quot;step-3-draw-the-main-path-then-the-branches&quot;&gt;Step 3: Draw the main path, then the branches&lt;/h2&gt;
&lt;p&gt;Put the steps of a normal, successful run in a straight line from the start to the end. Then add each branch to the side. This keeps the common case easy to follow and the exceptions easy to find.&lt;/p&gt;
&lt;p&gt;If more than one role is involved, use a swimlane diagram: one lane for each role, each step in the lane of the person who does it. An operator should be able to read down their own lane.&lt;/p&gt;
&lt;p&gt;Use the SOP&amp;#x27;s own step numbers in the labels (&amp;quot;4. Record the result&amp;quot;) so that the reader can move between the diagram and the text.&lt;/p&gt;
&lt;h2 id=&quot;step-4-give-every-path-an-end&quot;&gt;Step 4: Give every path an end&lt;/h2&gt;
&lt;p&gt;Follow each line with your finger. Every path must reach an end: the procedure is complete, the batch is rejected, the supervisor takes over. A path that stops at a step with no line out of it is a place where someone will have to guess.&lt;/p&gt;
&lt;p&gt;Mark the ends that mean a failure or a hand-over differently from the end that means success, so the outcome of each path is clear at a glance.&lt;/p&gt;
&lt;h2 id=&quot;making-the-flowchart-from-the-document&quot;&gt;Making the flowchart from the document&lt;/h2&gt;
&lt;p&gt;If the SOP exists as a file, you do not have to redraw it by hand. VamimiAI&amp;#x27;s &lt;a href=&quot;/document-to-flowchart&quot;&gt;document to flowchart&lt;/a&gt; page describes how it works: attach the SOP as a PDF, Word, Excel, CSV or text file, ask for the flowchart, and the steps and decisions in the document are drawn in order. Logic checks then point at a decision with only one exit, a step that leads nowhere and a step nothing reaches.&lt;/p&gt;
&lt;aside class=&quot;try-card&quot;&gt;&lt;p&gt;Try it in VamimiAI&lt;/p&gt;&lt;a href=&quot;/app?q=Turn%20this%20SOP%20into%20a%20flowchart%20with%20a%20lane%20for%20each%20role%2C%20and%20show%20what%20happens%20when%20a%20check%20fails&quot;&gt;&lt;q&gt;Turn this SOP into a flowchart with a lane for each role, and show what happens when a check fails&lt;/q&gt;&lt;span&gt;Try this prompt →&lt;/span&gt;&lt;/a&gt;&lt;small&gt;The prompt goes into the message box for you to edit; nothing is sent until you press Send.&lt;/small&gt;&lt;/aside&gt;
&lt;p&gt;Whichever way you draw it, read the result against the document once, line by line. The flowchart is a second view of the procedure, and both must say the same thing.&lt;/p&gt;
&lt;h2 id=&quot;keep-the-sop-and-the-flowchart-in-step&quot;&gt;Keep the SOP and the flowchart in step&lt;/h2&gt;
&lt;p&gt;A flowchart that disagrees with the text is worse than no flowchart, because people follow the picture.&lt;/p&gt;
&lt;ul&gt;&lt;li&gt;Give both the same version number and date, and change them together.&lt;/li&gt;&lt;li&gt;Put the flowchart on the first page of the SOP, not in an appendix.&lt;/li&gt;&lt;li&gt;When the procedure changes, change the flowchart first. It shows at once whether the change leaves a path without an end.&lt;/li&gt;&lt;li&gt;Keep the editable diagram, not only an image, so the next change takes minutes.&lt;/li&gt;&lt;/ul&gt;
&lt;p&gt;For the method behind all of this, see the &lt;a href=&quot;/blog/process-mapping-guide&quot;&gt;process mapping guide&lt;/a&gt;.&lt;/p&gt;
&lt;h2 id=&quot;questions&quot;&gt;Questions&lt;/h2&gt;
&lt;h3 id=&quot;what-is-an-sop-flowchart&quot;&gt;What is an SOP flowchart?&lt;/h3&gt;
&lt;p&gt;An SOP flowchart is a diagram of a standard operating procedure. It shows the procedure&amp;#x27;s steps in order, its decisions with a line for each answer, and how the procedure ends, so that a reader can see the whole procedure on one page.&lt;/p&gt;
&lt;h3 id=&quot;does-a-flowchart-replace-the-written-sop&quot;&gt;Does a flowchart replace the written SOP?&lt;/h3&gt;
&lt;p&gt;No. The flowchart gives the overview and the route; the written steps give the detail of how to carry out each one. Use the same step numbers in both.&lt;/p&gt;
&lt;h3 id=&quot;what-symbols-does-an-sop-flowchart-use&quot;&gt;What symbols does an SOP flowchart use?&lt;/h3&gt;
&lt;p&gt;A rounded shape for the start and the ends, a rectangle for a step, a diamond for a decision and arrows for the order. Label every line that leaves a decision with its answer.&lt;/p&gt;
&lt;h3 id=&quot;how-long-should-an-sop-flowchart-be&quot;&gt;How long should an SOP flowchart be?&lt;/h3&gt;
&lt;p&gt;One page. If it does not fit, the SOP probably covers more than one procedure. Split it, and let a single box in the main flowchart stand for each sub-procedure.&lt;/p&gt;
&lt;h3 id=&quot;can-i-make-a-flowchart-from-a-pdf-of-our-sop&quot;&gt;Can I make a flowchart from a PDF of our SOP?&lt;/h3&gt;
&lt;p&gt;Yes. In VamimiAI, attach the PDF and ask for the flowchart of the procedure it describes. Then check the result against the document.&lt;/p&gt;</content></entry><entry><title>Process Mapping: How to Map a Business Process in 6 Steps</title><link href="https://www.vamimi.com/blog/process-mapping-guide"/><id>https://www.vamimi.com/blog/process-mapping-guide</id><published>2026-11-16T00:00:00Z</published><updated>2026-11-16T00:00:00Z</updated><author><name>VamimiAI team</name></author><summary>How to map a business process step by step, which kind of process map to use, an example, and the mistakes that make a process map useless.</summary><content type="html">&lt;p&gt;Process mapping is drawing one piece of work from the event that starts it to the result it produces: the steps in order, the decisions, and who does each part. To map a process, pick one process with a clear start and end, list its steps with the people who do them, draw the steps and decisions in order, then walk through the map with those people and correct it.&lt;/p&gt;
&lt;p&gt;The six steps below work on paper, on a whiteboard or in any diagram tool.&lt;/p&gt;
&lt;h2 id=&quot;what-a-process-map-is-and-which-kind-to-draw&quot;&gt;What a process map is, and which kind to draw&lt;/h2&gt;
&lt;p&gt;A process map is any picture of how work flows. Three kinds cover almost every need:&lt;/p&gt;
&lt;div class=&quot;table&quot;&gt;&lt;table&gt;&lt;thead&gt;&lt;tr&gt;&lt;th&gt;Kind&lt;/th&gt;&lt;th&gt;Shows&lt;/th&gt;&lt;th&gt;Use it when&lt;/th&gt;&lt;/tr&gt;&lt;/thead&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td&gt;Process flowchart&lt;/td&gt;&lt;td&gt;Steps and decisions in order&lt;/td&gt;&lt;td&gt;One person or one team does the work&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Swimlane diagram (cross-functional flowchart)&lt;/td&gt;&lt;td&gt;The same, divided into a lane for each role&lt;/td&gt;&lt;td&gt;Work passes between people or teams&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;High-level map&lt;/td&gt;&lt;td&gt;Five to seven big steps, no decisions&lt;/td&gt;&lt;td&gt;You need to agree on scope before the detail&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt;&lt;/div&gt;
&lt;p&gt;If you are unsure, start with a flowchart and add lanes when the question becomes &amp;quot;who does this?&amp;quot;.&lt;/p&gt;
&lt;h2 id=&quot;step-1-pick-one-process-and-its-boundaries&quot;&gt;Step 1: Pick one process and its boundaries&lt;/h2&gt;
&lt;p&gt;Name the process by its result (&amp;quot;pay a supplier invoice&amp;quot;, not &amp;quot;finance&amp;quot;). Then write down two things: the event that starts it and the result that ends it. &amp;quot;An invoice arrives by email&amp;quot; and &amp;quot;the supplier is paid and the invoice is filed&amp;quot; are boundaries; &amp;quot;invoicing&amp;quot; is not.&lt;/p&gt;
&lt;p&gt;A map without boundaries grows until nobody can read it.&lt;/p&gt;
&lt;h2 id=&quot;step-2-list-the-steps-with-the-people-who-do-them&quot;&gt;Step 2: List the steps with the people who do them&lt;/h2&gt;
&lt;p&gt;Ask the people who do the work, not the person who designed it. Write each step as a verb and a thing: &amp;quot;Check the invoice against the order&amp;quot;, &amp;quot;Approve the payment&amp;quot;. Beside each step, write who does it.&lt;/p&gt;
&lt;p&gt;Collect what really happens, including the workarounds. If a manager approves by replying to a chat message, that is the step.&lt;/p&gt;
&lt;h2 id=&quot;step-3-put-the-steps-in-order-and-find-the-decisions&quot;&gt;Step 3: Put the steps in order and find the decisions&lt;/h2&gt;
&lt;p&gt;Arrange the steps from the start event to the end result. Wherever the next step depends on something, you have a decision. Write it as a question with a short answer: &amp;quot;Does the invoice match the order?&amp;quot;.&lt;/p&gt;
&lt;p&gt;Every decision needs a line for each answer. The answer people forget is the unwelcome one: the invoice does not match, the approver is on holiday, the customer does not reply. Those branches are where the real process lives.&lt;/p&gt;
