Automation: Build vs Buy vs Hire, and What Each Really Costs

A workflow finally got bad enough to fix. Leads come in through three forms, someone re-keys each one into the CRM, tags it, and pings sales in Slack. Last week two of them slipped through, and nobody noticed until the follow-up was a week late. So you sit down to solve it and hit a fork: wire it together yourself on a tool you half-know, buy some product that claims to do the whole thing, or hire someone whose actual job is this kind of plumbing.

Each path has a sticker price and a real price, and they are almost never the same number. A $19.99 Zapier plan is not the cost of building on Zapier. A six-figure salary is not the cost of a hire. Get the two confused and a lean team picks the option that looks cheapest, then pays for it for a year.

What build, buy, and hire actually mean here

The three words get used loosely, so pin them down first. For automation, they map to three different ways of getting a workflow running.

• Build. Your existing team wires it together on a no-code platform like Zapier, Make, or n8n, or writes custom scripts. You own the tool subscription and every hour of building and fixing.
• Buy. You purchase an off-the-shelf product that already does the specific job, a purpose-built SaaS tool, and configure it. You rent an outcome someone else maintains.
• Hire. You bring on a full-time person, an ops generalist or an automation engineer, to own the plumbing across your whole stack.

There is a fourth path that sits between buy and hire, and it is the one most teams skip: pay a studio to design and build the workflow for you, without carrying a full-time salary. That is what bottta does, and for most lean teams it is the honest recommendation. We will get to why the math favors it, but first the three classic options, priced.

Build it yourself: cheap sticker, expensive reality

Building it yourself looks like the frugal choice because the tool prices are small. Here is what the main no-code platforms charge right now.

| Platform | Entry paid plan | Included volume | Billing unit | |---|---|---|---| | Zapier | $19.99/mo (billed annually) | 750 tasks/mo | A task is each successful step in a Zap | | Make | $9/mo (Core) | 10,000 operations/mo | One credit per module action | | n8n Cloud | $20/mo (Starter, billed annually) | 2,500 executions/mo | One execution is a full workflow run | | n8n self-hosted | Free, open source | Unlimited (your server) | You run and maintain it |

Sources: Zapier's pricing page, Make's pricing page, and n8n's pricing page.

Note how differently they count. Zapier bills per task, and a task is each successful action, so a five-step Zap burns five tasks every time it runs, and many actions cost more than one. Make bills per operation, roughly one credit per module. n8n bills per execution, a whole workflow run regardless of how many steps it has, which is why complex flows get much cheaper there. Pick the wrong billing model for your shape of work and a workflow that fires thousands of times a month can quietly cost more than you planned. We ranked these platforms in detail in the best workflow automation tools for lean teams and the best no-code automation platforms.

The subscription is the small number. The real cost of building is your team's time, and it does not stop at launch. Someone maps the process, handles the lead whose company name has a comma in it, adds error handling so a failure does not vanish, and then owns the thing forever. Tools change their APIs. A vendor CSV grows a column. The automation needs upkeep, and the person who built it is usually the only one who understands it. Build makes sense when the workflow is simple, the inputs are stable, and you have someone who genuinely has the hours. It stops making sense the moment "quick automation" becomes a second job nobody signed up for.

Buy an off-the-shelf tool: right when your problem is common

Sometimes the smartest move is to buy a product that already solves your exact problem and skip the workflow entirely. If you need scheduling, buy a scheduling tool. If you need e-signatures, buy an e-signature tool. A purpose-built product is maintained by a vendor, updated without you, and usually cheaper than building the same thing.

Buy works when your need is common and your process can bend to fit the tool. It breaks down fast when your process is specific to how you operate. Off-the-shelf products assume a standard flow. The second you need "route it to this queue only when the deal is over $5,000 and the contact is in HubSpot but not in Stripe yet," no product does that, because that logic is yours. You end up buying the tool and still building the glue that connects it to everything else. That glue, the integrations and the conditional logic between your apps, is the actual work, and it is exactly the part a product does not sell you.

The trap is buying a product and assuming the integration comes for free. It rarely does. Most teams end up with a shelf of subscriptions that each do one thing well and none of which talk to each other, which is the copy-paste problem wearing a nicer logo.

Hire someone to own it: only past a real threshold

Hiring a full-time person to own automation gives you something the other paths do not: someone who holds the whole system in their head and is there when it breaks. That is genuinely valuable. It is also the most expensive option by a wide margin, and the number people quote themselves is too low.

In the US, the Bureau of Labor Statistics puts the median annual wage for software developers at $133,080 as of May 2024. Even in lower-cost regions, a capable full-time automation hire is a five-figure-plus annual commitment, and base salary is only part of it. Add payroll taxes, benefits, equipment, and the weeks of ramp before they ship anything. Then add the risk that a generalist ops hire can run a handful of Zaps but cannot write the custom integration you actually needed, or that a strong engineer is bored and gone in a year because there is not enough deep work to hold them.

A hire pays off when automation is constant and central: you have a steady backlog of new workflows, ongoing systems to maintain, and enough volume that one person is busy with it every week. For most teams under 50 people, that threshold is further away than it feels. You do not need a full-time salary. You need a specific set of workflows built well and kept running.

