We are an automation firm writing a guide about hiring automation firms. You should read this with appropriate suspicion — so we have written it as the checklist we would want a friend to use on us.
“Small business automation consultant” is not a regulated title. It covers a solo operator wiring up Zapier for $2,000, a boutique firm building custom software for $200,000, and a reseller whose real business is the software license they will place with you. All three will use the same words on their website. Most of the money wasted in this category is wasted by hiring the wrong category of help, not by hiring a bad firm.
So this guide is about telling them apart, what fair pricing looks like, and the specific questions that reveal which one you are talking to.
The Title Covers Four Different Jobs
Before you can evaluate anyone, you need to know which of these you are actually buying. Firms will happily let you assume they do all four.
The expensive mistake is hiring a custom build firm for a tool-connection problem, or hiring a Zapier specialist for something that genuinely needs a database and an audit trail. The second mistake is more common and more painful, because it usually surfaces a year in, when the automation has quietly been dropping records.
Which One You Actually Need
A rough but reliable test. Answer honestly:
- Do off-the-shelf products exist that do roughly what you need, and you simply have not connected them? You need tool connection. Do not buy a custom build.
- Do products exist, but you cannot decide between them or have failed at rollout before? You need implementation help, possibly a diagnosis first.
- Have you evaluated products and found that the way you work is genuinely not what they assume? Now custom is on the table — but verify this claim hard, because “we are different” is usually false and occasionally true.
- Do you not actually know where the time goes? Start with diagnosis. Buying anything before this is gambling.
If you cannot yet answer these, run the automation diagnostic first — it produces the ranked list a good firm would charge you for. Most small businesses that think they need custom software need the second or third option. We say that as a firm whose revenue comes from the fourth.
What Fair Pricing Looks Like
Published rates in this industry are mostly fiction, so here is how to reason about it instead of chasing a number.
For a worked example of what a genuinely specific engagement looks like in one vertical, see workflow automation for behavioral health. Diagnosis should be a fixed fee and should be small relative to what follows. If a firm wants a large open-ended discovery budget before committing to any deliverable, that is a warning sign. A competent firm can scope a small business in weeks, not months.
Tool connection is usually hourly or small fixed-price. If someone quotes five figures to connect two SaaS products, ask specifically what makes it hard — sometimes the answer is legitimate (error handling, reconciliation, volume), and sometimes it is padding.
Custom builds are where the numbers get serious, and where the range is genuinely wide because scope drives everything. The useful question is not “what is your rate” but “what is the smallest version of this that delivers value, and what does that cost?” A firm that cannot describe a smaller first phase is a firm that has not thought about your risk.
One structural point that matters more than rate: ask whether the firm makes money on software licenses they recommend. Not because that is disqualifying — certified implementation partners are often exactly what you need — but because you deserve to know whether the recommendation is advice or distribution.
Five Questions That Reveal What You Are Buying
- “What would make you tell us not to do this project?” Every honest firm has an answer, usually involving volume thresholds, internal ownership, or process maturity. A firm with no disqualifying conditions will take any project, including yours when it should not.
- “Who owns the code and the accounts when we are done?” The answer should be you, unambiguously, including repository access and the cloud accounts things run in. If anything meaningful lives in the vendor's account, you have hired a landlord, not a builder.
- “What happens if we stop working with you in six months?” Listen for whether the system keeps running and whether another firm could pick it up. Documentation and standard technology are the answer; proprietary frameworks are the red flag.
- “Show me something you built that failed or got scaled back.” Firms with real delivery history have these stories and will tell them. Firms without one are either new or editing.
- “Have you looked at the specific systems we run?” Ask before they quote. A number produced without reading your systems' integration documentation is a guess wearing a suit.
Red Flags in Proposals
- A single fixed price for a large scope with no phases. This transfers all risk to you while appearing to do the opposite — because when scope shifts, and it will, you are renegotiating from a weak position.
- Percentage-of-savings pricing without an agreed measurement method defined in advance. The argument about what was saved arrives later, guaranteed.
- No named people. If the proposal does not say who is building it, you may be buying a subcontracted team you will never meet.
- Hosting in the firm's cloud account “for convenience.” Convenient for whom.
- Timelines with no dependency on you. Real projects need your people for decisions and testing. A plan that assumes zero client time is a plan that has not been delivered before.
