Scope
Do they propose one workflow first, or a big plan?
For a small business, the wrong AI partner costs more than no AI partner. Look for agencies that start narrow, show real deliveries, and talk about data and operations - not just models. That is how we work at Aprex.
Last updated: September 18, 2026
An agency proposing a three-month analysis before anything is built sells hours. For a small business, the first step should be one workflow with clear scope: what should improve, which data is needed, and how it is measured. Our 2-6 week pilot model is made for exactly this.
Anonymous claims about delivery counts say little. Ask to talk to customers and see products in operation. Aprex has built TapInn, in daily use, plus the 24AI and 24Markets labs - all viewable yourself.
Ask how the agency handles your data: where it is stored, who sees it, what happens on error. Vague answers here are a red flag no matter how impressive the demo is. Privacy and approval must be clarified before building, not after.
We are a small team in Tonsberg that designs, builds, and operates ourselves. You talk to the person who builds. We start with one workflow, agree clear scope, and give go/no-go after the pilot - continue, change scope, or stop.
Ask these questions of any AI agency - including us.
Do they propose one workflow first, or a big plan?
Can you talk to past customers and see deliveries in operation?
How are storage, access, and errors handled concretely?
Do you get an assessment of your case, or a generic package price?
Who operates after launch, and what do changes cost?
See what a good 2-6 week AI pilot should include.
Read more →We do not claim to be best for everyone. Large enterprises with their own AI teams rarely need us. Our niche is smaller companies that want one workflow solved properly - with people who answer when you call.
Send one sentence about the workflow that takes most time. We answer honestly whether a pilot makes sense.
Contact Aprex →