For sales & solutions engineering
Build the demo data.
Not the deal-breaker.
Sales engineers, solutions engineers, solutions architects, and forward deployed engineers all end up doing the same unpaid job: building the data behind a demo, from scratch, per deal, usually the night before. Misata is the part of that job you can hand to a declaration instead of a script.
What you actually need
A demo dataset shaped like the prospect's own business, not a generic template.
Numbers that land where the demo needs them: a churn rate, a revenue curve, an SLA breach, not whatever a script happens to produce.
Every table to hold up if the prospect clicks into it, because a buyer evaluating a data product will run their own query.
To turn this around in hours, on a Thursday night, without opening an engineering ticket.
A real workflow
“Series B logistics company, 40 warehouses, EU and US, the champion cares about carrier cost overruns.” Plain English or a schema you already have, either works.
On-time delivery at 87%, breaching in Q3, worst in the Nordics. State it directly rather than shaping rows by hand until they look right.
Misata solves the rows to match every declaration exactly, then checks the result: every foreign key resolves, every roll-up reconciles, the aggregate lands where you said.
“Make the Nordics worse. Add a carrier late 30% of the time.” The rest of the declarations hold; only what you asked for changes.
What you get
A relational dataset in the format your demo environment already reads (CSV, JSON, SQL, Parquet, or a live database), plus a verification you can show your own manager before it goes in front of a prospect: every foreign key resolves, every declared number checked against what the rows actually contain.
Why this holds up technically
Misata is a declarative engine, not a row-by-row generator: it solves the whole schema together against your constraints, so a revenue curve, a group share, and a foreign key all land correctly at once instead of fighting each other. It’s open source and MIT licensed, so you can read exactly how a declared aggregate gets solved rather than taking a vendor’s word for it, which matters when you’re the one demoing it to a buyer who will ask.
Questions
How can sales engineers create realistic demo data?
By declaring the schema and the outcome the demo needs to show, rather than building it by hand. Misata solves the relational data to match the declaration exactly, verifies every foreign key resolves, and hands back a dataset a prospect can click into without it falling apart.
Can I customize the data for a specific prospect?
Yes, that's the primary case. Describe the prospect's vertical and the story the demo needs to tell, revenue growth, a churn spike, an SLA breach in a specific region, and Misata builds a schema and dataset shaped around that, not a generic template.
Do I need engineering help to do this?
No. It runs as a Python library, an MCP tool inside Claude or Cursor, or the no-code Studio in your browser. A sales engineer can go from a plain-English brief to a verified dataset without a ticket to another team.
Read the full case, or just start
The deeper argument for why demo data is the hard part of demo engineering, sourced and evidenced, lives on its own page, alongside the original research behind it. Or skip straight to building something.