- Discomfort with a paid pilot. A firm confident in its work will take a small, paid, scoped first phase. Insisting on all-or-nothing is a signal.
What a Good Engagement Actually Looks Like
Fixed fee, short duration. You should end up holding a written map of your processes, where the time goes, what it costs, and a prioritized list — useful to you even if you never hire this firm again. If the diagnosis is only a sales document for phase two, it was not a diagnosis.
One workflow, in production, used by real people. Not a prototype and not a demo. The purpose is to test the firm and the assumption at the same time, at a cost you can absorb if it goes badly. Insist on this even if the firm is happy to build everything at once.
Now you have evidence rather than projections. Expand the parts that worked, and be willing to kill the parts that did not. The firms worth keeping will agree with that on the record.
Documentation, credentials in your accounts, a runbook, and a named internal owner. Support afterwards should be a choice you make, not a dependency you are stuck with. Agree on what handover means before phase 2, not after phase 3.
When You Should Not Hire Anyone Yet
Three situations where consulting money is wasted regardless of who you hire:
Your process is not settled. Automating a workflow that three people perform three different ways encodes the disagreement. Standardize first — that is internal work, and it is free.
Nobody internally will own it. Every automation needs someone who notices when it breaks and knows what it was for. Without that person, you are buying something that will quietly stop working and nobody will notice for a quarter.
You have not measured anything. If you cannot say roughly how many hours a week go to the process in question, you cannot evaluate a proposal about it, and you certainly cannot tell afterwards whether it worked.
Where WorkflowUnity Fits — Plainly
We are the fourth category: a US-based custom software firm. We build on AWS, we hand over code and accounts to you, and we take small paid first phases specifically so you can test us cheaply.
Whether you should build at all is a separate question with its own four gates. We are the wrong call if your problem is connecting two SaaS tools, if you need a certified partner for a specific platform, or if you have not yet standardized the process you want built. We say so on first conversations regularly, because a custom build on an unsettled process is how firms like ours generate unhappy clients.
If you have measured the problem, own the process, and have found that the products genuinely do not fit — that is the work we do, and the questions above are fair to ask us.
Frequently Asked Questions
What does a small business automation consultant do?
The title covers four distinct jobs: diagnosing where time and money leak in your processes, connecting existing SaaS tools together, implementing and rolling out a specific platform such as a CRM or ERP, and building custom software when no product fits. Most firms specialize in one or two of these while marketing as though they cover all four, so identifying which job you need is the first and most consequential decision.
How much should automation consulting cost for a small business?
It depends entirely on which of the four jobs you are buying, and published rates are unreliable. Diagnosis should be a modest fixed fee with a deliverable you keep. Tool connection is typically hourly or small fixed-price. Custom builds are substantially larger and scope-driven. Rather than comparing hourly rates, ask each firm for the smallest first phase that would deliver real value and compare those — it reveals how they think about your risk.
How do I know if I need custom software or just better tools?
Test the claim that you are different. Evaluate two or three products properly — not demos, actual trials with your data — and document specifically where they fail to match how you work. If the gaps are cosmetic or process preferences, buy the product and adapt. If the gaps are structural, meaning the product assumes a fundamentally different operating model than yours, custom becomes reasonable. Most businesses that assume they are exceptions are not.
Who should own the code and accounts after the project?
You should, without qualification — the source code, the repository, and the cloud accounts the system runs in. If anything essential lives in the consultant's account or depends on a proprietary framework only they can maintain, you have created a dependency rather than an asset. Settle this in writing before work begins, not at handover.
What are the warning signs of a bad automation proposal?
A single fixed price for a large scope with no phases; percentage-of-savings pricing with no agreed measurement method; no named individuals on the delivery team; hosting in the vendor's cloud account; timelines that require no time from your staff; and reluctance to do a small paid pilot. Any one of these is worth a direct question. Two or more together is worth walking away.
When should a small business not hire an automation consultant?
When the process is not standardized yet, because automation will encode the inconsistency. When no one internally will own the result, because unowned systems decay silently. And when you have not measured how much time the process actually consumes, because you will be unable to evaluate the proposal or verify the outcome. All three are fixable internally at no cost, and all three should be fixed first.