There is a version of the AI consulting pitch you have probably heard by now. It goes something like this: your business needs to be “AI-ready,” which requires a thorough audit, a strategic roadmap, a phased implementation plan, and a twelve-month engagement to see it through. It sounds responsible. It sounds thorough. It is, in most cases, a way of billing you for work that shouldn’t take twelve months.
I want to be direct about this, because I think it matters for anyone trying to figure out where AI actually fits in their business.
The transformation story is built for consultants, not for you
“AI transformation” is a compelling framing if you are a large consulting firm with a team to staff and a quarterly revenue target to hit. For a small business owner, it is usually the wrong unit of work entirely.
Transformation implies that your whole operation needs to change before AI can be useful. It implies complexity, sequencing, and long lead times. It gives a consultant room to spend month three on a strategy document and month seven on a “pilot program” that hasn’t shipped anything yet.
The small businesses I work with don’t need transformation. They need one or two specific things built well. A proposal process that used to take four hours and now takes forty-five minutes. A client intake system that no longer falls through the cracks. A way to get a first draft of something in ten minutes instead of ninety. These are not transformations. They are targeted improvements to specific, already-painful problems. And they can be built, tested, and working within sixty days.
If someone is quoting you a twelve-month timeline before they’ve identified a single concrete workflow to improve, that timeline is doing more work for their business model than it is for yours.
What actually takes sixty days, and what genuinely takes longer
I want to be fair here, because not every long engagement is padding. Some AI work is genuinely complex: integrating AI into legacy systems, building something that handles sensitive data with real compliance requirements, training a model on proprietary information at scale. If that describes your situation, longer timelines are honest.
But that does not describe most small businesses. Most small businesses have a handful of repetitive, time-consuming tasks that AI tools can touch right now, with no custom infrastructure required.
The pattern I see across the founders I work with is this: the first useful thing we build together is almost never the most sophisticated thing. It is usually the most obvious one. The task that everyone on the team does slightly differently. The thing that should have a template but doesn’t. The question that lands in the owner’s inbox because there’s no good system to catch it first.
You can build something genuinely useful around a problem like that in sixty days. In week one or two, you are identifying the exact workflow and what’s actually slow about it. By week four or five, you have something running that real people are using. By day sixty, you know whether it worked and what to improve. That’s a complete loop. It’s not a phase. It’s a result.
If a project can’t show you something working in sixty days, one of two things is true: the scope is genuinely large and complex, in which case you should be told that clearly upfront, or the project isn’t ready and the honest thing to do is say so.
The question worth asking before you sign anything
Before you engage anyone to help you with AI, including me, ask them this: what will I have that’s working at the sixty-day mark?
Not “what will we have mapped out” or “what will we have planned.” What will be running. What will your team be using. What problem will have gotten measurably smaller.
If the answer is vague, or if they tell you sixty days is too soon to expect results, treat that as information. It doesn’t mean the work is bad. It means the engagement hasn’t been structured around your outcomes.
Good AI work for a small business starts with a specific problem, builds something to address it, and proves it out quickly enough that you can decide whether to go further. The consulting model that works against that instinct, that slows down the feedback loop and stretches out the timeline, is one that serves the consultant’s margin more than it serves your business.
You are allowed to expect something useful, and soon. That’s not impatience. That’s good judgment.