&lt;h2 id=&quot;step-4-draw-it&quot;&gt;Step 4: Draw it&lt;/h2&gt;
&lt;p&gt;Use the same few shapes throughout:&lt;/p&gt;
&lt;ul&gt;&lt;li&gt;A rounded shape for the start and for each end.&lt;/li&gt;&lt;li&gt;A rectangle for a step.&lt;/li&gt;&lt;li&gt;A diamond for a decision, with a label on each line that leaves it.&lt;/li&gt;&lt;li&gt;An arrow for &amp;quot;then&amp;quot;.&lt;/li&gt;&lt;/ul&gt;
&lt;p&gt;Keep the main path in a straight line, top to bottom or left to right, and let the exceptions branch off to the side. If work passes between people, divide the page into lanes and put each step in the lane of whoever does it.&lt;/p&gt;
&lt;p&gt;You can describe the process to VamimiAI&amp;#x27;s &lt;a href=&quot;/process-flowchart-maker&quot;&gt;process flowchart maker&lt;/a&gt; and get this drawing made for you, with the lanes if you name the roles.&lt;/p&gt;
&lt;aside class=&quot;try-card&quot;&gt;&lt;p&gt;Try it in VamimiAI&lt;/p&gt;&lt;a href=&quot;/app?q=Process%20map%20of%20paying%20a%20supplier%20invoice%2C%20with%20lanes%20for%20the%20requester%2C%20the%20approver%20and%20finance%2C%20including%20an%20invoice%20that%20does%20not%20match%20the%20order&quot;&gt;&lt;q&gt;Process map of paying a supplier invoice, with lanes for the requester, the approver and finance, including an invoice that does not match the order&lt;/q&gt;&lt;span&gt;Try this prompt →&lt;/span&gt;&lt;/a&gt;&lt;small&gt;The prompt goes into the message box for you to edit; nothing is sent until you press Send.&lt;/small&gt;&lt;/aside&gt;
&lt;h2 id=&quot;step-5-walk-through-it-with-the-people-who-do-the-work&quot;&gt;Step 5: Walk through it with the people who do the work&lt;/h2&gt;
&lt;p&gt;Take one real case from last week and follow it through the map, step by step. Then take a case that went wrong. Each time the map and the story disagree, the map is what you change.&lt;/p&gt;
&lt;p&gt;This review is the point of the exercise. People who have worked side by side for years often find they describe the same process differently.&lt;/p&gt;
&lt;h2 id=&quot;step-6-mark-what-to-fix-then-keep-the-map-alive&quot;&gt;Step 6: Mark what to fix, then keep the map alive&lt;/h2&gt;
&lt;p&gt;With an accurate map, look for:&lt;/p&gt;
&lt;ul&gt;&lt;li&gt;&lt;strong&gt;Hand-offs.&lt;/strong&gt; Every line that crosses from one lane to another is a place where work waits.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Loops.&lt;/strong&gt; A step that sends work back (&amp;quot;return for correction&amp;quot;) shows where information was missing earlier.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Decisions with no owner.&lt;/strong&gt; If nobody can say who decides, the decision is being made by whoever is least busy.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Steps nobody can explain.&lt;/strong&gt; Ask what would break if the step were removed.&lt;/li&gt;&lt;/ul&gt;
&lt;p&gt;Give the map an owner and a date. When the process changes, change the map the same day, or it becomes a picture of how things used to work.&lt;/p&gt;
&lt;h2 id=&quot;a-process-mapping-example&quot;&gt;A process mapping example&lt;/h2&gt;
&lt;p&gt;Take a purchase request as an illustration. The start event is an employee needing something; the end result is a paid invoice. Suppose that drawing it shows the manager approving before anyone checks the budget, so finance sometimes refuses an approved request and the requester starts again. Moving the budget check before the approval removes that loop. Nobody had noticed it, because each person saw only their own part.&lt;/p&gt;
&lt;p&gt;The &lt;a href=&quot;/swimlane-diagram-maker&quot;&gt;swimlane diagram maker&lt;/a&gt; page has a similar example with one lane for each role.&lt;/p&gt;
&lt;h2 id=&quot;mistakes-that-make-a-process-map-useless&quot;&gt;Mistakes that make a process map useless&lt;/h2&gt;
&lt;ul&gt;&lt;li&gt;&lt;strong&gt;Mapping the process as it should be.&lt;/strong&gt; Map what happens now first. The improved version is a second map.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Too much on one page.&lt;/strong&gt; If a step needs its own flowchart, give it one and keep a single box for it here.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Only the path that goes well.&lt;/strong&gt; A map with no refusals, errors or delays describes a process that exists only on good days.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Vague step names.&lt;/strong&gt; &amp;quot;Processing&amp;quot; tells the reader nothing. Use a verb and a thing.&lt;/li&gt;&lt;/ul&gt;
&lt;h2 id=&quot;questions&quot;&gt;Questions&lt;/h2&gt;
&lt;h3 id=&quot;what-is-process-mapping&quot;&gt;What is process mapping?&lt;/h3&gt;
&lt;p&gt;Process mapping is drawing a process from its start to its end, showing the steps in order, the decisions and who does each part, so that everyone involved sees the same picture of the work.&lt;/p&gt;
&lt;h3 id=&quot;what-is-the-difference-between-a-process-map-and-a-flowchart&quot;&gt;What is the difference between a process map and a flowchart?&lt;/h3&gt;
&lt;p&gt;A process map is the general term for a picture of how work flows. A flowchart is the most common form of process map. A swimlane diagram is a flowchart divided into lanes, one for each person or team.&lt;/p&gt;
&lt;h3 id=&quot;how-detailed-should-a-process-map-be&quot;&gt;How detailed should a process map be?&lt;/h3&gt;
&lt;p&gt;Detailed enough that a new person could follow it, and no more. If the map does not fit on one page, split it: keep one box for the sub-process and draw that sub-process on its own page.&lt;/p&gt;
&lt;h3 id=&quot;who-should-be-involved-in-mapping-a-process&quot;&gt;Who should be involved in mapping a process?&lt;/h3&gt;
&lt;p&gt;The people who do the work every day, plus whoever receives the result. A map drawn by a manager alone usually shows the designed process, not the real one.&lt;/p&gt;
&lt;h3 id=&quot;how-often-should-a-process-map-be-updated&quot;&gt;How often should a process map be updated?&lt;/h3&gt;
&lt;p&gt;Whenever the process changes. In practice, give the map an owner and review it when a step, a tool or a role changes, and at least once a year.&lt;/p&gt;</content></entry><entry><title>Employee Onboarding Process: Steps, Checklist and Flowchart</title><link href="https://www.vamimi.com/blog/employee-onboarding-process"/><id>https://www.vamimi.com/blog/employee-onboarding-process</id><published>2026-11-16T00:00:00Z</published><updated>2026-11-16T00:00:00Z</updated><author><name>VamimiAI team</name></author><summary>The employee onboarding process in four phases, a checklist for each, who owns every step, and how to draw the onboarding flowchart for your team.</summary><content type="html">&lt;p&gt;An employee onboarding process has four phases: before the first day (paperwork, accounts, equipment), the first day (welcome and setup), the first week (training and a first task) and the first month or more (check-ins and a review). It works when every step has one owner, and when HR, IT and the manager can each see what the others are waiting for.&lt;/p&gt;
&lt;p&gt;Below is a checklist for each phase, then how to turn it into a flowchart that shows the hand-offs.&lt;/p&gt;
&lt;h2 id=&quot;who-owns-what&quot;&gt;Who owns what&lt;/h2&gt;
&lt;p&gt;Onboarding fails at the joins between people, so name the owners before the steps.&lt;/p&gt;
&lt;div class=&quot;table&quot;&gt;&lt;table&gt;&lt;thead&gt;&lt;tr&gt;&lt;th&gt;Owner&lt;/th&gt;&lt;th&gt;Typically responsible for&lt;/th&gt;&lt;/tr&gt;&lt;/thead&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td&gt;HR or people team&lt;/td&gt;&lt;td&gt;Contract, forms, payroll and benefits, policies, the overall schedule&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;IT&lt;/td&gt;&lt;td&gt;Accounts, access, laptop and other equipment&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Hiring manager&lt;/td&gt;&lt;td&gt;First-week plan, introductions, goals, check-ins&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;New hire&lt;/td&gt;&lt;td&gt;Returning forms, completing training, asking questions&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;A buddy or mentor&lt;/td&gt;&lt;td&gt;Day-to-day questions in the first weeks&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt;&lt;/div&gt;
&lt;p&gt;In a small company one person may hold several of these roles. Write the roles down anyway; the list of steps stays the same.&lt;/p&gt;
&lt;h2 id=&quot;phase-1-before-the-first-day&quot;&gt;Phase 1: Before the first day&lt;/h2&gt;
&lt;p&gt;Most first-day problems are decided here.&lt;/p&gt;
&lt;ul&gt;&lt;li&gt;Send the contract and the forms, and confirm they come back complete.&lt;/li&gt;&lt;li&gt;Set up payroll and benefits.&lt;/li&gt;&lt;li&gt;Create the email account and the accounts for the tools the role needs.&lt;/li&gt;&lt;li&gt;Order the laptop and equipment, with enough time for delivery.&lt;/li&gt;&lt;li&gt;Tell the team who is joining and when.&lt;/li&gt;&lt;li&gt;Write the plan for the first week and choose a buddy.&lt;/li&gt;&lt;li&gt;Send the new hire the practical details: start time, place or link, who will meet them.&lt;/li&gt;&lt;/ul&gt;
