Where do we start with AI?
Start with a problem you know. Make one useful change. Measure what gets better.
Many business owners can see the potential of AI. The harder question is where to begin. Which part of the business needs it? How much work will it take? Will it make a difference?
We run businesses too. Our starting point is a conversation about the work: what takes too long, what keeps getting stuck, and what drains the team’s energy.
Look across the business.
We walk through the organisation together: operations, finance, sales, customer service. Where does information come from? Who needs it next? What has to be checked, copied or chased along the way?
A finance team may spend hours gathering information that already exists elsewhere. An operations manager may answer the same question several times a day. Each frustration tells us something about how the business works.
This gives us a shared picture of the organisation and helps us choose where to focus.
Count the repetition.
A task that takes ten minutes can feel too small to matter. If it happens thirty times a week, it adds up to five hours.
Frequency changes the calculation. So does the effort needed to improve the task. A recurring problem with a straightforward solution can be a useful place to start. Another problem may be worth tackling later because it needs more preparation or changes across several teams.
We look at these questions together: how often does this happen, what does it cost in time and effort, and how much work would it take to improve?
Start with one useful change.
We choose a manageable problem that matters to the team. Something they recognise, use regularly and can help us improve.
We build, try it in their working day and adjust. Their feedback shows us what still takes effort and what genuinely becomes easier. That first project gives everyone a practical basis for deciding what to do next.
Make the result visible.
Before the new workflow goes live, we agree on what improvement should look like and how to track it.
If we estimate that it saves ten minutes each time, a simple dashboard can count completed uses and show the estimated time saved. We check that estimate against actual use, including any review or corrections still needed.
Time is one measure. Depending on the problem, we might track fewer errors, shorter turnaround times or less work waiting for someone’s attention. When a benefit is harder to quantify, we still agree on what evidence would show that things have improved.
The first project should leave you with something useful, a result you can assess, and a clearer idea of where to go next.