When can you easily build a Make.com automation yourself?

Make is designed to be used without programming knowledge, and that really works. With a few hours of practice you can click together a scenario that passes a form on to your CRM or sends a message when an order comes in. Make has its own free learning environment for this, the Make Academy.

Building it yourself is a good choice when most of these points apply:

  • The scenario has a few steps and goes from one system to one other system.
  • Little is at stake. If it stops for a day, you notice and pick it up by hand.
  • Both systems have a ready-made app in Make. So you do not have to work with an API yourself.
  • You or a colleague has time for it on an ongoing basis. Not just to build it, but also to look at it later when something changes.

For scenarios like these, outsourcing often costs more than it delivers. Start small, see what it saves you and learn how Make thinks along the way. Ideas for a first scenario are in Make.com workflows for SMEs.

When does it pay to have your automation built?

Building the "happy path" is rarely the problem. The work is in everything that deviates: a customer who already exists, an empty field, an API that briefly does not respond, an invoice that comes in twice. Help pays off when one of these situations applies:

  • The scenario touches money or customer data. You only see a mistake in an invoice or order flow when a customer calls or the books do not balance.
  • More than two systems are involved, or one system first has to look something up in the other before it can write.
  • There is no ready-made app and you have to talk to an API through an HTTP module or webhook. Then you are dealing with authentication, pagination and limits.
  • The volume is high. Then the setup matters for your credit usage, and so for your bill. How that works is covered in what Make.com automation costs.
  • Nobody in the business has time for it. A half-built scenario that someone maintains alongside their actual job is often the most expensive option.

There is also a middle way: you build it yourself and have someone with experience review your setup and error handling before it goes live. That takes little time and prevents the best-known mistakes.

How do you make the trade-off concrete?

Take a scenario you want to automate and answer these five questions. The example below is fictional.

  1. What does the manual work cost now? Minutes per time multiplied by the number of times per month.
  2. What happens if the scenario stops for a day? Nothing, an annoyance, or a customer who notices?
  3. How many systems and exceptions are involved? Write them down before you start.
  4. Is there a ready-made app for every system? Check this in Make's overview.
  5. Who looks at it when it goes wrong a year from now? Name a person, not a department.

Example. An installation company with twenty employees wants to automate two things. First: an alert in Teams when a new request comes in through the website form. That is two steps between two systems with a ready-made app, and if it stops for a day, someone simply checks the mailbox. The office manager builds that one.

Second: turning job sheets from the planning software into draft invoices in the accounting package, with the right customer, the right item and the right VAT rate. That is three systems, money is involved and the planning software has no app in Make, so it has to go through the API. That is where a builder pays off, with agreements on maintenance.

What should you look for when you have a Make.com automation built?

A scenario that works on delivery is not yet a good scenario. Agree on these points upfront:

  • It lives in your account. Have the scenario built in a Make organisation in your own company's name, and give the builder access to it. That way your scenarios, connections and logs stay yours, even if the collaboration ends.
  • Connections use company accounts. A connection that runs on the builder's personal account stops when that account disappears. Use a fixed company account or a service account per system.
  • Error handling is part of the job. Ask what happens when a system does not respond, and who gets alerted. How this works in Make is covered in Make.com error handling.
  • There is documentation. Which systems, which steps, which exceptions and what to do when it goes wrong. Short is fine, as long as it exists.
  • You get an export. Make can export a scenario as a blueprint, a JSON file. Keep it after delivery, so you always have a copy.
  • It is agreed who maintains it. Passwords expire and APIs change. Agree whether the builder does the maintenance or you take it on yourself, and how quickly someone responds when it stops.

How do you recognise a good builder?

Make has a partner programme for agencies and consultants, with tiers that depend on, among other things, certified people and previous projects. That says something about experience with the platform. It says less about whether someone understands your process. So also look for these signals:

  • They ask about the process first, not the tools. Anyone who starts talking about modules straight away will end up automating exactly the detour you now take by hand.
  • They start with one clearly defined scenario. A big plan for ten automations at once sounds efficient, but you only learn what works once the first scenario is running.
  • They also tell you when Make is not the best choice. For example when n8n fits better, or when it is really about reporting and the data should go to a database in bulk. That trade-off is covered in Make vs n8n.
  • They show how they test. With real examples, including the odd cases, not just with one test record.
  • They are clear about the price. A fixed price per defined phase, or a clear estimate in hours, with maintenance named separately.

The same questions apply to whoever you hire for dashboards, by the way. See also hiring a Power BI consultant.

What does a good collaboration look like?

A practical setup that works for most SME processes has four steps:

  1. Walk through the process. Together with the person who does the work now: what happens, in what order, and where does it deviate?
  2. Build one scenario. With error handling and alerts included, tested on real cases.
  3. Hand over. Documentation, a short explanation for the people who work with it and the blueprint as a backup.
  4. Agree on maintenance. In-house or outsourced, but with a name attached.

Only when the first scenario has run for a few weeks without trouble is it time for the next one. That way the number of automations grows with what your organisation can handle. Which processes qualify is covered in workflow automation for SMEs.

Need help with your Make.com automation?

Not sure whether you can build a scenario yourself, or want someone to review your setup before it goes live? We build workflows with Make.com and also work with n8n. We start with one clearly defined workflow, and after the intake you receive a proposal with a fixed price for that phase. See what we do in workflow automation, which integrations we often build, or read how we work.