&lt;h2 id=&quot;phase-2-the-first-day&quot;&gt;Phase 2: The first day&lt;/h2&gt;
&lt;ul&gt;&lt;li&gt;Welcome them in person or on a call. Somebody must be expecting them.&lt;/li&gt;&lt;li&gt;Hand over the equipment and check that every account works.&lt;/li&gt;&lt;li&gt;Introduce the team and the buddy.&lt;/li&gt;&lt;li&gt;Walk through the first-week plan.&lt;/li&gt;&lt;li&gt;Cover what cannot wait: working hours, how to ask for help, anything required for safety or security.&lt;/li&gt;&lt;/ul&gt;
&lt;p&gt;Keep the day light. A person who has logged in to everything and knows what tomorrow holds has had a good first day.&lt;/p&gt;
&lt;h2 id=&quot;phase-3-the-first-week&quot;&gt;Phase 3: The first week&lt;/h2&gt;
&lt;ul&gt;&lt;li&gt;Role-specific training, spread across the week.&lt;/li&gt;&lt;li&gt;A first small task with a real result.&lt;/li&gt;&lt;li&gt;Introductions to the people in other teams they will work with.&lt;/li&gt;&lt;li&gt;A short conversation with the manager at the end of the week: what is clear, what is not.&lt;/li&gt;&lt;/ul&gt;
&lt;h2 id=&quot;phase-4-the-first-month-and-after&quot;&gt;Phase 4: The first month and after&lt;/h2&gt;
&lt;ul&gt;&lt;li&gt;A regular check-in with the manager.&lt;/li&gt;&lt;li&gt;Goals for the first 30, 60 and 90 days, written down.&lt;/li&gt;&lt;li&gt;The rest of the training.&lt;/li&gt;&lt;li&gt;A review at 30 days, and again at the end of any probation period.&lt;/li&gt;&lt;li&gt;Ask the new hire what was missing from their onboarding, and fix the process for the next person.&lt;/li&gt;&lt;/ul&gt;
&lt;h2 id=&quot;from-the-checklist-to-a-flowchart&quot;&gt;From the checklist to a flowchart&lt;/h2&gt;
&lt;p&gt;A checklist tells each person what to do. It does not show order or dependency: that IT cannot create accounts until HR has the signed forms, or that the manager&amp;#x27;s plan depends on the equipment arriving. A flowchart shows those links.&lt;/p&gt;
&lt;p&gt;To draw it:&lt;/p&gt;
&lt;ol&gt;&lt;li&gt;&lt;strong&gt;Draw a lane for each owner&lt;/strong&gt; from the table above.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Place each step in its owner&amp;#x27;s lane&lt;/strong&gt;, in the order it happens, starting from the signed offer.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Add a decision wherever something can come back incomplete&lt;/strong&gt;: &amp;quot;Forms returned?&amp;quot;, &amp;quot;Laptop delivered?&amp;quot;. Draw a line for each answer, and say what happens on &amp;quot;No&amp;quot;.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Look at every line that crosses lanes.&lt;/strong&gt; Each one is a hand-off. Ask how the next person finds out that it is their turn.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;End with the review&lt;/strong&gt;, not with day one.&lt;/li&gt;&lt;/ol&gt;
&lt;p&gt;VamimiAI&amp;#x27;s &lt;a href=&quot;/onboarding-process-flowchart&quot;&gt;onboarding process flowchart&lt;/a&gt; page shows an example with lanes for HR, IT and the manager, and you can start from the prompt below. If your checklist is already in a document or a spreadsheet, attach it and ask for the flowchart.&lt;/p&gt;
&lt;aside class=&quot;try-card&quot;&gt;&lt;p&gt;Try it in VamimiAI&lt;/p&gt;&lt;a href=&quot;/app?q=Onboarding%20process%20flowchart%20for%20a%20new%20employee%2C%20with%20lanes%20for%20HR%2C%20IT%2C%20the%20hiring%20manager%20and%20the%20new%20hire%2C%20from%20signed%20offer%20to%20the%2030-day%20review&quot;&gt;&lt;q&gt;Onboarding process flowchart for a new employee, with lanes for HR, IT, the hiring manager and the new hire, from signed offer to the 30-day review&lt;/q&gt;&lt;span&gt;Try this prompt →&lt;/span&gt;&lt;/a&gt;&lt;small&gt;The prompt goes into the message box for you to edit; nothing is sent until you press Send.&lt;/small&gt;&lt;/aside&gt;
&lt;p&gt;An &lt;a href=&quot;/org-chart-maker&quot;&gt;org chart&lt;/a&gt; is a useful companion: give it to the new hire on day one so the names they hear have a place.&lt;/p&gt;
&lt;h2 id=&quot;common-gaps-the-flowchart-exposes&quot;&gt;Common gaps the flowchart exposes&lt;/h2&gt;
&lt;ul&gt;&lt;li&gt;&lt;strong&gt;No step for a late laptop.&lt;/strong&gt; Draw what happens if the equipment does not arrive: a loan device, or a later start.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Access requested after day one.&lt;/strong&gt; Account creation should start from the returned forms, not from the first morning.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Nobody owns the first-week plan.&lt;/strong&gt; If it sits in no lane, it does not exist.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Onboarding ends on day one.&lt;/strong&gt; With no check-in drawn, there is none.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Remote hires follow the office process.&lt;/strong&gt; Shipping, time zones and a first video call need their own steps.&lt;/li&gt;&lt;/ul&gt;
&lt;p&gt;For the general method of drawing any process, see the &lt;a href=&quot;/blog/process-mapping-guide&quot;&gt;process mapping guide&lt;/a&gt;.&lt;/p&gt;
&lt;h2 id=&quot;questions&quot;&gt;Questions&lt;/h2&gt;
&lt;h3 id=&quot;what-are-the-steps-of-the-employee-onboarding-process&quot;&gt;What are the steps of the employee onboarding process?&lt;/h3&gt;
&lt;p&gt;Before the first day: paperwork, payroll, accounts and equipment. On the first day: a welcome, setup and introductions. In the first week: training and a first task. In the first month and after: check-ins, goals and a review.&lt;/p&gt;
&lt;h3 id=&quot;how-long-should-onboarding-last&quot;&gt;How long should onboarding last?&lt;/h3&gt;
&lt;p&gt;Longer than the first day or week. Many teams plan for the first 90 days, with reviews at 30, 60 and 90 days. The right length is the time it takes for the person to do the job without help.&lt;/p&gt;
&lt;h3 id=&quot;what-is-the-difference-between-an-onboarding-checklist-and-an-onboarding-flowchart&quot;&gt;What is the difference between an onboarding checklist and an onboarding flowchart?&lt;/h3&gt;
&lt;p&gt;A checklist lists the tasks, usually for one owner. A flowchart shows the order of the tasks, who does each one and where work passes between people, including what happens when a step is late or incomplete.&lt;/p&gt;
&lt;h3 id=&quot;who-is-responsible-for-onboarding&quot;&gt;Who is responsible for onboarding?&lt;/h3&gt;
&lt;p&gt;Several people: HR for paperwork and the schedule, IT for accounts and equipment, and the manager for the plan, the introductions and the check-ins. The process works when each step has exactly one owner.&lt;/p&gt;
&lt;h3 id=&quot;how-do-i-onboard-a-remote-employee&quot;&gt;How do I onboard a remote employee?&lt;/h3&gt;
&lt;p&gt;Use the same phases, and add the steps an office hides: ship the equipment early, schedule the first-day call, introduce people by video, and agree how and when to ask questions.&lt;/p&gt;</content></entry><entry><title>n8n vs Make vs Zapier for Client Work: Failures and Handover</title><link href="https://www.vamimi.com/blog/n8n-vs-make-vs-zapier-for-client-work"/><id>https://www.vamimi.com/blog/n8n-vs-make-vs-zapier-for-client-work</id><published>2026-11-01T00:00:00Z</published><updated>2026-11-01T00:00:00Z</updated><author><name>VamimiAI team</name></author><summary>n8n vs Make vs Zapier compared on what matters after you hand an automation to a client - what happens when a step fails, who is told, and how you document it.</summary><content type="html">&lt;p&gt;Most n8n vs Make vs Zapier comparisons are about plans and the number of integrations. For client work, the differences that cost you later are elsewhere: what happens when a step fails, whether anyone is told, whether a failed run can be retried, and how easily you can show the client what you built. In short, n8n gives you the most control over failures but expects you to set it up; Make has the richest built-in error handling and keeps failed runs for a retry if you turn that on; Zapier is the simplest to hand over and the least explicit about failure paths.&lt;/p&gt;
&lt;p&gt;Plans, limits and features change often, so check each platform&amp;#x27;s current documentation for the details of your plan.&lt;/p&gt;
&lt;h2 id=&quot;at-a-glance&quot;&gt;At a glance&lt;/h2&gt;
