Every week brings a new AI product with a bold promise. You cannot try them all, and you should not. Here is the filter I use with clients.
Start with the job, not the tool
Write the job in one sentence: "Reply to order-status questions in Arabic and English within five minutes." Now you are comparing tools against a job, not against each other's marketing.
Prefer what you already pay for
Your email, documents and chat tools increasingly include AI features. If one of them can do 80 percent of the job, use it. A new subscription must earn its place.
Check four things
- Data: where does your information go, and is it used to train the vendor's models?
- Cost at scale: the price for ten requests a day is not the price for a thousand.
- Exit: can you export your work if you leave?
- Language: does it handle Arabic and mixed Arabic-English text well? Test it with your own messages.
Run a small bake-off
Take ten real examples, run them through two options and compare the results side by side. It takes an afternoon and it ends most debates.
Be wary of these signals
"Fully autonomous", "replaces your team", prices that only appear after a demo call, and no way to see what the tool actually did. Good tools show their work.
Remember the tool is the cheap part
The expensive part is the process around it: who checks the output, what happens when it is wrong, and who owns it. Spend your time there.
