AI tools at work: how to choose the right ones for the way you work

You’ve almost certainly read a “50 best AI tools of the year” list. You may have tried five or six of them. And if you’re honest, you still use one — two at most. That isn’t a lack of discipline. It’s that the question you started with was the wrong one.

The right question isn’t “what’s the best AI tool?” but “which tool strengthens the way I naturally work, rather than fighting it?”. Two people in exactly the same role, with the same targets, don’t need the same toolkit — because they don’t get stuck in the same places.

Why “best tools” lists don’t work

A tool list sorts by feature: writing, images, automation, note-taking. That’s catalogue logic, not usage logic. It answers “what can this tool do?”, when your real problem is “what slows me down every single week?”.

The outcome is predictable. You adopt a tool because the demo was impressive, use it for three days, then slide back into old habits — because it solved a problem you didn’t actually have. The cost isn’t just the subscription. It’s the energy you spent learning an interface you then abandoned.

The criterion that actually matters: how you work

Research on strengths at work — Gallup’s work on natural talents in particular — keeps arriving at the same point: people improve much faster by building on what they already do well than by fixing their weak spots. The same logic applies to AI.

An AI tool can do two very different jobs. It can amplify a strength: you’re already quick at structuring an idea, and AI takes that from ten minutes to two. Or it can compensate for a weakness: you hate writing, so AI writes for you. Both are legitimate, but the gains are of a different order. Amplifying a strength builds a lasting advantage; compensating for a weakness brings relief.

So start with the uses that amplify, then add one or two that compensate. Not the other way round. That’s the difference between an AI setup that makes you better and one that simply makes you a little less slow.

Four broad ways of working

To make this usable, ways of working can be grouped into four broad families. Nobody belongs to just one: you’ll have a lead family and often a strong second. The point isn’t to put you in a box; it’s to know where your effort will pay off.

Execution — you turn ideas into results

You need to finish things. A day without something concrete to show for it leaves a bad taste. Your strength: you move while others are still deliberating. Your trap: you get on with a task before checking it was worth doing, and you’re quick to absorb other people’s work.

• What AI amplifies: automatic formatting of recurring deliverables (minutes, reports, meeting summaries), first drafts you edit rather than write from scratch, and automations that remove the copy-paste work between two tools.
• What works against you: open-ended brainstorming tools with nothing to show at the end. They feel like progress and produce nothing.
• The habit to build: always ask for a precise output format (“a five-row table”, “a 120-word email”). Without one, you’ll lose in reformatting what you gained in drafting.

Influence — you take ideas to other people

You’re comfortable persuading, presenting and bringing people on board. Your strength: you make an idea land with people who would never have read it in a document. Your trap: you produce a lot, and how solid the substance is depends on how much time is left after the delivery.

• What AI amplifies: adapting one message into several formats and tones, preparing for objections before a negotiation, and reworking a pitch for a particular audience.
• What works against you: mass content generation. It amplifies precisely your weak spot — saying a lot, checking less.
• The habit to build: use AI as a critic before you use it as a writer. “What are the three weakest points in this argument?” beats “write me this post”.

Relationship building — you build what holds teams together

You read people, you defuse tension, you make sure the group holds together. Your strength is invisible on a dashboard and decisive in reality. Your trap: your time gets broken into fragments, and your own work happens in the evening.

• What AI amplifies: preparing for difficult conversations (wording tough feedback, anticipating reactions), summarising what was said so nobody gets forgotten, and drafting individual follow-up messages at scale.
• What works against you: anything that automates the relationship itself — generated replies sent as they are, scheduled congratulations. You’d lose the very thing that makes you valuable.
• The habit to build: point AI at the preparation and the paper trail, never at the conversation itself.

Strategic thinking — you analyse, structure and anticipate

You want to understand before you act. Your strength: you see consequences three moves ahead and spot the flaw in an argument. Your trap: you keep analysing well past the point where a decision would have done.

• What AI amplifies: reading and summarising large volumes of documents, testing several scenarios against each other, and getting a quick lay of the land in an unfamiliar field before going to the primary sources.
• What works against you: tools that generate even more material to analyse. Access to information isn’t your problem.
• The habit to build: force a decision. “Give me two options, recommend one, and tell me the biggest risk in your recommendation.”

At a glance: where to start for each family

A four-step method for choosing a tool

This takes an hour, once. It saves months of subscriptions you never use.

1. List your recurring tasks, not your goals. For one week, note everything you do more than twice. Those are your candidates — you can’t automate an annual objective, but you can automate a weekly task.
2. Tag each task as “amplifies a strength” or “compensates for a weakness”. Start with the first column: the payback is faster and the habit sticks more naturally.
3. Choose one tool for the task at the top of the list. Just one. Give it three weeks of real use before you even look at a second.
4. Measure the time you actually saved, not how it felt. If you can’t say what you gained after three weeks, cancel it without guilt.

Five mistakes that come up again and again

• Stacking tools. Three tools used at 20% are worth less than one tool used at 80%.
• Choosing on hype rather than on the task. A popular tool solves somebody else’s problem.
• Handing over your judgement. AI produces a first draft; the call, the decision and the accountability stay with you.
• Ignoring confidentiality. Before pasting an internal document into a consumer tool, check what happens to it. The question applies whether you’re covered by the GDPR in the EU or the UK, Switzerland’s data protection act, or your own company’s policy — which often turns out to be the strictest of all.
• Never cancelling anything. Review your tools every three months: anything you haven’t opened in a month goes.

How to tell which family you belong to

Self-diagnosis has its limits: we’re poor judges of what comes to us easily, precisely because it seems obvious. Three reliable clues: what colleagues always come to you for, what you do instinctively when a day frees up, and the kind of task that wears you out unusually fast.

If you want something more structured, an assessment of your dominant talents gives you a firmer starting point than gut feeling — and, crucially, it translates straight into which AI uses to prioritise. The four families are covered in more depth in our guide to the 34 talents, and how far along you already are with AI is the subject of the A1 to C2 AI literacy scale.