&lt;div class=&quot;table&quot;&gt;&lt;table&gt;&lt;thead&gt;&lt;tr&gt;&lt;th&gt;&lt;/th&gt;&lt;th&gt;n8n&lt;/th&gt;&lt;th&gt;Make&lt;/th&gt;&lt;th&gt;Zapier&lt;/th&gt;&lt;/tr&gt;&lt;/thead&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td&gt;When a step fails, by default&lt;/td&gt;&lt;td&gt;The run stops at that step&lt;/td&gt;&lt;td&gt;The run stops; repeated failures switch the scenario off&lt;/td&gt;&lt;td&gt;The Zap run errors; repeated errors can turn the Zap off&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Retry a failing step&lt;/td&gt;&lt;td&gt;Retry On Fail on the node&lt;/td&gt;&lt;td&gt;Break error handler with automatic retries&lt;/td&gt;&lt;td&gt;Autoreplay of failed runs, depending on the plan&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Handle the failure in the flow&lt;/td&gt;&lt;td&gt;An error output on the node&lt;/td&gt;&lt;td&gt;Error handler routes: Ignore, Resume, Commit, Rollback, Break&lt;/td&gt;&lt;td&gt;Error handling inside the Zap, depending on the plan&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Tell someone&lt;/td&gt;&lt;td&gt;An error workflow (Error Trigger) per workflow&lt;/td&gt;&lt;td&gt;Notifications from Make, or a handler route that alerts&lt;/td&gt;&lt;td&gt;Error emails from Zapier&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Keep failed runs for later&lt;/td&gt;&lt;td&gt;Failed executions in the Executions list&lt;/td&gt;&lt;td&gt;Incomplete executions, if storing them is turned on&lt;/td&gt;&lt;td&gt;Zap history, and replay where available&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Export for documentation&lt;/td&gt;&lt;td&gt;Workflow JSON (Download)&lt;/td&gt;&lt;td&gt;Blueprint JSON (Export Blueprint)&lt;/td&gt;&lt;td&gt;Zap export file, where your account offers it&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt;&lt;/div&gt;
&lt;h2 id=&quot;n8n-the-most-control-if-you-set-it-up&quot;&gt;n8n: the most control, if you set it up&lt;/h2&gt;
&lt;p&gt;Every node can retry (&lt;strong&gt;Retry On Fail&lt;/strong&gt;) and can send its failure down a second, error output instead of stopping the run. A separate &lt;strong&gt;error workflow&lt;/strong&gt; that starts with an &lt;strong&gt;Error Trigger&lt;/strong&gt; catches anything else, but only for the workflows that name it in their settings, and only for runs started by a trigger, not for tests you run by hand. That is the trap for client work: a workflow without an error workflow fails silently. The &lt;a href=&quot;/blog/n8n-error-handling-guide&quot;&gt;n8n error handling guide&lt;/a&gt; goes through each option.&lt;/p&gt;
&lt;h2 id=&quot;make-the-richest-built-in-error-handling&quot;&gt;Make: the richest built-in error handling&lt;/h2&gt;
&lt;p&gt;An error handler route on a module ends in one of five directives that decide what happens to the run and to the bundle that failed. &lt;strong&gt;Break&lt;/strong&gt; with &lt;strong&gt;Allow storing of incomplete executions&lt;/strong&gt; turned on keeps a failed run so it can be retried, which is the safest default for anything that calls an outside service. Without handlers, repeated failures switch the scenario off. &lt;a href=&quot;/blog/make-error-handlers-explained&quot;&gt;Make error handlers explained&lt;/a&gt; covers all five, and &lt;a href=&quot;/blog/make-scenario-deactivated&quot;&gt;Make scenario deactivated&lt;/a&gt; covers what to do when it happens.&lt;/p&gt;
&lt;h2 id=&quot;zapier-the-simplest-to-hand-over&quot;&gt;Zapier: the simplest to hand over&lt;/h2&gt;
&lt;p&gt;A Zap is a straight line of steps with Paths for branches, so a client can follow it on the canvas. Failure handling is less visible: failed runs show up in the Zap history, Zapier emails about errors, and replaying failed runs depends on your plan. Before handing a Zap over, decide who receives those emails and what they should do.&lt;/p&gt;
&lt;h2 id=&quot;what-this-means-for-the-handover&quot;&gt;What this means for the handover&lt;/h2&gt;
&lt;p&gt;Whichever platform you build on, the client needs the same three answers: what starts it, what each step does, and what happens when a step fails. The platform&amp;#x27;s canvas answers the first two for you and rarely the third. Importing the export into VamimiAI draws every path, including the failure paths, for &lt;a href=&quot;/n8n-workflow-diagram&quot;&gt;n8n workflows&lt;/a&gt; and &lt;a href=&quot;/make-scenario-diagram&quot;&gt;Make blueprints&lt;/a&gt;, and reads Zapier exports on a best-effort basis because Zapier does not publish its export format. The &lt;a href=&quot;/automation-handover-document&quot;&gt;handover document&lt;/a&gt; then puts the diagram, a step table and care notes on one page.&lt;/p&gt;
&lt;aside class=&quot;try-card&quot;&gt;&lt;p&gt;Try it in VamimiAI&lt;/p&gt;&lt;a href=&quot;/app?q=Compare%20n8n%2C%20Make%20and%20Zapier%20for%20a%20client%20with%202%2C000%20orders%20a%20month%20in%20a%20table&quot;&gt;&lt;q&gt;Compare n8n, Make and Zapier for a client with 2,000 orders a month in a table&lt;/q&gt;&lt;span&gt;Try this prompt →&lt;/span&gt;&lt;/a&gt;&lt;small&gt;The prompt goes into the message box for you to edit; nothing is sent until you press Send.&lt;/small&gt;&lt;/aside&gt;
&lt;h2 id=&quot;questions&quot;&gt;Questions&lt;/h2&gt;
&lt;h3 id=&quot;is-n8n-make-or-zapier-best-for-freelancers-building-for-clients&quot;&gt;Is n8n, Make or Zapier best for freelancers building for clients?&lt;/h3&gt;
&lt;p&gt;It depends on the client more than the platform. n8n suits clients who want control or self-hosting and a builder who sets up error handling carefully; Make suits complex scenarios that need retries and partial failures handled; Zapier suits simple automations the client may look after themselves.&lt;/p&gt;
&lt;h3 id=&quot;which-platform-has-the-best-error-handling&quot;&gt;Which platform has the best error handling?&lt;/h3&gt;
&lt;p&gt;Make has the most built-in options (five error handler directives and stored incomplete executions). n8n can do as much with retries, error outputs and error workflows, but you have to set each one up. Zapier&amp;#x27;s options depend more on the plan.&lt;/p&gt;
&lt;h3 id=&quot;can-i-move-an-automation-between-n8n-make-and-zapier&quot;&gt;Can I move an automation between n8n, Make and Zapier?&lt;/h3&gt;
&lt;p&gt;Not directly: each platform exports its own format. A diagram of the automation, with every step and failure path, is the most useful starting point for rebuilding it on another platform.&lt;/p&gt;</content></entry><entry><title>n8n Error Workflow Not Triggering? 6 Causes and Fixes</title><link href="https://www.vamimi.com/blog/n8n-error-workflow-not-triggering"/><id>https://www.vamimi.com/blog/n8n-error-workflow-not-triggering</id><published>2026-11-01T00:00:00Z</published><updated>2026-11-01T00:00:00Z</updated><author><name>VamimiAI team</name></author><summary>Why an n8n error workflow does not run when a workflow fails, how to check each cause in a few minutes, and how to test it safely before a client relies on it.</summary><content type="html">&lt;p&gt;When an n8n error workflow does not trigger, the cause is almost always one of six things: the failing workflow does not point to it, it does not start with an Error Trigger, the failure happened in a run you started by hand, the node was set to continue instead of fail, the failure was inside a sub-workflow, or the error workflow ran and failed itself. Check them in this order; the first three cover most cases.&lt;/p&gt;
&lt;h2 id=&quot;1-the-failing-workflow-does-not-point-to-it&quot;&gt;1. The failing workflow does not point to it&lt;/h2&gt;
&lt;p&gt;The link is set on the workflow that fails, not on the error workflow. Open the failing workflow, open &lt;strong&gt;Workflow settings&lt;/strong&gt;, and check that &lt;strong&gt;Error Workflow&lt;/strong&gt; names your error workflow. A new workflow, a duplicated one and one imported from a file start with no error workflow set, so a workflow you copied from a working one may not have it.&lt;/p&gt;
&lt;h2 id=&quot;2-the-error-workflow-does-not-start-with-an-error-trigger&quot;&gt;2. The error workflow does not start with an Error Trigger&lt;/h2&gt;
&lt;p&gt;An error workflow is an ordinary workflow whose first node is the &lt;strong&gt;Error Trigger&lt;/strong&gt;. If it starts with a webhook or a schedule instead, n8n has nothing to call when a run fails.&lt;/p&gt;
&lt;h2 id=&quot;3-you-tested-by-running-the-workflow-by-hand&quot;&gt;3. You tested by running the workflow by hand&lt;/h2&gt;
&lt;p&gt;n8n starts the error workflow for runs started by a trigger: a schedule, a webhook, an app event. A failure in a run you start yourself from the editor shows the error on the canvas instead. This is the most common reason people think it is broken.&lt;/p&gt;
&lt;p&gt;To test it properly, put a &lt;strong&gt;Stop and Error&lt;/strong&gt; node where you want the failure, make sure the workflow runs from its trigger (call the production webhook URL, or let the schedule fire), and watch the error workflow&amp;#x27;s executions.&lt;/p&gt;
&lt;h2 id=&quot;4-the-node-was-set-to-continue-not-to-fail&quot;&gt;4. The node was set to continue, not to fail&lt;/h2&gt;
