HowMuchDoesBusinessAutomationCost?APlainPricingGuide

What drives the cost of an automation

Two automations that sound similar can take very different amounts of work. The price follows the work, and the work mostly comes from four things.

A useful rule: the more clearly you can describe the task as "when this happens, do that", the cheaper it is to build. Every "it depends" in your description adds time.

• How many tools are connected. Moving a form entry into a Google Sheet is one connection. Moving it into a sheet, a CRM, an accounting tool and a WhatsApp message is four, and each one has to be set up, tested and kept working.
• How messy the data is. If phone numbers are typed in five formats, customer names are spelled differently in each system, or half the rows have blanks, the automation needs cleaning steps before it can do anything useful.
• Whether AI is involved. Rule-based steps are predictable and quick to test. Steps where an AI model reads an email, sorts a request or drafts a reply need more testing, a check for wrong answers, and they add a usage fee.
• How many exceptions need a human. Refunds over a limit, VIP customers, orders with missing details. Each exception needs a rule for when the automation stops and who it hands the case to.

The three ways automation is sold

Most agencies and freelancers sell automation in one of three ways. None of them is wrong. Each one moves the risk to a different place.

Moonlab uses the first two. A single automation starts from $99 as a one-time project. The monthly plan is $199 per month, which covers up to 2 new automations a month with monitoring included, and there is a larger plan at $399 per month. Every project starts with a free 30-minute audit, and the fixed price is agreed before work starts. Current details are at moonlab.space/pricing.

• One-off fixed-price project. You agree the scope and the price, and the work is delivered for that price. You know the total before you start, and the builder carries the risk if it takes longer. The trade-off is that the scope has to be written down clearly, and changes after that are a new conversation. It also does not cover what happens when something breaks six months later, unless support is included.
• Monthly plan. You pay a fixed amount each month for a set amount of new work plus monitoring and fixes. This suits a business with a list of tasks to automate over time, or one that wants someone watching the automations. The trade-off is that you pay in quiet months too, so it only makes sense if you will use it.
• Hourly. You pay for time spent. It is flexible and works for small fixes or work that cannot be scoped yet. The trade-off is that you carry the risk. The total is unknown at the start, and a slow builder costs you more than a fast one.

The running costs people forget

The build price is only part of the cost. An automation runs on other companies' tools, and those have their own charges. We do not quote them here because they vary by plan, by country and by volume, and they change. Check each vendor's own pricing page before you commit, not a number in a blog post.

Ask for these costs to be listed in the proposal, next to the build price, so you can see the full monthly figure before you agree.

• Tool subscriptions. Automation platforms such as Zapier, Make or n8n, plus any CRM, form or inbox tool the workflow depends on. Many have tiers based on how many tasks you run a month, so your volume decides which tier you need.
• Message and API usage fees. WhatsApp template messages, SMS, email sending and AI model calls are usually charged by use. A workflow that runs 50 times a month and one that runs 5,000 times can have very different bills.
• Maintenance when a tool changes. Tools update their apps, rename fields and retire old connections. When that happens the automation can stop or, worse, keep running with wrong data. Someone has to notice and fix it, and that time is a cost whether you pay an agency or do it yourself.

How to work out whether it pays back

You can test any automation idea with one sum. Take the hours it saves per week, multiply by what an hour of that person's time costs you, and compare the yearly total against the build cost plus the running cost.

Here is a made-up example to show the arithmetic. Say a task takes 5 hours a week, and the automation removes 4 of them because 1 hour of checking and exceptions remains. Say that person's time costs you $15 an hour. The saving is 4 x $15 = $60 a week, and $60 x 52 weeks = $3,120 a year. If the build were a one-time project at a fixed quote, the payback time is the quote divided by $60. At the $99 starting price that is 99 / 60 = 1.65, so under two weeks. On a $199 monthly plan the yearly cost is $199 x 12 = $2,388, which leaves $3,120 - $2,388 = $732 a year from this one task, before tool fees. That is a thin margin for one task, and it shows why a monthly plan makes sense when you have several tasks to automate, not one.

Then subtract the running costs from the section above. If the number is still clearly positive, the project is worth doing. If it only works when every estimate goes your way, it is not. The same sum works in taka. Use your own hours and your own wage figure, not ours.

• Hours saved per week x hourly cost = weekly saving.
• Weekly saving x 52 = yearly saving.
• Build cost + (monthly running cost x 12) = first-year cost.
• Yearly saving minus first-year cost = what you actually keep in year one.

What to ask an agency before signing

A good proposal answers most of these without being asked. If the answers are vague, expect the invoice to be vague too.

• Is the price fixed, and what exactly is included in it?
• Which tool subscriptions and usage fees will I pay on top, and who pays them, me or you?
• Whose accounts will the automation run in? It should be yours, so you keep it if we stop working together.
• What happens when it breaks? Who finds out, how fast, and is the fix included?
• What does the automation do when it meets a case it cannot handle?
• How will it be tested before it touches real customers or real data?
• Will I get a written description of how it works, in plain language?
• What does a change cost after delivery?

Mistakes that make projects expensive

Most cost overruns come from decisions made before any building starts. These are the common ones.

• Automating a process that is not settled. If the steps change every week, you pay for the build and then pay again for each change. Fix the process on paper first.
• Starting with the biggest, most complicated workflow. Start with one small task that runs often. It pays back sooner and teaches you what the next one should be.
• Vague scope. "Automate our sales" cannot be priced. "When a lead fills in the form, add it to the CRM and send a WhatsApp reply within a minute" can.
• Skipping the data clean-up. Messy data does not go away when you automate. It just moves faster.
• Using AI where a simple rule would do. It adds usage fees and testing time for no gain.
• Trying to automate every exception. Handling the common cases and handing the rare ones to a person is usually far cheaper than covering everything.
• No owner after launch. If nobody is responsible for the automation, small faults sit unnoticed until they become costly ones.

Get a fixed price for your own case

The quickest way to get a real number is to describe one task to someone who builds these for a living. Book a free 30-minute audit at moonlab.space/contact. We look at the task, tell you whether it is worth automating, list the running costs you should expect, and give you a fixed price before any work starts. If the sum does not work in your favour, we will say so.

Related services: operations automation at moonlab.space/services/operations, custom builds at moonlab.space/services/custom-build, and sales and lead follow-up at moonlab.space/services/sales-and-leads.