Marketing automation

Make.com automation expert

For UK businesses choosing Make, or living with scenarios that break without warning. I design, build and document Make scenarios in your own account, with error alerts and a plain runbook, so the automation keeps working after I hand it over.

Make (the platform once called Integromat) is a visual automation tool. You build a scenario on a canvas, one module at a time, and it passes data between the apps your business already runs on: your website forms, CRM, accounting software, shop and spreadsheets. This page is for UK businesses that have picked Make, are deciding whether to, or have inherited scenarios that fail without anyone noticing. I design, build and document Make scenarios, and everything I build lives in your own account.

What Make does well

Make earns its place when a process has more than one path. A router can send the same record down several branches, each with its own filter, so a new enquiry can go to sales, to the email list and to a reporting sheet in one run. Iterators and aggregators handle lists properly, which matters for things like order line items or a batch of invoices.

It also reaches well beyond its list of ready-made app modules. The HTTP module can call almost any API, and a custom webhook lets a website form or another system push data into a scenario the moment something happens. When a supplier has an API but no Make app, that is rarely a dead end.

The run history is the other strength. Every execution is logged, and you can open each module to see exactly what data went in and what came out. When a lead goes missing, that turns an afternoon of guessing into a few minutes of reading.

Where Make is weaker

  • Polling costs add up. A trigger that checks an app for new records uses part of your allowance on every check, even when nothing has arrived. A scenario polling every five minutes runs 288 times a day before it has done any work.
  • Big scenarios get hard to read. The canvas is pleasant at ten modules and confusing at sixty. Without naming conventions and notes, the next person to open it has to reverse-engineer it.
  • It is hosted only. You cannot run Make on your own servers. Data passes through Make’s infrastructure, in the hosting region chosen when your organisation was set up.
  • Mapping takes some getting used to. Staff who are comfortable with a simple step-by-step editor often find arrays, bundles and field mapping in Make harder to maintain without help.

What I build on Make

Lead capture and routing

Enquiries from website forms and Facebook and Instagram lead ads land in your CRM as they arrive, with their source recorded, duplicates checked and an owner assigned by rule. The person enquiring gets an immediate reply, and someone on your side gets a task. The rules that decide who gets which lead are written down with you first, because lead routing that nobody agreed on gets bypassed.

Orders, invoices and fulfilment

New orders from Shopify or WooCommerce can create invoices in Xero or QuickBooks, alert a warehouse or courier, and flag low stock. These scenarios need careful handling of VAT, refunds and part shipments, so I test them against your real order history rather than one tidy example.

Keeping tools in step

When the same customer lives in your CRM, your email platform and a spreadsheet, Make can keep them aligned. A two-way sync needs the most care: without a clear rule about which system wins, two scenarios can update each other in a loop.

Scheduled reporting feeds

Ad spend, leads and sales can be pulled into a Google Sheet each morning so the numbers you look at are already joined up. Reporting only helps if the tracking underneath it is sound, so I check that first; this guide on testing whether your conversion tracking works covers the same checks.

AI steps with a person approving

Make has modules for the main AI model providers, which makes it practical to sort incoming emails, extract details from documents or draft a first reply. I build these with a human in the loop: the AI proposes, a named person approves, and nothing reaches a customer unchecked.

When I would recommend something else

Make is not always the right answer, and I would rather say so before you pay for it. The trade-offs are covered in more depth in the glossary entry on Zapier and Make, but in short:

  • For a handful of simple two-step automations that a non-technical team will maintain alone, Zapier is usually easier to live with.
  • If data must stay on servers you control, or volumes are high enough that usage pricing hurts, self-hosted n8n is worth comparing.
  • If the whole process happens inside Google Workspace, Apps Script may do it with no extra subscription.
  • If both ends of the process sit in one platform, such as HubSpot or Klaviyo, its own built-in workflows are simpler and cheaper to keep running.
  • If the process itself is still changing every week, it is too early to automate. Settle it by hand first.

