The AI analyst for your store.
Every answer shows its work.
Ask questions about your store in plain language. Every number in the answer carries a receipt: the governed metric definition, the exact SQL, and a hash you can replay without the AI in the loop.
Request early accessAsk in plain language
"What was total revenue?" "Compare June and July." "Which countries drive revenue?" The model picks the right governed metric and writes the prose. It never writes the SQL and never touches your data directly.
Every answer carries a receipt
Answers come in two visible sections: receipted facts (numbers produced by governed SQL, each with a receipt id and hash) and interpretation (model prose, clearly labeled as not receipted). Any receipt can be replayed model-free; a doctored one fails.
Your warehouse is yours
Tilltrend builds a real bronze/silver/gold warehouse from your store data: on PostgreSQL you control, exportable, inspectable. Built so that leaving is easy, which is exactly why you won't have to.
What an answer actually looks like
This is not a mockup. Everything below comes from one real run of the product answering "What was total revenue?", executed 2026-07-26 against the demo warehouse. Reproduce it yourself with:
python app\scripts\ask.py "What was total revenue?"In the app, the answer opens as a readable card built from the stored receipt. Two renderings of one stored value: the card formats it as money at the display layer, the receipt keeps the exact number.
- Gross revenue only: the source has no cancellation/refund data.
Total revenue across all time was 29,356,250. This figure represents gross revenue only; the underlying source lacks cancellation and refund data, so the net revenue after adjustments may differ from this amount.
The exact audited text stays one click away in the app, unchanged to the character. Here it is in full: the verbatim terminal output of that same run.
question: What was total revenue? turn: turn_7333a781e9ba mode: model model: claude-haiku-4-5-20251001 RECEIPTED FACTS =============== (1) [run_metric] total_revenue (all time) <rcp_1e047258878c43cba3b21d673a7ccbc5 | sha256:2bb5a5c95744d84e> total_revenue = 29,356,250 caveat: Gross revenue only: the source has no cancellation/refund data. INTERPRETATION (model, not receipted) ===================================== | Total revenue across all time was 29,356,250. This figure represents gross revenue only; the underlying source lacks cancellation and refund data, so the net revenue after adjustments may differ from this amount. receipts: rcp_1e047258878c43cba3b21d673a7ccbc5 | sha256:2bb5a5c95744d84e replay: python app\scripts\replay.py rcp_1e047258878c43cba3b21d673a7ccbc5 postcheck: PASS (attempts=1)
Every number traces to one query. Here is the governed SQL Tilltrend runs
for total_revenue: the exact statement the deterministic
layer executes for that metric and stores in the receipt's execution
block. The model only chose the metric; it never writes SQL. Tilltrend's Ask page now
shows this beneath every answer:
SELECT SUM(f.sales_amount) AS total_revenue
FROM gold.fact_sales fThe 29,356,250 is a property of the demo dataset, not a revenue claim about this product or anyone's business. Replaying that receipt re-executes the same governed SQL with no model involved and grades the result. Observed verdict: MATCH. Doctor the receipt file and it fails. See every claim with its reproduction command →
Where this is honest by construction
Tilltrend is pre-launch. No merchant counts, no testimonials, no revenue-lift percentages: we have none, so none appear on this site. What we do have is a working product whose every claim is reproducible: the evidence page lists each one with the exact command and the observed result, and the changelog records only shipped, verified work.
Early access
We're looking for a small number of merchants to shape the product. The Shopify app is real and running: install is by private link today, and the public App Store listing waits on our legal pack. If "an analyst that shows its work" sounds like what your store is missing, write to us. A human reads every mail.