&lt;p&gt;A node whose &lt;strong&gt;On Error&lt;/strong&gt; setting is &lt;strong&gt;Continue&lt;/strong&gt; or &lt;strong&gt;Continue (using error output)&lt;/strong&gt; does not fail the run. The run carries on, finishes, and the error workflow never hears about it. That is intended: the failure is handled by the path you built. If you also want an alert, send one from the error output, or end that path with a &lt;strong&gt;Stop and Error&lt;/strong&gt; node so the run fails on purpose.&lt;/p&gt;
&lt;h2 id=&quot;5-the-failure-happened-inside-a-sub-workflow&quot;&gt;5. The failure happened inside a sub-workflow&lt;/h2&gt;
&lt;p&gt;When a workflow calls another one with &lt;strong&gt;Execute Workflow&lt;/strong&gt;, the failure happens in the sub-workflow. Set the error workflow on the sub-workflow as well, and check the executions of both, so a failure is caught wherever it happens.&lt;/p&gt;
&lt;h2 id=&quot;6-the-error-workflow-ran-and-failed-too&quot;&gt;6. The error workflow ran, and failed too&lt;/h2&gt;
&lt;p&gt;Open the error workflow&amp;#x27;s own executions. If its Slack, email or ticket step uses a login that has expired, the alert fails silently. Keep the error workflow small: one alert step with a retry, using a login that is not shared with the workflows it watches.&lt;/p&gt;
&lt;h2 id=&quot;make-it-part-of-the-handover&quot;&gt;Make it part of the handover&lt;/h2&gt;
&lt;p&gt;Before you hand a workflow to a client, every production workflow should name the error workflow, and the error workflow should have been tested from a real trigger. Importing the workflow into &lt;a href=&quot;/n8n-workflow-diagram&quot;&gt;VamimiAI&amp;#x27;s n8n workflow diagram&lt;/a&gt; flags a workflow with no error workflow set and draws each call to an outside service that nothing protects as a red end, so you can see what is still unprotected. The &lt;a href=&quot;/blog/n8n-error-handling-guide&quot;&gt;n8n error handling guide&lt;/a&gt; covers retries and error outputs, and the &lt;a href=&quot;/automation-handover-document&quot;&gt;handover document&lt;/a&gt; tells the client where failed runs show up.&lt;/p&gt;
&lt;aside class=&quot;try-card&quot;&gt;&lt;p&gt;Try it in VamimiAI&lt;/p&gt;&lt;a href=&quot;/app?q=Draw%20an%20n8n%20error%20workflow%20that%20alerts%20ops%20in%20Slack%20when%20any%20workflow%20fails&quot;&gt;&lt;q&gt;Draw an n8n error workflow that alerts ops in Slack when any workflow fails&lt;/q&gt;&lt;span&gt;Try this prompt →&lt;/span&gt;&lt;/a&gt;&lt;small&gt;The prompt goes into the message box for you to edit; nothing is sent until you press Send.&lt;/small&gt;&lt;/aside&gt;
&lt;h2 id=&quot;questions&quot;&gt;Questions&lt;/h2&gt;
&lt;h3 id=&quot;does-the-n8n-error-workflow-run-on-manual-executions&quot;&gt;Does the n8n error workflow run on manual executions?&lt;/h3&gt;
&lt;p&gt;No. It runs when a run started by a trigger fails. To test it, run the workflow from its trigger with a Stop and Error node where you want the failure.&lt;/p&gt;
&lt;h3 id=&quot;do-i-need-one-error-workflow-per-workflow&quot;&gt;Do I need one error workflow per workflow?&lt;/h3&gt;
&lt;p&gt;No. One error workflow can serve every workflow of a client; each workflow names it in its own settings.&lt;/p&gt;
&lt;h3 id=&quot;why-does-my-workflow-finish-successfully-even-though-a-node-failed&quot;&gt;Why does my workflow finish successfully even though a node failed?&lt;/h3&gt;
&lt;p&gt;The node&amp;#x27;s On Error setting is Continue or Continue (using error output), so the failure is handled by the path you built and the run does not fail. Send an alert from that path, or end it with a Stop and Error node.&lt;/p&gt;</content></entry><entry><title>n8n Error Handling: Catch Every Failure Before a Client Does</title><link href="https://www.vamimi.com/blog/n8n-error-handling-guide"/><id>https://www.vamimi.com/blog/n8n-error-handling-guide</id><published>2026-11-01T00:00:00Z</published><updated>2026-11-01T00:00:00Z</updated><author><name>VamimiAI team</name></author><summary>Retry On Fail, error outputs, Continue, error workflows and the Error Trigger in n8n, what each one does and which to use for steps that call outside services.</summary><content type="html">&lt;p&gt;n8n gives you three places to handle a failure: on the node (Retry On Fail and the On Error setting), in the workflow (an error output route), and for the whole workflow (an error workflow that starts with an Error Trigger). A production workflow you hand to a client should use all three: retries for passing problems, an error path for failures you can handle, and an error workflow so that anything else at least tells someone.&lt;/p&gt;
&lt;h2 id=&quot;what-happens-by-default&quot;&gt;What happens by default&lt;/h2&gt;
&lt;p&gt;When a node fails and you changed nothing, the run stops at that node. The execution is marked as failed in the Executions list, the remaining nodes never run, and unless the workflow has an error workflow set, nobody is notified. For a workflow that runs on a schedule or a webhook, that means a failure is silent until someone goes looking.&lt;/p&gt;
&lt;p&gt;The nodes most likely to fail are the ones that call another service: HTTP Request, any app node (HubSpot, Slack, Google Sheets, Airtable...), and AI nodes. They fail when the service is down, when you hit a rate limit, or when a login expires.&lt;/p&gt;
&lt;h2 id=&quot;on-the-node-retry-on-fail&quot;&gt;On the node: Retry On Fail&lt;/h2&gt;
&lt;p&gt;In a node&amp;#x27;s &lt;strong&gt;Settings&lt;/strong&gt;, turn on &lt;strong&gt;Retry On Fail&lt;/strong&gt; and set &lt;strong&gt;Max Tries&lt;/strong&gt; and &lt;strong&gt;Wait Between Tries&lt;/strong&gt;. This handles passing problems such as a timeout or a rate limit without any extra nodes.&lt;/p&gt;
&lt;p&gt;Use it on every node that calls an outside service, with a wait long enough for a rate limit to reset. It does not help with a failure that will happen every time, such as a deleted record or a revoked login: those need a path or an alert.&lt;/p&gt;
&lt;h2 id=&quot;on-the-node-the-on-error-setting&quot;&gt;On the node: the On Error setting&lt;/h2&gt;
&lt;p&gt;The same Settings tab has &lt;strong&gt;On Error&lt;/strong&gt;, with three choices:&lt;/p&gt;
&lt;div class=&quot;table&quot;&gt;&lt;table&gt;&lt;thead&gt;&lt;tr&gt;&lt;th&gt;Setting&lt;/th&gt;&lt;th&gt;What happens when the node fails&lt;/th&gt;&lt;th&gt;Use it when&lt;/th&gt;&lt;/tr&gt;&lt;/thead&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td&gt;Stop Workflow&lt;/td&gt;&lt;td&gt;The run stops here (the default)&lt;/td&gt;&lt;td&gt;A failure must not be hidden&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Continue&lt;/td&gt;&lt;td&gt;The run goes on, with the error passed along as data&lt;/td&gt;&lt;td&gt;The next steps can cope with missing data&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Continue (using error output)&lt;/td&gt;&lt;td&gt;The node gets a second, error output you connect like any other path&lt;/td&gt;&lt;td&gt;You want to handle the failure: alert, log, retry later&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt;&lt;/div&gt;
&lt;p&gt;&lt;strong&gt;Continue&lt;/strong&gt; is the risky one: later nodes receive an error item instead of the data they expect, and the run can finish &amp;quot;successfully&amp;quot; having done the wrong thing. If you use it, check for the error in the next node.&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Continue (using error output)&lt;/strong&gt; is the clearest. The failure becomes a visible path on the canvas, and on a diagram, so you can send it somewhere deliberate: a Slack message to ops, a row in a &amp;quot;failed leads&amp;quot; sheet, or a queue to try again later.&lt;/p&gt;
&lt;h2 id=&quot;for-the-whole-workflow-an-error-workflow&quot;&gt;For the whole workflow: an error workflow&lt;/h2&gt;
&lt;p&gt;Create a separate workflow that starts with an &lt;strong&gt;Error Trigger&lt;/strong&gt; node, and send yourself or the client an alert from it (Slack, email, a ticket). Then, in each production workflow, open &lt;strong&gt;Workflow settings&lt;/strong&gt; and choose it under &lt;strong&gt;Error Workflow&lt;/strong&gt;.&lt;/p&gt;
&lt;p&gt;When a run fails, the error workflow receives the workflow name, the failed execution and the error message, so the alert can link straight to the failed run. One shared error workflow for all of a client&amp;#x27;s workflows is usually enough.&lt;/p&gt;
&lt;p&gt;Two things catch people out:&lt;/p&gt;
&lt;ul&gt;&lt;li&gt;The error workflow runs for failed automatic runs (triggered by the schedule, a webhook or an app event), not when you click to test the workflow in the editor. Test it with a real trigger, or force a failure with a &lt;strong&gt;Stop and Error&lt;/strong&gt; node.&lt;/li&gt;&lt;li&gt;The setting is per workflow. A new workflow, or a copy, has no error workflow until you set one.&lt;/li&gt;&lt;/ul&gt;
