Agentic data platform
Intelligence needs a foundation
A commerce-native semantic layer with every entity (orders, products, customers, inventory, marketing, costs) modeled for agents. Not a chatbot on dashboards. Real AI reasoning over structured commercial data.
Semantic layer
Trusted by commercial teams at leading brands
Semantic layer
Commerce-native schema that unifies orders, marketing, customers, and inventory into one model the agent can reason over.
Profit depth
Revenue to CM3. Real profit after every cost, discount, return, and fee. Not vanity ROAS.
Smart modeling
Returns, cost rules, joins, and tracking configured for real-world complexity, not spreadsheet hacks.
Data domains
Every domain of the business, in one model.
Orders, marketing, customers, inventory, sessions, creatives, stores, costs, and returns — modeled together, so the agent reasons across all of them at once.
Orders
Revenue, COGS, fulfillment, shipping, returns, discounts, profit at every level.
Marketing
Spend, impressions, clicks, campaigns, ad sets, ads, all connected to orders, not siloed.
Customers
New vs. returning, cohorts, order sequence, LTV predictions, acquisition source.
Inventory
Stock levels, incoming, size curves, warehouses, days of supply, sell-through forecasts.
Sessions
Visits, product views, impressions, landing pages, devices, down to product and country level.
Creatives
Images, video, headlines, CTAs, format, linked to performance, not just ad metrics.
Stores
Physical store sales and inventory unified with online data for true omnichannel visibility.
Costs
Transaction fees, shipping, customs, pick and pack, per country, per provider. Real P&L, not estimates.
Returns
Return rates, reasons, refund amounts, and restocking timelines tracked per product, country, and channel.
Behavioral events
One product. Every behavioral signal.
Capture views, favourites, recommendation impressions, add-to-carts, and outfits as they happen — joined to profit, inventory, and returns in the same model the agent reasons over. Install via GTM or the Shopify one-click app.
Configured for real-world complexity
Returns, cost rules, joins, and tracking. The details that separate a toy model from a production one. Every rule updates profit metrics in real time.
Return #4821
| Order placed | Day 0 | €89.00 | |
| Return estimated (33% risk) | Day 5 | −€9.80 | |
| Return requested | Day 12 | ||
| Return shipped | Day 15 | −€7.43 | |
| Return received | Day 19 | Inspected | |
| GP3 adjusted | Day 20 | −€15.33 |
Return handling
Every return event (shipping refund, restocking, inspection) adjusts GP3 automatically. No manual reconciliation.
CostsCurrencyPer orderPer itemPer shipment% of value
Transaction costsEUR ▾0000%
Shipping costsEUR ▾10000%
Country: AustriaEUR ▾7,43000%
DHLEUR ▾3,75000%
UPSEUR ▾7,5000%
DHL ExpressEUR ▾4,3000%
Add row
CostsCurrencyPer orderPer itemPer shipment% of value
Transaction costsEUR ▾0000%
Shipping costsEUR ▾10000%
Country: AustriaEUR ▾7,43000%
DHLEUR ▾3,75000%
UPSEUR ▾7,5000%
DHL ExpressEUR ▾4,3000%
Add row
Cost rules
Transaction fees, shipping, customs, pick and pack, per country, per provider. Your P&L reflects reality.
Automatic joins
Marketing, orders, products, and inventory join automatically across sources. No manual data stitching or SQL.
Own tracking
Dema's tracking captures visits, product views, and impressions at product and country level, independent of GA4.
MCP integration
Model Context Protocol lets agents read from and write to external tools natively, including Slack, sheets, and ad platforms.
Profit depth
Dema speaks P&L, not just ROAS.
Every profit metric computed per order line item in real time — and ML-powered predictions available as regular metrics right next to the actuals.
01
From gross sales to true profit after every cost and return
The full chain: Gross sales → Net sales → GP1 (after COGS) → GP2 (after shipping & payment) → GP3 (after returns) → NGP3 (after pick & pack and overhead). Every metric computed per order line item in real time.
Profit waterfall
Revenue to CM3 breakdown for the current period. Each row shows the contribution or cost deducted.