The fourth option: buy the build from a studio

This is where the math usually lands for a lean team, and it is why we recommend it first. A studio like bottta designs and builds the workflow for you, then keeps it running, without you carrying a full-time salary or teaching yourself a platform you will half-forget in a month.

The difference from "build" is that you are not spending your team's hours or inheriting the maintenance. The difference from "hire" is the commitment: no $133K salary, no recruiting, no risk of a bad hire, and the work starts this week instead of after a two-month search. The difference from "buy a product" is that the workflow fits how you actually operate, integrations and edge cases included, because we build it around your process instead of bending your process around a tool.

Concretely, there are two ways to work with us:

• The $4K project is fixed scope and fixed price. We map the process, build the integrations, handle the inputs that do not fit the happy path, and hand it over with 30 days of post-launch support. This is the right call when you have one clear workflow that needs to exist and run reliably.
• The $3K/month retainer keeps up to three active workflows built and monitored at a time, with flexible hours, async access, and a weekly call. This is for teams with a steady backlog, the ones who would otherwise be tempted to hire.

The design is the durable part, not the wiring. Clear triggers, sane handling for the record that does not fit the shape, and monitoring that tells a human when something needs attention instead of failing silently at 2am. That is the difference between an automation that saves time for years and one that becomes a liability in a month. If you want the deeper comparison of studios and in-house options, we wrote how to pick an AI automation agency for a growing team and in-house vs outsourced automation.

Worked example: one workflow, three ways to pay for it

Take the lead-to-CRM-to-billing workflow from the top. A form submission creates a CRM contact, tags it, posts to Slack, and when the deal closes, generates the Stripe invoice and drops it in a finance sheet. Call it a few hundred runs a month across five or six steps. Here is roughly what each path costs.

| Path | Up front | Ongoing | Who maintains it | |---|---|---|---| | Build it yourself | Your team's build time | $19.99-$29/mo tool plan, plus your team's fix time | You, indefinitely | | Buy a product | Setup time | Multiple subscriptions plus the glue you still build | You, for the glue | | Hire in-house | Recruiting + ramp | $133,080+/yr fully loaded (US median base) | Them, if they stay | | bottta project | $4K fixed | 30 days support, then optional retainer | Us |

The tool subscriptions are real and small. The number that decides this is the time, and time is where DIY quietly loses to a fixed-price build. Spending an engineer's month, or your own, wiring and re-wiring a workflow that a studio ships in days is the expensive choice dressed as the cheap one. And a full-time hire for one workflow is not a comparison, it is an order of magnitude too much. The honest read is that build wins for simple and stable, hire wins only at real ongoing volume, and for the wide middle, a fixed-price build is the cheapest path to something that actually runs.

How to decide

Match the path to your situation instead of your instinct.

• One simple, stable workflow and spare hours on the team. Build it yourself on Make or n8n. Keep it small, add error handling, and write down how it works so it does not live in one person's head. Our test for whether a task is even worth automating is in when to automate a task, and when not to, and first processes to automate covers which one to start with.
• A common problem a product already solves. Buy the product. Do not rebuild scheduling or e-signatures. Just budget for connecting it to the rest of your stack.
• One or a few workflows that need to work reliably and fit your process. Bring in bottta on the $4K project. Fixed price, integrations included, running in days.
• A steady backlog of automation and systems to keep alive. The retainer at $3K/month, or a full-time hire once the volume genuinely fills a role. Start with the retainer, because it is cheaper, faster to start, and easy to stop if the backlog dries up.

The mistake that costs the most is defaulting to build because the tool is cheap, then paying for it in months of your own time. The second most expensive is hiring for volume you do not have yet.

Frequently asked questions

Is it cheaper to build automation in-house or hire an agency? For a small number of workflows, an agency or studio is almost always cheaper in total cost, because the sticker price of building in-house hides the time to build, fix, and maintain it. A fixed-price studio build like bottta's $4K project is a known number, where in-house build time is open-ended and lands on your team.

When does hiring a full-time automation person make sense? When automation is constant and central: a steady backlog of new workflows plus ongoing systems to maintain, enough to keep one person busy every week. Below that, the US median software developer wage of $133,080 buys far more value spread across a studio retainer than parked in a single underused role.

Can I just use Zapier and skip all this? For a simple, stable, low-stakes workflow, yes. Zapier starts at $19.99/month billed annually for 750 tasks, and it is genuinely good. The reason teams outgrow it is not the tool, it is the edge cases, the integrations it does not cover, and the maintenance when an API changes. That is the work a build path underestimates.

What does bottta cost compared to a tool subscription? More than a subscription, less than a hire. A no-code plan runs $10 to $30 a month but only rents you the canvas. bottta's $3K/month retainer or $4K project covers the design, the build, the integrations, and the monitoring, which is the part that decides whether the automation actually runs.

Weigh it by total cost, not sticker price, and the middle path wins more often than a lean team expects. The honest version of this call sometimes ends in build it yourself or buy a product, and we will say so when it does. When it lands on a build, that is the one we are here for. Book a call.