AI strategy01
Before you buy another AI tool, map one day of work.
A practical way to find a useful AI starting point: follow one task, identify the friction, and define a result worth measuring.
The journal / Leadership / Note 05
Leadership · Field note 05A Toronto business can start smaller than an AI department.
For a Toronto owner with a full calendar, “develop an AI strategy” can sound like another large project. A useful first question can be much smaller: which piece of work deserves a closer look?
Begin with the person doing the task
Choose a recurring task that someone knows well. Ask what makes it slow or frustrating, and what a good result looks like. Give that person a meaningful role in deciding whether an experiment is worth running.
This is a Toronto-focused business note, not a report on local adoption rates. The point is relevant to an owner, an independent consultant, or a small team: a first experiment should be close enough to the work that someone can evaluate it honestly.
Use an ordinary example
Consider a hypothetical local service business that writes an internal summary after each project meeting. The team already has notes and an agreed template. A limited trial could explore turning approved, non-sensitive notes into a draft for the project lead to check.
The first question is whether the notes are clear enough. If decisions, suggestions, and unresolved issues are mixed together, begin by improving the note-taking template. If the source is sound, test whether assistance reduces the total effort while preserving the distinctions the team needs.
Keep the boundaries visible
Choose the allowed information, the expected output, and the person responsible for approval. Make clear that the experiment does not authorize the system to contact customers, promise dates, or make decisions on behalf of the business.
Keep the existing workflow available. If a trial creates more checking than it saves, or requires information the tool is not approved to receive, pause and reconsider the scope. A small experiment is easier to learn from when the organization can step back without disrupting its work.
Choose a question you can answer
Instead of asking whether AI will transform the business, ask whether this particular task becomes more useful or less burdensome. Track preparation, drafting, checking, and correction. Ask the person receiving the output whether it is clearer and more reliable.
The answer might be yes, with a narrow scope. It might be that a shared template solved most of the problem. Either result gives the owner something more useful than a broad statement about being ready for the future.
Let the next step earn its place
If the trial works, document the conditions under which it worked. Keep the approved brief, the review routine, and the ownership clear. Only then consider whether a nearby task has enough in common to justify another experiment.
My interest in Toronto’s business conversations is the connection between ambition and everyday work. You can take the opportunity seriously while keeping the first step manageable. Start with a problem worth understanding, then see what the evidence supports.
One thing to take away
One recurring task, one responsible owner, and one clear decision can be a sensible beginning. A company-wide rollout can wait.
Pasqualino Mazza
Entrepreneur, global AI consultant, and content creator. Writing about AI, business, and leadership from a Toronto perspective.