Revenue
€284,000
COGS
−€112,000
Gross Profit
€172,000
Ad Spend
−€52,300
Shipping & Returns
−€28,400
Fixed Costs
−€18,700
CM3
€72,600
02
Not just historical reporting
Live data and ML-powered predictions available as regular metrics in every dashboard and report. Order forecasts, inventory sell-through, and profit-based LTV all update daily and sit alongside actuals.
CM3 forecast
4-week forward forecast for Contribution Margin 3 based on current spend trajectory and seasonal patterns.
68k
W10
72k
W11
65k
W12
78k
W13
82k
W14
85k
W15
88k
W16
91k
W17
Actual
Forecast
Predicted CM3 (4w)
€91,200+12%
100+
Pre-built commerce metrics (profit, acquisition, retention, inventory) available from day one
Clarity and control
“Dema has been a game-changer for our brand. It's given us the clarity and control we needed to scale profitably, ensuring we make the right decisions across inventory, marketing, and product strategy.”
Integrations
Connects to your full commerce stack
Ad platforms, e-commerce, analytics, email, and more. Data flows in, the semantic layer normalizes it, and the agent reasons across all of it.
Plus warehouse, POS, and ERP integrations via CSV, sFTP, and APIs.
View all integrationsOne foundation
The foundation for everything Dema does
The agent, self-serve analytics, MMM, and incrementality testing all read from the same modeled data. No separate pipelines, no conflicting numbers.
AI Agent
Ask questions in plain language. The agent translates them into queries across the full data model, including profit, inventory, marketing, and customers in one answer.
Self-serve analytics
100+ metrics, 40+ dimensions, real-time dashboards. The same modeled data, available through a point-and-click explorer.
Profit-first measurement
Always-on MMM and geo-based incrementality tests that read profit, spend, and channel data directly from the model. Budget optimization grounded in real margins, calibrated by causal experiments.
Trusted by category leaders
Real teams, real numbers — from cutting acquisition costs to doubling profit to automating whole planning cycles.
“Since we started using Dema, we can optimize our budgets for the most profitable growth in real time.”
How Myrqvist Use Dema to Optimize for Profitable Growth
Sebastian Öhrn
Founder, Myrqvist
83% Contribution margin increase
Enterprise-ready
Built for production.
Where your data is processed, how it is protected, and what an agent is allowed to change are controls you set — not defaults you inherit.
EU & US inference
Choose where your data is processed. Model inference and data pipelines run on European or US infrastructure with full residency control.
Never used to train models
Your data is never used to train models. Processing stays in your chosen region and runs in isolated, encrypted sessions.
Always the latest models
New reasoning models roll out as they ship, so your agents keep improving automatically without migration work.
Permissions and approvals
Each agent has its own scope and integrations. Actions can require a named approval before anything changes.
What you can create
Real setups, linked to how they run.
Every example below is a real prompt from the library. Open the use case behind it to see the full run — what it checks, what comes back, and where it lands.
Questions
Frequently asked questions
A semantic layer defines what metrics mean, how entities relate, and what calculations are correct before anyone queries the data. The agent, dashboards, and reports all use the same definitions, so numbers always match.
GA4 tracks events and sessions. Shopify shows gross margin. Neither connects marketing spend to inventory to profit after all costs. Dema unifies the full picture and adds forecasting and an AI agent on top.
Behavioral events capture every product view, favourite, recommendation impression, add-to-cart, and outfit creation, joined directly to profit and inventory inside Dema's semantic layer. Install via GTM or the Shopify one-click app and events flow into the agent and dashboards immediately.
The full hierarchy: Gross Profit 1 (after COGS), GP2 (after shipping & payment fees), GP3 / NGP3 (after returns), and contribution margin. All computed per order line item in real time.
Connect your e-commerce platform and ad accounts, configure your cost structure, and you're live. Most teams are running within a day.
Yes. Dema layers on top of your stack. It ingests data from Shopify, Google, Meta, TikTok, Klaviyo, and more. You don't need to replace anything.
Bring one commercial job