How a build runs

  1. Map the process. We talk it through, then I write it down step by step: what starts it, which apps are involved, who acts at each point and what should happen when something fails.
  2. Set up access properly. The Make organisation is in your name and on your billing, and I join as a member. App connections are authorised with shared business accounts where possible, not someone’s personal login.
  3. Build and test on sample data. I build in a copy, then run it with awkward real-world records: blank fields, duplicate contacts, phone numbers written as 07 and +44, postcodes with and without spaces.
  4. Add error handling. Each scenario gets error routes, an alert to an inbox or chat channel you watch, and stored incomplete executions so a failed run can be fixed and replayed instead of lost.
  5. Go live and watch. For a period we agree before work starts, I check the run history after launch and tune schedules and filters.
  6. Hand over. A walkthrough with whoever will own it, plus the written documentation below.

Problems I see most often in inherited Make accounts

  • Scenarios sitting in a former freelancer’s personal account, so the business does not actually own them.
  • No error handling at all. Failures go unnoticed, and depending on its settings Make can stop scheduling a scenario after repeated errors.
  • Polling triggers set to run every few minutes on processes that need checking once an hour, quietly using up the plan.
  • Connections authorised by someone who has since left, which break the day their account is closed.
  • Customer details copied into data stores and Google Sheets and kept indefinitely. Under UK GDPR you should hold only what the process needs, for as long as it needs it; data minimisation applies to automations as much as to forms.
  • Modules left with default names, so nobody can tell what “HTTP 4” does without opening it.

What you receive

  • Working scenarios in your own Make organisation, consistently named, with a note on every module that is not obvious.
  • Exported blueprints of each scenario, stored with you as a backup.
  • A short runbook for each scenario: what triggers it, what it changes, where errors are reported, how to pause it and how to replay a failed run.
  • A list of the personal data each scenario touches and where it ends up, for your own records. Make processes that data on your behalf, so its data processing agreement and hosting region belong in the same file.

Pricing

There are two separate costs. The first is your Make subscription, which you pay to Make directly. Make prices by usage: each module action in each run draws on a monthly allowance, so the bill depends on how many steps your scenarios have and how often they run. Plan names, prices and allowances change, so check Make’s current price list before you choose. Before you commit, I estimate the monthly usage of the scenarios we plan, so the plan you choose fits.

The second is my fee for the build. I write each quote in GBP after a free first call, once I know which scenarios are needed, and we agree it in writing before work starts, along with any support after launch. I do not resell Make, add a margin to your subscription or hold your account on my side. If we stop working together, you keep everything and I remove my access.

Next step

Send me a short description of the process you want automated, the apps involved and roughly how many records go through it in a month. If you already have Make scenarios, tell me what they are meant to do and what keeps going wrong. You can book a free 30-minute call and I will tell you whether Make is the right tool for it, and if not, what I would use instead.

Frequently asked questions

Is Make the same thing as Integromat?

Yes. Integromat was renamed Make in 2022. Older guides, forum answers and notes from a previous freelancer may still use the old name; they describe the same product, although the interface and some module names have changed since.

Can you take over scenarios someone else built?

Yes. I start by reviewing what exists: what each scenario does, whose account and connections it depends on, and how failures are handled. Sometimes a few fixes are enough. Sometimes a tangled scenario is quicker and safer to rebuild, and I will tell you which before any work starts.

Is it safe to put customer data through Make under UK GDPR?

It can be, but no tool is compliant on its own. You remain the controller and Make acts as your processor, so you need its data processing agreement in place, a sensible choice of hosting region and scenarios that only move the fields they need. For a UK business I normally recommend the EU region, which the UK recognises as providing adequate protection.

Will Make be cheaper than Zapier for us?

It depends on the shape of your automations, not on the headline price. Multi-step scenarios with branches often cost less on Make, while a few simple two-step automations may cost about the same anywhere. I count the steps and expected runs for both before recommending one.

Who owns the scenarios if we stop working together?

You do. They are built in a Make organisation registered and billed in your name, with your own app connections, and you have exported blueprints as a backup. When the work ends I remove my membership and nothing stops running.

Can Make send data to ChatGPT or other AI tools?

Yes, Make has modules for the main AI model providers and can call others through their APIs. Remember that the AI provider becomes another processor of whatever you send it, so I keep personal data out of prompts where the task allows and put a person in the approval step before anything reaches a customer.

Ready to talk about your project?

A straight answer about what would move the numbers, and a written proposal if we are a fit.