Most businesses that hire me for automation work have the same problem: they’ve tried to automate before, it didn’t stick, and now they’re not sure why. Usually the answer is simple. They automated the wrong things first, or they automated a broken process and just made the broken process faster.
Business process automation isn’t complicated in principle. You find the work that repeats on a predictable schedule, you remove the human from the middle of it, and you redirect that time toward work that actually requires judgment. What makes it hard in practice is the sequencing — knowing which processes to touch first, which tools to use, and what the thing needs to look like before a machine can run it reliably.
I’ve built 80+ automations across businesses doing $500K–$10M. This is the guide I give clients at the start of an engagement: what business process automation actually is, which five processes move the needle fastest, what tools do what, and what build order produces results instead of half-finished workflows that nobody maintains.
What is business process automation?
Business process automation (BPA) is the use of software to execute recurring business tasks without human intervention. When a client signs a contract and a folder structure appears, a Slack message fires, and a welcome email goes out — without anyone on your team doing anything — that’s business process automation.
It’s distinct from regular software. Every SaaS tool you use involves some automation. BPA means deliberately designing workflows that connect your tools and trigger actions based on defined events, so that the operational layer of your business runs on rules rather than people.
The scope ranges from simple (a Zapier trigger that copies a new row from one sheet to another) to complex (a multi-branch Make.com scenario that handles client onboarding across seven tools without a single manual step). For businesses doing $500K–$10M, you almost never need the simple stuff — and you almost always need something closer to the complex end.
The 5 processes every small business should automate first
Not every automation has the same ROI. These five show up at almost every business I work with, and they’re ranked by time savings and downstream impact.
1. Client onboarding
Onboarding is the highest-leverage automation for most service businesses because it happens with every client and it’s almost always overly manual. The typical pre-automation version: send a questionnaire by email, wait for it back, build a folder, create an account somewhere, set up a Slack channel, update a tracker. Six to eight steps, all done by a person, for every new client.
Nomads Cast had exactly this process — eight manual steps per client, each touching a different tool. We replaced it with a Make.com automation connecting Notion, Go High Level, Jotform, Google Drive, Gmail, and Slack. Client enters the system, everything fires in sequence, zero human steps. Onboarding time dropped from hours to under a minute. The team recovered 5 hours per client — time that previously went to setup and coordination.
Onboarding automation is worth doing first because the ROI compounds. Five hours saved per client means the 10th client saves you 50 hours compared to the manual alternative.
2. Lead follow-up and client retention
The second process is follow-up — the messages your team sends after a meeting, after a visit, when a client hasn’t been heard from in 30 days. Manual follow-up is inconsistent (it happens when someone remembers) and slow (it goes out the next day, not within the hour). Both problems cost you deals.
Scenthound’s team was manually managing all post-visit communication and lapsed-member re-engagement across a growing franchise. We rebuilt their CRM logic around client lifecycle states and automated every routine touchpoint — post-visit messages, 25-day re-engagement sequences, cancellation win-back flows. The result was 40 hours per week recovered from manual follow-up tasks and 80% faster response times. Mike Schoen, COO, described it as saving a full-time equivalent of manual work.
Workflow automation for lead follow-up is the difference between a CRM that stores data and a CRM that drives revenue from the data it already has.
3. Internal notifications and status updates
Someone finishes a deliverable and doesn’t tell the next person in line. A project moves to a new stage but the project tracker doesn’t reflect it. A client pays an invoice and the account manager doesn’t know for two days. These are small failures individually; they add up to a business where coordination happens via Slack messages and memory rather than systems.
Automating internal notifications means defining: when X happens in system A, notify person B and update system C. No manual handoffs. No one has to remember to tell the next person. Status is always current because the update is automated at the moment the trigger fires.
4. Invoice and payment reminders
Collections are one of the most predictable processes in any service business — and one of the last to get automated, because it feels awkward and founders put it off. Once it’s set up, a payment reminder sequence requires nothing from your team: a reminder goes out at 3 days, another at 7, another at 14. If payment clears, the sequence stops automatically.
The time savings are modest, but this automation pays for itself in the first month and runs forever.
5. Reporting and data aggregation
Every business I’ve audited has reporting that either doesn’t happen or happens manually once a week when someone pulls data from three tools and builds a spreadsheet. Automated reporting means the numbers are always current, the dashboard always reflects reality, and nobody spends Friday afternoon doing data entry.
This one tends to get built last because it requires the first four to be working. You can’t automate reporting on a process that isn’t running cleanly yet.
Tools for business process automation (and when each makes sense)
The three tools worth knowing for small business automation are Make.com, Zapier, and Airtable’s native automations. Here’s when each is the right choice.
Make.com is the tool I use for almost everything complex. It handles multi-step, multi-branch workflows well, the visual builder is intuitive once you understand the logic, and the pricing is dramatically more favorable than Zapier at volume. The Nomads Cast and Scenthound automations were both built on Make. If you’re doing anything that involves more than two or three tools or any conditional logic, Make is the right choice. Our full breakdown of Make.com automation work covers this in more detail.
