Enterprise AI
What to Prepare Before You Hire Someone to Automate with AI
Most AI automation projects stall on what the business has not written down yet, not on the technology. Half an hour of preparation saves weeks of back-and-forth.
1. The task, written as it is done today
One task, not a department. Write what triggers it, the steps in order, and what it produces. If you cannot write the steps, the task is not ready to be automated: it first needs someone to decide how it should be done.
2. Ten real examples, including the awkward ones
Collect ten real cases: ten incoming requests, ten client replies, ten quotes. Include the ones that went wrong or needed a manager. Anyone building the automation will test it on these; without them, it gets tested on invented cases that are always easier than yours.
3. Where the data lives and who opens the door
List the tools the task touches (mailbox, CRM, spreadsheet, accounting software) and who can grant access to each. Access requests are the most common reason a small project waits for weeks.
4. Where a human must approve
Decide in advance which outputs a person checks before they leave the company: anything that sends money, a price, or a promise to a client. Writing this down early keeps the first version safe and makes the team more willing to use it.
5. How you will know it works
Pick one check you can do by hand: compare the automation's output with what your team would have done, on the ten examples, then for one or two weeks of real work. If you cannot say what a good result looks like, nobody can build toward it.
If you want this prepared for you
Venkai's free AI Scan (venkai.fr/en/scan/) reads your website and returns three concrete automation leads in a few minutes. The X-Ray goes further: 3 to 5 sized opportunities, each with today's workflow, the automated workflow, the leverage, the complexity and the first action. It costs €290 excl. VAT, credited to your pilot if it starts within 30 days. Both are produced by an AI system (Venkai, running on Anthropic's Claude).
FAQ
Do I need clean data before starting?
No. You need to know where the data for one task lives and who can give access to it. Cleaning everything first is a common way to never start.
Who on my team should be involved?
The person who does the task today. They know the exceptions, and the exceptions are what decide whether the automation works.