&lt;p&gt;Still not firing? &lt;a href=&quot;/blog/n8n-error-workflow-not-triggering&quot;&gt;Six causes of an n8n error workflow not triggering&lt;/a&gt;, and how to test it.&lt;/p&gt;
&lt;h2 id=&quot;which-to-use-where&quot;&gt;Which to use where&lt;/h2&gt;
&lt;p&gt;A simple rule for client work:&lt;/p&gt;
&lt;ol&gt;&lt;li&gt;Every node that calls an outside service gets &lt;strong&gt;Retry On Fail&lt;/strong&gt;.&lt;/li&gt;&lt;li&gt;Every failure the business cares about (a lost lead, an unpaid invoice) gets an &lt;strong&gt;error output&lt;/strong&gt; to a deliberate place.&lt;/li&gt;&lt;li&gt;Every production workflow gets the client&amp;#x27;s &lt;strong&gt;error workflow&lt;/strong&gt;.&lt;/li&gt;&lt;li&gt;Use &lt;strong&gt;Continue&lt;/strong&gt; only when the next step checks for the error.&lt;/li&gt;&lt;/ol&gt;
&lt;h2 id=&quot;see-every-failure-path-at-once&quot;&gt;See every failure path at once&lt;/h2&gt;
&lt;p&gt;The hard part is not the settings, it is knowing which nodes still have no protection. Importing the workflow into &lt;a href=&quot;/n8n-workflow-diagram&quot;&gt;VamimiAI&amp;#x27;s n8n workflow diagram&lt;/a&gt; draws the error outputs dashed, and gives every outside-service node with no retry and no error path a red end that says what really happens: the run stops, and whether anyone is told. The checks on import also flag a missing error workflow and errors passed on as data with Continue.&lt;/p&gt;
&lt;aside class=&quot;try-card&quot;&gt;&lt;p&gt;Try it in VamimiAI&lt;/p&gt;&lt;a href=&quot;/app?q=Draw%20an%20n8n%20error%20workflow%20that%20alerts%20ops%20in%20Slack%20when%20any%20workflow%20fails&quot;&gt;&lt;q&gt;Draw an n8n error workflow that alerts ops in Slack when any workflow fails&lt;/q&gt;&lt;span&gt;Try this prompt →&lt;/span&gt;&lt;/a&gt;&lt;small&gt;The prompt goes into the message box for you to edit; nothing is sent until you press Send.&lt;/small&gt;&lt;/aside&gt;
&lt;p&gt;When everything is covered, the &lt;a href=&quot;/automation-handover-document&quot;&gt;handover document&lt;/a&gt; puts the diagram and a table of what happens when each step fails in front of your client. The &lt;a href=&quot;/blog/how-to-document-n8n-workflows-for-clients&quot;&gt;n8n documentation checklist&lt;/a&gt; covers the rest of the handover.&lt;/p&gt;
&lt;h2 id=&quot;questions&quot;&gt;Questions&lt;/h2&gt;
&lt;h3 id=&quot;why-is-my-n8n-error-workflow-not-running&quot;&gt;Why is my n8n error workflow not running?&lt;/h3&gt;
&lt;p&gt;Check that the failing workflow has it selected under Error Workflow in its own settings, that the error workflow starts with an Error Trigger node, and that the failure happened in an automatic run: testing a workflow by hand in the editor does not start the error workflow.&lt;/p&gt;
&lt;h3 id=&quot;what-is-the-difference-between-continue-and-continue-using-error-output&quot;&gt;What is the difference between Continue and Continue (using error output)?&lt;/h3&gt;
&lt;p&gt;Continue passes the error along as data on the normal path, so later nodes must check for it. Continue (using error output) gives the node a separate error output, so the failure gets its own visible path.&lt;/p&gt;
&lt;h3 id=&quot;should-every-node-retry-on-fail&quot;&gt;Should every node retry on fail?&lt;/h3&gt;
&lt;p&gt;Every node that calls an outside service should, with a wait long enough for rate limits. Retrying does not fix failures that will happen every time, such as a revoked login, so those still need an error path or an alert.&lt;/p&gt;</content></entry><entry><title>Make Scenario Deactivated? Errors and Incomplete Executions</title><link href="https://www.vamimi.com/blog/make-scenario-deactivated"/><id>https://www.vamimi.com/blog/make-scenario-deactivated</id><published>2026-11-01T00:00:00Z</published><updated>2026-11-01T00:00:00Z</updated><author><name>VamimiAI team</name></author><summary>Why Make switches a scenario off, what &quot;not enough space to add a new incomplete execution&quot; means, and the settings that keep a scenario running.</summary><content type="html">&lt;p&gt;Make switches a scenario off when its runs keep failing, and it can no longer set failed runs aside when the store of incomplete executions is full. Both come from the same place: modules that fail with nothing to handle the failure, or a failure that keeps coming back. Fix the module that fails, clear what has piled up, and the scenario can run again.&lt;/p&gt;
&lt;h2 id=&quot;why-make-deactivated-the-scenario&quot;&gt;Why Make deactivated the scenario&lt;/h2&gt;
&lt;p&gt;Every scenario has a setting for how many failed runs in a row Make accepts before it stops scheduling the scenario (&lt;strong&gt;Number of consecutive errors&lt;/strong&gt; in the scenario settings). When a module without an error handler fails again and again (a login that expired, an app that changed a field, a record that no longer exists), the count is reached and the scenario is switched off. Depending on the error and on your settings this can happen sooner; Make&amp;#x27;s help pages on scenario settings and on incomplete executions describe the exact rules for your account.&lt;/p&gt;
&lt;p&gt;What to do:&lt;/p&gt;
&lt;ol&gt;&lt;li&gt;Open the scenario&amp;#x27;s run history and find the module that failed, and why.&lt;/li&gt;&lt;li&gt;Fix the cause: reconnect the login, update the mapping, or handle the missing record.&lt;/li&gt;&lt;li&gt;Turn the scenario back on, and run it once while you watch.&lt;/li&gt;&lt;/ol&gt;
&lt;h2 id=&quot;there-is-not-enough-space-to-add-a-new-incomplete-execution&quot;&gt;&amp;quot;There is NOT ENOUGH SPACE to add a new incomplete execution&amp;quot;&lt;/h2&gt;
&lt;p&gt;With &lt;strong&gt;Allow storing of incomplete executions&lt;/strong&gt; on, a run that fails can be kept so it can be resolved or retried later. Those stored runs take space, and when the space is used up Make cannot store another one, so the next failure is reported with this message instead of being kept.&lt;/p&gt;
&lt;p&gt;It usually means the same failure has been happening for a while, often behind a &lt;strong&gt;Break&lt;/strong&gt; error handler that keeps retrying a problem that will not pass on its own.&lt;/p&gt;
&lt;p&gt;What to do:&lt;/p&gt;
&lt;ol&gt;&lt;li&gt;Open the scenario&amp;#x27;s &lt;strong&gt;Incomplete executions&lt;/strong&gt; list and look at what is stored: most often it is one cause, many times.&lt;/li&gt;&lt;li&gt;Fix that cause first, otherwise the list fills up again.&lt;/li&gt;&lt;li&gt;Retry the stored runs that should still go through, and delete the ones that should not.&lt;/li&gt;&lt;li&gt;Check how much storage your plan includes in Make&amp;#x27;s help pages if the list is legitimately large.&lt;/li&gt;&lt;/ol&gt;
&lt;h2 id=&quot;settings-to-check-before-handing-a-scenario-over&quot;&gt;Settings to check before handing a scenario over&lt;/h2&gt;
&lt;ul&gt;&lt;li&gt;&lt;strong&gt;Allow storing of incomplete executions&lt;/strong&gt; is on, so a failed run is kept instead of lost.&lt;/li&gt;&lt;li&gt;Every module that calls an outside service has an error handler, and &lt;strong&gt;Break&lt;/strong&gt; is used for failures that pass (timeouts, rate limits) with a sensible number of retries.&lt;/li&gt;&lt;li&gt;Someone receives Make&amp;#x27;s notifications when a scenario fails or is switched off, and the client knows who that is.&lt;/li&gt;&lt;/ul&gt;
&lt;p&gt;Importing the blueprint into &lt;a href=&quot;/make-scenario-diagram&quot;&gt;VamimiAI&amp;#x27;s Make blueprint to flowchart&lt;/a&gt; flags modules with no error handler and scenarios that do not keep failed runs, and draws each error handler route ending in what its directive does. &lt;a href=&quot;/blog/make-error-handlers-explained&quot;&gt;Make error handlers explained&lt;/a&gt; covers each directive, and the &lt;a href=&quot;/automation-handover-document&quot;&gt;handover document&lt;/a&gt; tells the client where failed runs show up.&lt;/p&gt;
&lt;aside class=&quot;try-card&quot;&gt;&lt;p&gt;Try it in VamimiAI&lt;/p&gt;&lt;a href=&quot;/app?q=Draw%20the%20error%20handling%20for%20a%20Make%20scenario%20that%20syncs%20Shopify%20orders%20to%20Airtable&quot;&gt;&lt;q&gt;Draw the error handling for a Make scenario that syncs Shopify orders to Airtable&lt;/q&gt;&lt;span&gt;Try this prompt →&lt;/span&gt;&lt;/a&gt;&lt;small&gt;The prompt goes into the message box for you to edit; nothing is sent until you press Send.&lt;/small&gt;&lt;/aside&gt;
&lt;h2 id=&quot;questions&quot;&gt;Questions&lt;/h2&gt;
&lt;h3 id=&quot;how-do-i-turn-a-deactivated-make-scenario-back-on&quot;&gt;How do I turn a deactivated Make scenario back on?&lt;/h3&gt;
&lt;p&gt;Fix the module that kept failing first, then switch the scenario on again from the scenario list or the editor. Turning it on without a fix only lets the same failures switch it off again.&lt;/p&gt;
&lt;h3 id=&quot;what-fills-up-incomplete-executions-in-make&quot;&gt;What fills up incomplete executions in Make?&lt;/h3&gt;