Zapier is more expensive at equivalent operation volume and handles complexity less gracefully, but it has better native integration with certain tools and a lower barrier to entry for non-technical users who want to set up simple workflows themselves. If your automation needs are simple and you want to manage them in-house, Zapier works. If you’re hiring someone to build a real automation stack, Make almost always makes more financial sense.
Airtable native automations are underused. If your data already lives in Airtable, you can automate a surprising amount without touching Make or Zapier — record-triggered emails, status updates, notifications, basic conditional logic. The limitation is scope: Airtable automations don’t connect well to external tools. For anything that needs to span systems, you’ll want Make on top of it.
For businesses doing $500K–$10M, the typical setup is Airtable as the data foundation with Make.com connecting it to everything else. That combination handles 90% of business process automation use cases without requiring anything more sophisticated.
How much does business process automation cost?
Ranges depend heavily on complexity. Here’s an honest breakdown.
DIY tool cost: Make.com starts at $9/month, scales to $16–$29/month for most small businesses. Zapier is $19.99–$49/month for comparable usage. Airtable runs $20–$45/user/month depending on tier. If you’re building simple automations yourself, monthly tool cost is $50–$150.
Automation project fees (what you pay a consultant to build it):
- Simple automation (1-2 tools, straightforward trigger-action): $1,500–$3,000
- Medium complexity (3-5 tools, conditional logic, custom Airtable base): $3,000–$6,000
- Full automation layer (multi-tool stack, Make.com scenarios, integrated with data architecture): $6,000–$12,000
The ROI math usually closes fast. Nomads Cast was saving 5 hours per client. If they onboard 20 clients a year at an average hourly cost of $40, that’s $4,000 in recovered labor from a single automation. A project at $3,000 pays back in under a year and compounds indefinitely. Scenthound recovered a full-time equivalent — 40 hours per week, every week.
The automation and workflow integration page covers what a full engagement looks like and what you should expect at each complexity tier.
The build order that actually works
I’ve seen businesses skip this and regret it. The sequence matters.
Step 1: Fix the data first. Automations run on data. If your data is incomplete, in the wrong format, or spread across five places, your automations will fail silently or produce wrong outputs. Before you automate anything, get your data architecture right — one source of truth, clean records, fields that mean the same thing every time. This is the operations consulting work that has to happen before the automation work.
Step 2: Document the process before you automate it. A machine will do exactly what you tell it, including executing a broken process at scale. Map the workflow first: every step, every decision point, every tool involved. Then look for what doesn’t need to be there. Automate the simplified version, not the current one.
Step 3: Start with high-frequency, low-complexity processes. Client onboarding and follow-up sequences run constantly and have predictable inputs. They’re the right place to start. Don’t begin with the most complex automation in your business — begin with the one that runs the most often, produces the most savings, and has the cleanest inputs.
Step 4: Connect your tools before adding logic. Get Make.com talking to your CRM, your email platform, your project management tool. Verify data is flowing correctly. Then layer in the conditional logic and multi-branch sequences. Building complexity before the connections are stable means debugging two problems at once.
Step 5: Layer AI last. AI-powered automation — document processing, intelligent routing, voice agents — only works well on top of clean data and tested workflows. If you try to add AI before the foundation is solid, it amplifies the problems in your data rather than solving them. The AI readiness assessment is a good place to check whether your business is ready for this layer.
Common mistakes that kill automation projects
Automating a broken process. The most common one. If your onboarding process has six unnecessary steps and two approval gates that add no value, automating it makes those steps happen faster, not better. Simplify first.
Starting with the wrong automation. Every business has one automation they think is most important and three that would actually save more time. The “most important” one is usually the most complex and the hardest to get right. Build confidence with the high-frequency, high-ROI processes first. The complex ones are easier after you understand how the system behaves.
Not documenting the logic. Automations built without documentation are a liability. When something breaks six months later — and something always breaks — the person who fixes it needs to understand the logic, not just the steps. Document every scenario, every conditional branch, every error state.
Choosing the wrong tool for the job. Using Zapier for something that needs Make.com is like using a spreadsheet for a database problem — it works until it doesn’t, and then it fails expensively. Know your tools before you build.
Building automations that nobody understands. If your team doesn’t understand how the automation works, they won’t trust it and they’ll work around it. Build with your team, not for them. Train them on the logic, not just the interface.
Business process automation done right is infrastructure — it runs in the background, reliably, without anyone managing it. Done wrong, it’s a collection of half-maintained workflows that creates more problems than it solves.
If you’re not sure where to start or which processes in your business are worth automating first, book a call. We’ll map your current workflows, identify the 5–10 highest-ROI automation opportunities, and tell you what order to build them in — before you commit to anything.
