Ask your store a question.
“What sold best last week? Why did revenue dip on Tuesday?” Plain language in, plain language out — with the working shown and the action one approval away.
Today, without it
Ten dashboards, no answers.
Analytics tells you what happened. The why — and the what-now — is still homework.
Ask in plain language
The dashboard answers back.
Lumnya reads your live store data and answers the question you actually asked — no query language, no chart-building, no waiting for someone to run it.
>refund order #1043 and email the customer an apology
… reading order #1043 · drafting refund + apology email
preview
Refund €42.00 (full) — order #1043
Email: “We’re sorry about the delay — your refund is on its way.”
Refunded & email sent · logged 14:32 · approved by you
It shows its work
An answer you can check.
Revenue Detective decomposes a move into quantified drivers — volume, basket size, refunds, mix — so “sales dropped” becomes a list of reasons with numbers next to them. Every specialist attaches its reasoning to what it reports.
>how are sales this week?
Revenue
€12.4k
+8%
Orders
214
+5%
AOV
€58
+2%
Up 8% on last week, driven by the hoodie restock. One watch-out: Oxygen Pro margin dipped after Tuesday’s discount code.
From answer to action
Then tell it to fix the thing.
Prices down to the right variant, refunds, discount codes, customer emails, settings, triggers and FAQ — you ask, Lumnya shows you the exact change first, and nothing happens until you approve it.
multi-model engine
How it runs
Four steps, no project plan.
Ask
In plain language, the way you'd ask a colleague who knows the store.
Read the reasoning
The answer arrives with its drivers and its working attached.
Preview the change
If it proposes an action, you see exactly what would change.
Approve — or don't
One tap. Approved actions execute and land in the audit trail.
The honest edges
What it won't do
An honest read on the analysis you get.
Why it matters
Not a chart. An answer, and then a fix.
The gap between knowing and doing is where most stores lose the week.
Straight answers
Questions merchants ask about this one.
No. It reads live store data to answer, and it can act on what it finds — price edits down to the variant, refunds, discount codes, emails, settings — with every consequential action previewed and confirm-gated.
It tells you what it doesn't have rather than filling the gap with something plausible. Low confidence surfaces as low confidence.
Your store's own isolated data. Nothing is pooled across merchants, and your data isn't used to train foundation models.
The other four
Something else eating your day?
Private beta
Hand over this job first.
Join the private beta and be first in line when Lumnya opens to your store.