&lt;p&gt;Runs that failed while storing incomplete executions is on, often the same failure many times behind a Break handler. Fix the cause, then retry or delete the stored runs.&lt;/p&gt;
&lt;h3 id=&quot;should-i-turn-off-storing-incomplete-executions-to-stop-the-error&quot;&gt;Should I turn off storing incomplete executions to stop the error?&lt;/h3&gt;
&lt;p&gt;No. Then failed runs are simply lost. Keep it on, fix the failing module, and clear the stored runs.&lt;/p&gt;</content></entry><entry><title>Make Error Handlers: Break, Rollback, Commit, Ignore, Resume</title><link href="https://www.vamimi.com/blog/make-error-handlers-explained"/><id>https://www.vamimi.com/blog/make-error-handlers-explained</id><published>2026-11-01T00:00:00Z</published><updated>2026-11-01T00:00:00Z</updated><author><name>VamimiAI team</name></author><summary>What each Make error handler directive does to the run, the bundle and your data, when to use it, and the scenario settings that decide if a failure is lost.</summary><content type="html">&lt;p&gt;In Make, an error handler is a route you attach to a module; it runs when that module fails and ends in a directive that decides what happens next. &lt;strong&gt;Ignore&lt;/strong&gt; and &lt;strong&gt;Resume&lt;/strong&gt; let the run carry on, &lt;strong&gt;Commit&lt;/strong&gt; and &lt;strong&gt;Rollback&lt;/strong&gt; stop it, and &lt;strong&gt;Break&lt;/strong&gt; sets the failed bundle aside as an incomplete execution so it can be retried. Which one you pick decides whether a client&amp;#x27;s order or lead is handled, skipped or lost.&lt;/p&gt;
&lt;h2 id=&quot;what-happens-with-no-error-handler&quot;&gt;What happens with no error handler&lt;/h2&gt;
&lt;p&gt;When a module without an error handler fails, the run stops with an error. If the scenario keeps failing, Make switches it off after the number of consecutive errors set in the scenario settings, and every run after that simply does not happen until someone turns it back on.&lt;/p&gt;
&lt;p&gt;Two scenario settings matter here, and both are worth checking before a handover:&lt;/p&gt;
&lt;ul&gt;&lt;li&gt;&lt;strong&gt;Allow storing of incomplete executions.&lt;/strong&gt; With it on, a run that fails can be kept and resolved or retried later instead of being lost.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Number of consecutive errors.&lt;/strong&gt; How many failed runs in a row it takes for Make to switch the scenario off.&lt;/li&gt;&lt;/ul&gt;
&lt;h2 id=&quot;the-five-directives&quot;&gt;The five directives&lt;/h2&gt;
&lt;p&gt;Add an error handler by right-clicking a module and choosing &lt;strong&gt;Add error handler&lt;/strong&gt;, then end the route with one of these:&lt;/p&gt;
&lt;div class=&quot;table&quot;&gt;&lt;table&gt;&lt;thead&gt;&lt;tr&gt;&lt;th&gt;Directive&lt;/th&gt;&lt;th&gt;The run&lt;/th&gt;&lt;th&gt;The failed bundle&lt;/th&gt;&lt;th&gt;Use it when&lt;/th&gt;&lt;/tr&gt;&lt;/thead&gt;&lt;tbody&gt;&lt;tr&gt;&lt;td&gt;Ignore&lt;/td&gt;&lt;td&gt;Carries on with the next bundle&lt;/td&gt;&lt;td&gt;Skipped&lt;/td&gt;&lt;td&gt;One bad record must not stop the rest&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Resume&lt;/td&gt;&lt;td&gt;Carries on as if the module had worked&lt;/td&gt;&lt;td&gt;Replaced by output you specify&lt;/td&gt;&lt;td&gt;A sensible default value exists&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Commit&lt;/td&gt;&lt;td&gt;Stops, marked as a success&lt;/td&gt;&lt;td&gt;Not processed further&lt;/td&gt;&lt;td&gt;Partial work is fine to keep&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Rollback&lt;/td&gt;&lt;td&gt;Stops, marked as an error&lt;/td&gt;&lt;td&gt;Not processed further&lt;/td&gt;&lt;td&gt;Partial work must be undone&lt;/td&gt;&lt;/tr&gt;&lt;tr&gt;&lt;td&gt;Break&lt;/td&gt;&lt;td&gt;Stops for that bundle&lt;/td&gt;&lt;td&gt;Saved as an incomplete execution, retried automatically if you set it to&lt;/td&gt;&lt;td&gt;The failure is likely to pass, such as a timeout&lt;/td&gt;&lt;/tr&gt;&lt;/tbody&gt;&lt;/table&gt;&lt;/div&gt;
&lt;h3 id=&quot;ignore&quot;&gt;Ignore&lt;/h3&gt;
&lt;p&gt;The error is swallowed: the bundle that failed is dropped and the scenario goes on. The run looks successful, so nobody will notice the skipped record unless the handler route does something visible first, such as posting to Slack or adding a row to a log sheet before the Ignore.&lt;/p&gt;
&lt;h3 id=&quot;resume&quot;&gt;Resume&lt;/h3&gt;
&lt;p&gt;You give the output the failed module should have produced, and the run continues with it. Good for optional enrichment (a missing company size becomes &amp;quot;unknown&amp;quot;), dangerous for anything that matters, because the substitute value flows into everything after it.&lt;/p&gt;
&lt;h3 id=&quot;rollback-vs-commit&quot;&gt;Rollback vs Commit&lt;/h3&gt;
&lt;p&gt;Both stop the run. Commit keeps what was already done and marks the run as a success; Rollback reverts the changes made by modules that support transactions and marks the run as an error. Most app modules cannot be undone, so Rollback mainly protects data stores and databases: an email that was sent stays sent.&lt;/p&gt;
&lt;h3 id=&quot;break&quot;&gt;Break&lt;/h3&gt;
&lt;p&gt;Break is the safest choice for outside services. The failed bundle is stored as an incomplete execution (this needs &lt;strong&gt;Allow storing of incomplete executions&lt;/strong&gt; on), and with automatic completion turned on, Make retries it after the interval you choose, for the number of attempts you set. If the retries run out, it waits in the list of incomplete executions for someone to resolve.&lt;/p&gt;
&lt;h2 id=&quot;a-sensible-default-for-client-scenarios&quot;&gt;A sensible default for client scenarios&lt;/h2&gt;
&lt;ol&gt;&lt;li&gt;Turn on &lt;strong&gt;Allow storing of incomplete executions&lt;/strong&gt; for every production scenario.&lt;/li&gt;&lt;li&gt;Give every module that calls an outside service an error handler that ends in &lt;strong&gt;Break&lt;/strong&gt; with automatic retries.&lt;/li&gt;&lt;li&gt;Where skipping a record is acceptable, log it first and then &lt;strong&gt;Ignore&lt;/strong&gt;, so the skip is visible.&lt;/li&gt;&lt;li&gt;Keep &lt;strong&gt;Resume&lt;/strong&gt; for values that are truly optional.&lt;/li&gt;&lt;/ol&gt;
&lt;h2 id=&quot;see-where-a-scenario-will-lose-data&quot;&gt;See where a scenario will lose data&lt;/h2&gt;
&lt;p&gt;The scenario editor shows error handler routes, but not what each one means for the client&amp;#x27;s data, and it never shows the modules that have no handler at all. Importing the blueprint into &lt;a href=&quot;/make-scenario-diagram&quot;&gt;VamimiAI&amp;#x27;s Make scenario diagram&lt;/a&gt; draws each error handler route dashed and ends it in what its directive does, in plain words such as &amp;quot;Saved to incomplete runs for a retry&amp;quot;. The checks on import flag modules with no error handler and scenarios that do not keep failed runs.&lt;/p&gt;
&lt;aside class=&quot;try-card&quot;&gt;&lt;p&gt;Try it in VamimiAI&lt;/p&gt;&lt;a href=&quot;/app?q=Draw%20the%20error%20handling%20for%20a%20Make%20scenario%20that%20syncs%20Shopify%20orders%20to%20Airtable&quot;&gt;&lt;q&gt;Draw the error handling for a Make scenario that syncs Shopify orders to Airtable&lt;/q&gt;&lt;span&gt;Try this prompt →&lt;/span&gt;&lt;/a&gt;&lt;small&gt;The prompt goes into the message box for you to edit; nothing is sent until you press Send.&lt;/small&gt;&lt;/aside&gt;
&lt;p&gt;Then create the &lt;a href=&quot;/automation-handover-document&quot;&gt;handover document&lt;/a&gt; so the client sees the same picture, with a table of what happens when each module fails. Scenario already switched off, or seeing &amp;quot;not enough space to add a new incomplete execution&amp;quot;? See &lt;a href=&quot;/blog/make-scenario-deactivated&quot;&gt;Make scenario deactivated&lt;/a&gt;. Building on n8n as well? The &lt;a href=&quot;/blog/n8n-error-handling-guide&quot;&gt;n8n error handling guide&lt;/a&gt; covers the same ground there.&lt;/p&gt;
&lt;h2 id=&quot;questions&quot;&gt;Questions&lt;/h2&gt;
&lt;h3 id=&quot;what-is-the-difference-between-ignore-and-break-in-make&quot;&gt;What is the difference between Ignore and Break in Make?&lt;/h3&gt;
&lt;p&gt;Ignore drops the failed bundle and lets the run carry on, so the record is skipped. Break stops processing that bundle and saves it as an incomplete execution, so it can be retried automatically or resolved by hand.&lt;/p&gt;
&lt;h3 id=&quot;why-did-my-make-scenario-turn-itself-off&quot;&gt;Why did my Make scenario turn itself off?&lt;/h3&gt;
&lt;p&gt;Make switches a scenario off after the number of consecutive errors set in its settings. Add error handlers to the modules that fail, and keep incomplete executions so failed runs are not lost.&lt;/p&gt;
&lt;h3 id=&quot;does-rollback-undo-everything-the-scenario-did&quot;&gt;Does Rollback undo everything the scenario did?&lt;/h3&gt;
