Paralysed by where to start with AI? One diagnostic question cuts through the noise. Here's how to find your starting point in under five minutes.
Every sales leader has heard the same pitch by now. AI will close more deals, train your reps, predict your pipeline, and write every email. The tools multiply every month. The vendor decks are relentless. And most heads of sales end up in the same place: knowing AI matters, but not knowing where to actually begin.
The paralysis is understandable. There are too many options, real consequences for choosing wrong, and no shortage of people selling certainty they haven’t earned. So instead of adding another framework to the pile, this piece answers the question plainly, using the AI Maturity Scale to show you where you sit today, and one diagnostic question that tells you where to go first.
Where you sit right now: the AI Maturity Scale
The AI Maturity Scale runs from [0] to [100] across four bands. It’s a useful benchmark because it stops the conversation from jumping straight to tools.
Laggard (0 to 25). AI isn’t part of the business yet. The team is aware of it but hasn’t made a move.
Curious (26 to 50). You’re watching, reading, maybe experimenting with a free tool. Nothing has been built for the business specifically.
Explorer (51 to 75). You’ve stood something up and you’re learning from it. A pilot is running, or one workflow has changed.
Leader (76 to 100). AI is running real work at scale. The team has shifted how it operates around it.
Most sales leaders who come to us sit in the Curious band. They’ve seen the demos. They believe it works. They just haven’t committed to a first move, because the first move feels like a guess.
It doesn’t have to be.
The one diagnostic question
The source of the paralysis is usually the same thing: people start with the tool, not the problem. They ask “which AI should we use?” before they’ve asked “what are we actually trying to fix?”
The question that cuts through it, drawn from talking to all of our customers, is this: what’s one thing that drives your bottom line, that AI has already demonstrated it can do well, and where you know your business well enough to execute it properly?
That question has three parts, and all three matter. The bottom-line condition stops you chasing novelty. The “already demonstrated” condition keeps you out of experimental territory where the risk is real. And the domain knowledge condition is what separates a build that works from one that looks good in a demo and fails in practice.
For most sales teams (and other teams), the answer lands in the same place.
What that looks like for a sales team
Picture a mid-sized sales team with 12 reps. Each rep spends roughly 10 hours a week on admin, writing up call notes, logging activity in the CRM, and drafting follow-up emails after meetings. That’s time not spent on prospects, not spent on deals, and not spent on the calls that actually move pipeline.
The work is predictable. A call happens. Notes need to be written. A follow-up needs to go out within 24 hours or the deal cools. The pattern almost never changes, which is exactly the kind of work AI has already proven it can handle.
So they started there. Calls were transcribed and summarised automatically into the CRM. Follow-up drafts landed in each rep’s queue for review and approval before anything went out. Nobody’s judgement was replaced. The AI handled the grunt work, and the rep decided what to send.
Within six weeks, each rep had 9 hours a week back. Follow-up time dropped from days to minutes. The deals that used to go quiet because everyone got busy stopped going quiet.
That’s not a dramatic claim. It’s a narrow, well-defined problem that AI is genuinely good at, pointed at a team that knew their own process well enough to set it up properly.
What AI won’t do here
It won’t close deals. It won’t replace the rep on a complex negotiation or a relationship that needs a person in the room. A difficult objection still needs a human response, and the AI knows to flag those rather than draft past them. The CRM notes it writes are a first draft, not a final record. A rep still reads them before they’re locked. The follow-up emails go out with a person’s name on them, and a person approved every word.
That’s the model. AI clears the predictable work so the rep’s time goes where it actually counts.
What this means for you
If your reps are still typing notes at 9pm, or if deals are going quiet because follow-ups fell through the gaps, you already know the problem. The diagnostic question just confirms where to point the solution. You don’t need a bigger team, a magic tool, or a six-month implementation project to start. You need one well-defined problem, a process you understand, and a first move that’s low enough risk to learn from quickly.
The AI Strategy work we do with sales teams starts exactly here: mapping the work, finding the highest-value first move, and building something that fits the way your team actually operates, not a generic template. If you want to see how that applies to your specific situation, you can also read more about AI agents for sales teams on our website.
If you want to know where your team sits on the AI Maturity Scale and what your first move should be, take the AI Roadmap Interview. You’ll talk through your goals, your pipeline pain points, and where the admin is costing you the most time. You’ll leave with a concrete starting plan, not a brochure.
