Get startedGet started for free

Department-Wide Adoption

1. Department-Wide Adoption

Welcome back! Your Sales pilot is working, so now it's time to scale it up to the whole department.

2. The pilot worked, now scale it

In the previous chapter, you set up a Genie space on the payments table and made it answer reliably for a small Sales team. It worked. Now the Sales Director wants to roll it out across the whole department, and the Marketing team has noticed and wants in too. That means more people asking, and bigger questions than one table can answer.

3. One table isn't enough

Their questions go beyond payments. The Director wants revenue by booking outcome: confirmed, completed, or canceled. Marketing wants revenue by country, with business customers broken out. None of that lives in the payments table alone. To answer it, your space needs more tables, and it needs to know how those tables connect.

4. Connecting tables with relationships

So your first step is to add the bookings table, then tell Genie how it relates to payments. That connection is a relationship: a join on booking_id, set to many-to-one, because many payments map to one booking. Once that link exists, Genie can combine the two tables on its own. And the rules you set in the previous chapter, like excluding failed payments, still apply.

5. Watch the second money column

Adding bookings brings a trap. It has its own money column, total_amount, which is the quoted booking price, not the money actually received. With two money columns now in play, you'll add a clear description so Genie never mistakes the booking price for real revenue. It's insurance, the same idea as the synonyms you added before.

6. A three-table chain

Next, you'll add the users table for Marketing's segmentation, joined to bookings the same way. Now you have a three-table chain: payments, to bookings, to users. With entity matching on the segment columns, Genie can resolve real values like businesses and specific countries on its own and answer questions that no single table could.

7. Locking a shared definition

With more people in the space, one problem grows: net revenue starts meaning different things to different people. You want one definition, computed the same way every time. A plain text instruction is too loose for a multi-step formula like this. So you'll add a SQL example: a titled, prebuilt query that Genie reuses exactly whenever someone asks the matching question. One formula, locked for everyone.

8. One asset, every country

That locks one formula. But Marketing won't ask about net revenue just once; they'll ask for it country by country. So your last step turns that SQL example into a Trusted Asset: a saved, reusable query with a country placeholder that Genie fills in for each question. One asset answers every country. Just remember to use the word "country" when you test it, so that Genie picks this asset instead of the plain one.

9. Let's practice!

By the end, your space spans three tables and locks the formulas that the whole department relies on. Time to roll it out. Let's build it!

Create Your Free Account

or

By continuing, you accept our Terms of Use, our Privacy Policy and that your data is stored in the USA.