&lt;p&gt;No. Rollback reverts changes only in modules that support transactions, such as data stores. Actions in most apps, like a sent email or a created CRM record, cannot be undone.&lt;/p&gt;</content></entry><entry><title>How to Document an n8n Workflow for a Client</title><link href="https://www.vamimi.com/blog/how-to-document-n8n-workflows-for-clients"/><id>https://www.vamimi.com/blog/how-to-document-n8n-workflows-for-clients</id><published>2026-11-01T00:00:00Z</published><updated>2026-11-01T00:00:00Z</updated><author><name>VamimiAI team</name></author><summary>A handover (or handoff) checklist for n8n freelancers and agencies: what to write down, what to draw, and how to show a client what a failure does.</summary><content type="html">&lt;p&gt;To document an n8n workflow for a client, give them four things in plain words: what starts it, what it does step by step, what happens when a step fails, and what they must keep working (logins, sheets, webhook addresses). A diagram that shows every path, including the failure paths, does most of that work; a one-page handover document does the rest.&lt;/p&gt;
&lt;p&gt;This guide is a checklist you can follow for every workflow you hand over (some call it the handoff), whether you build for one client or fifty.&lt;/p&gt;
&lt;h2 id=&quot;why-the-workflow-file-is-not-documentation&quot;&gt;Why the workflow file is not documentation&lt;/h2&gt;
&lt;p&gt;The exported &lt;code&gt;.json&lt;/code&gt; file is a backup, not an explanation. A client who opens it sees node parameters and IDs. A screenshot of the canvas is better, but it shows node types (HTTP Request, Set, If) rather than what the business gets, and it never shows what happens when a node fails, because n8n draws only the paths you connected.&lt;/p&gt;
&lt;p&gt;That last part is what costs you later. Most nodes that call an outside service stop the whole run when that service is down, rate-limits you or loses its login. Unless the workflow has an error workflow set, nobody is told. The client finds out when a customer complains.&lt;/p&gt;
&lt;h2 id=&quot;the-handover-checklist&quot;&gt;The handover checklist&lt;/h2&gt;
&lt;p&gt;Work through these in order. Each one answers a question your client will otherwise ask you by email.&lt;/p&gt;
&lt;ol&gt;&lt;li&gt;&lt;strong&gt;What starts it.&lt;/strong&gt; The trigger in client words: &amp;quot;a new row in the Leads sheet&amp;quot;, &amp;quot;every weekday at 8:00&amp;quot;, &amp;quot;a payment in Stripe&amp;quot;. If it is a webhook, say which system calls it.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;What it does.&lt;/strong&gt; One line per step, named the way the client knows the apps: &amp;quot;Adds the lead to HubSpot&amp;quot;, not &amp;quot;HTTP Request 3&amp;quot;.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;What happens when each step fails.&lt;/strong&gt; For every step that calls another service: does it retry, does it continue, or does the run stop? Who is told?&lt;/li&gt;&lt;li&gt;&lt;strong&gt;What it depends on.&lt;/strong&gt; The accounts and credentials by name (never the values), the sheets and their column names, the webhook URLs, any limits on the plan it runs on.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;How to look after it.&lt;/strong&gt; Where failed runs show up (the Executions list in n8n), what is safe to change, and who to call.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;What was left out.&lt;/strong&gt; Anything you agreed not to build. It saves an argument about scope later.&lt;/li&gt;&lt;/ol&gt;
&lt;h2 id=&quot;what-the-exported-json-contains&quot;&gt;What the exported JSON contains&lt;/h2&gt;
&lt;p&gt;An n8n export carries the names and IDs of the credentials the workflow uses, and anything typed straight into a node, keys included. Before the file goes to a client, or anywhere else, check that every secret lives in a stored credential rather than in a node. Give the client the JSON as a backup they own, with the documentation beside it, not instead of it. If the documentation lives in a README or wiki, the diagram can go there as &lt;a href=&quot;/n8n-to-mermaid&quot;&gt;Mermaid text or a draw.io file&lt;/a&gt;.&lt;/p&gt;
&lt;h2 id=&quot;draw-it-including-the-failure-paths&quot;&gt;Draw it, including the failure paths&lt;/h2&gt;
&lt;p&gt;A diagram answers the first three questions at a glance, but only if it shows the failure paths. Draw the decisions with labelled exits (Yes / No, the names of Switch outputs), and for every step that can fail, a dashed line to what really happens next: a retry, an alert, or a red end that says the run stops.&lt;/p&gt;
&lt;p&gt;You can draw this by hand, or import the workflow into &lt;a href=&quot;/n8n-workflow-diagram&quot;&gt;VamimiAI&amp;#x27;s n8n workflow diagram&lt;/a&gt;. It reads the exported file as it is, with no AI guesswork: If and Switch branches are labelled, error outputs are drawn dashed, and each node that calls an outside service with no retry and no error path leads to a red end saying the run stops and whether anyone is told.&lt;/p&gt;
&lt;aside class=&quot;try-card&quot;&gt;&lt;p&gt;Try it in VamimiAI&lt;/p&gt;&lt;a href=&quot;/app?q=Design%20an%20n8n%20workflow%20that%20sends%20new%20form%20leads%20to%20HubSpot%20and%20Slack%2C%20with%20every%20failure%20path&quot;&gt;&lt;q&gt;Design an n8n workflow that sends new form leads to HubSpot and Slack, with every failure path&lt;/q&gt;&lt;span&gt;Try this prompt →&lt;/span&gt;&lt;/a&gt;&lt;small&gt;The prompt goes into the message box for you to edit; nothing is sent until you press Send.&lt;/small&gt;&lt;/aside&gt;
&lt;h2 id=&quot;check-it-before-you-hand-it-over&quot;&gt;Check it before you hand it over&lt;/h2&gt;
&lt;p&gt;Documenting a workflow is the moment you notice what will break. Before the handover, look for:&lt;/p&gt;
&lt;ul&gt;&lt;li&gt;&lt;strong&gt;Webhooks without authentication.&lt;/strong&gt; Anyone who finds the address can start the workflow.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Keys typed into a node&lt;/strong&gt; instead of stored as a credential. They travel with every export of the workflow.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;If exits that lead nowhere&lt;/strong&gt;, and nodes nothing reaches.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;No error workflow&lt;/strong&gt; in the workflow settings, so failures are silent.&lt;/li&gt;&lt;li&gt;&lt;strong&gt;Default node names&lt;/strong&gt; like &amp;quot;HTTP Request1&amp;quot;, which make the execution log unreadable for the next person.&lt;/li&gt;&lt;/ul&gt;
&lt;p&gt;VamimiAI runs these checks on import and lists them most serious first, with the fix for each. A secret is reported by the name of its field only; its value is never shown.&lt;/p&gt;
&lt;h2 id=&quot;write-the-handover-document&quot;&gt;Write the handover document&lt;/h2&gt;
&lt;p&gt;Put it all on one page the client can keep: the diagram first, then a table of what each step does and what happens if it fails, the accounts it needs, and short notes on looking after it. Add your name and contact details so they know who to call.&lt;/p&gt;
&lt;p&gt;The &lt;a href=&quot;/automation-handover-document&quot;&gt;automation handover document&lt;/a&gt; in VamimiAI builds exactly this page from an imported or described workflow. It is put together in your browser, and you can print it, save it as PDF or download it as a web page. The list of problems found is left out unless you choose to include it.&lt;/p&gt;
&lt;p&gt;For the failure paths specifically, the &lt;a href=&quot;/blog/n8n-error-handling-guide&quot;&gt;n8n error handling guide&lt;/a&gt; goes through each option n8n gives you and when to use it.&lt;/p&gt;
&lt;h2 id=&quot;questions&quot;&gt;Questions&lt;/h2&gt;
&lt;h3 id=&quot;what-should-an-n8n-handover-document-include&quot;&gt;What should an n8n handover document include?&lt;/h3&gt;
&lt;p&gt;The trigger, each step in plain words, what happens when each step fails, the accounts and sheets it depends on, where failed runs show up, and who to contact. A diagram that shows every path, including failures, should come first.&lt;/p&gt;
&lt;h3 id=&quot;should-i-give-the-client-the-workflow-json&quot;&gt;Should I give the client the workflow JSON?&lt;/h3&gt;
&lt;p&gt;Yes, as a backup they own, but not as the documentation. Keep credentials out of it: anything typed directly into a node travels with the export, while stored credentials do not.&lt;/p&gt;
&lt;h3 id=&quot;how-do-i-show-a-client-what-happens-when-a-step-fails&quot;&gt;How do I show a client what happens when a step fails?&lt;/h3&gt;
&lt;p&gt;Draw a dashed line from every step that calls an outside service to what really happens next: a retry, an alert, an error output or a red end where the run stops. Importing the workflow into VamimiAI draws those paths from the file.&lt;/p&gt;</content></entry></feed>
