Dema answers from your orders, stock, returns and ad spend, using the definitions your team already agreed on.
−14% net sales in Germany, mostly from three styles that went out of stock early in the week:
| Style | Out of stock | Lost sales | Back in stock |
|---|---|---|---|
 Heavyweight Hoodie, Grey | Tue 6 Oct | €18,400 | 21 Oct |
 Cable Knit Crewneck | Wed 7 Oct | €9,200 | 19 Oct |
 Leather Sneakers, White | Thu 8 Oct | €6,100 | 26 Oct |
Used by the teams at
Go beyond the top line
Revenue rarely tells the whole story. Ask about contribution after product costs, fulfilment, returns and advertising, with the detail behind the number.
Follow up by market, product, channel or customer group. Each question builds on the same commercial model.
Meet the data underneathPart of the platform
Marketing mix modeling, incrementality testing and causal attribution are built into Dema. Bring that evidence into your questions, budget reviews and recurring workflows, alongside the rest of your commerce data.
The foundation of a useful answer
Shared metrics, company context and reusable skills behind every answer.
Define once. Use across every answer and workflow.
After costs. After advertising.
Company rules, with room for personal preferences.
Instructions, code and references. One reusable command.
weekly-margin-reviewRun this for Sweden.
Share the work. Set who can change it.
Start with a real question
Explore questions teams ask about margin, demand and marketing performance.
Why did contribution margin fall last week, and which products and markets drove it?
Answer with 3 charts
Compare reported ROAS with modeled incremental return for every channel last month.
Reported vs incremental, every channel side by side
Which products will break their stock cover in the next 4 weeks at the current run rate?
Ranked breaks with the revenue at risk
Which products convert well but never get seen? Rank them by the revenue we would gain moving them up the page.
Buried converters ranked by revenue upside
Break yesterday's revenue by market, then channel, then product category, keeping the same definitions at every level.
Three-level drilldown, same definitions throughout
Which styles justify air freight this week? Compare the margin cost of flying against the full-price sales lost waiting for sea.
Fly-or-wait decisions with both costs modeled
How did return costs and payment fees move against plan this quarter?
Both cost lines against plan, quarter view
Which regions are growing fastest organically where we are not yet scaling paid?
Fastest-growing regions with no paid scale yet
From the very first prompt
“Dema worked perfectly right out of the box. It's the first tool that truly understands our data from the very first prompt.”
Robin Mostacero
CEO, aim'n
Orders, inventory, returns, marketing and omnichannel data connect once and stay maintained. Nobody on your team owns the pipeline.
Built on the same foundation
Give your team the capabilities to move faster, with the controls to make them part of everyday operations.
Inference and pipelines in the region you choose.
GDPR compliant.
Scoped permissions. Approval before the actions you choose.
New reasoning models. No migration for your team.
Questions
A generic assistant reasons over whatever file you pasted, as of whenever it was exported, with definitions re-explained each time. Dema answers from live modeled commercial data with your metrics defined once, and flags missing data instead of guessing.
Dema uses shared definitions for orders, costs, returns, inventory, marketing and customer behavior. People working with the same period, filters and data use the same calculation, rather than redefining the metric in each conversation.
Anything across the commercial picture: why margin moved, which products are slowing down, where paid growth stopped being profitable, how return costs track against plan, with drill-downs by market, channel, product, and customer cohort.
Missing sources and broken definitions get flagged instead of quietly published. The agent tells you what it would need, rather than inventing a confident number.
Yes. Any investigation can be scheduled as a recurring agent or turned into a shared application. The same data and definitions carry over, so nothing gets rebuilt.
No. You ask in plain language and the semantic layer handles the joins, definitions, and correctness underneath. Analysts can still go deeper when they want to.
