Use cases
Real questions teams ask, mapped to Tattva — the data you need, how to ask, what comes back, and how to keep watching it.
This section answers the question every new user has: "What can I actually use this for?"
Each use case is a concrete recipe: the business question, who asks it, the data you need, how to phrase it in Reasoning, what the answer looks like, and how to keep watching it with a tracker.
Use cases are grouped by team. The section covers Performance marketing.
Planned
These use cases are built on NRoAS, CAC, marketing % and targets, which aren't defined in the Library yet. Each page says which definitions it needs.
How to read a use case
Every page follows the same shape:
- The question: the business question, the way it gets asked.
- Who asks it: the people, and the meeting where it comes up.
- Frequency: how often it comes up.
- Data you need: the tables and columns the answer depends on.
- How to ask it: questions to type, with example phrasings.
- What you'll get back: the shape of a good answer.
- How to make it recurring: a tracker recipe with cadence, rule and delivery, or a report you save each cycle.
- Pitfalls: common mistakes and how to avoid them.
The figures and names on these pages are made up. Your answers will use your own data and your own metric definitions.
Before you start
The recipes work best once your data analyst has defined the metrics they use (NRoAS, marketing %, repeat share) as stored definitions. Figures from a definition are computed by code and shown in the Trail's Compute card; figures without one are the agent's own SQL and labelled as such. See For data analysts.
Where to start
Pick the use case closest to a problem you have now, and come back to the concepts when you need them.