Apps
Planners, review queues, dashboards that act — describe the workflow and get a working application on live commercial data. Share it with the team, adjust it in chat, and push approved changes to your connected systems.
126 applications built by commercial teams
Build a sell-through planner where the team reviews reorders and pushes approved changes to the ERP.
Shared with Planning
6 teammates · can edit
“Add a markdown column”
Adjusted in chat · 40s
2 reorders to the ERP
Named approval required
Actions execute in connected systemsTrusted by commercial teams at leading brands
Sound familiar?
Commerce workflows are too specific for generic software, and an AI chat can't reach your data — so they live in spreadsheets, pasted exports, and read-only dashboards.
planner_FINAL_v7 (2).xlsx| A | B | C | |
|---|---|---|---|
| 1 | Rust Tee | 412 | #REF! |
| 2 | Linen Shirt | 388 | 1.9 |
| 3 | Rib Knit | — | 6.1 |
The team’s real tool is a spreadsheet
Build a buying planner for the merch team
Here’s your buying planner! I’ve built it around the file you uploaded — just re-upload each week to refresh it.
✕ It runs on a pasted file, not your systems
✕ Wrong the day the data moves on
Elin 09:14
Rust Tee is at 14.8 weeks of cover — can someone update the markdown in the ERP?
Screenshot 09.12.44.png
PNG
How it works in Dema
An app is a described workflow on the modeled data — with sharing, chat-based adjustments, and actions built in. Every example below works that way, and links to its full use case.
Meta Creative Gallery
Facebook & Instagram creatives · ad-platform attribution · Online
Spend
€2,921,284
Blended ROAS
9.08×
Creatives shown
99
Live (total)
218
Needs attention
41
Scoring weights · ranks creatives within each objective
35% of score
0% of score
47% of score
18% of score
Objective
Market
Sort by
Search
Sales / Purchase 80
99 of 99
25 adsCream boxy tee — UGC
30 adsField overshirt — studio
35 adsOat crewneck — lifestyle
31 adsLinen polo — carousel
Pause 3 fatigued creative sets in MetaOne tool, every team
“Finance, marketing, operations, and management can collaborate around the same insights, even if they ask different questions.”
Sandra Lindberg
CFO, Rapunzel of Sweden
Why it works
Anything can produce an interface once. The difference is whether it runs on data that stays right, and whether the team can share and change it without breaking it.
Sell-through planner
Live · synced just nowA tool built in an AI chat runs on the file you pasted, and dies when the data moves on. Apps built here sit on the same live modeled data as everything else — adjust them, share them, and they’re still right next quarter.
#tradingDema APP
Weekly trading report — 3 movements explained.
CM3 = Revenue − COGS − fulfilment − marketing
Defined by your team — not inferred
Orders, inventory, returns, marketing and omnichannel data connect once and stay maintained. Nobody on your team owns the pipeline.
And dozens more via APIs and direct integrations.
View all integrationsWhat you can build
Every card is a real build prompt from the library. Open the use case behind it to see the working application — and how teams adjust it once it's live.
One foundation, three ways to work
Chat, Agents, and Apps all run on the same modeled commercial data — the answer, the workflow, and the application never disagree.
Which products drove margin last week?
Answer with 3 charts
12s
Ask why margin moved, which products slowed down, or where paid growth stopped being profitable — and get the number that is actually right.
Explore Chat
Anomaly scan completed
06:00
Report delivered to Slack
06:02
Budget change prepared
06:04
Meta DE
+€9.8k/wk
Status
Awaiting approval
Describe the report, monitor, or routine once. An agent runs it on schedule, delivers the result, and executes approved changes.
Explore Agents
Enterprise-ready
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.
Choose where your data is processed. Model inference and data pipelines run on European or US infrastructure with full residency control.
Your data is never used to train models. Processing stays in your chosen region and runs in isolated, encrypted sessions.
New reasoning models roll out as they ship, so your agents keep improving automatically without migration work.
Each agent has its own scope and integrations. Actions can require a named approval before anything changes.
Questions
Substantial applications for planning, merchandising, buying, creative review, forecasting, and commercial operations: planners with suggested actions, review queues with approvals, leaderboards, controlling workspaces, and other tools generic software does not fit.
No. Apps are described in plain language and adjusted the same way. There is no code to own, no deployment to manage, and no ticket queue between the team and the tool.
The same modeled commercial data as everything else in Dema: orders, costs, returns, inventory, marketing, and customer behavior, plus Dema's measurement suite. Apps never run on a stale upload.
Yes, within the permissions you set. A planner can push approved reorders to the ERP, a merchandising app can publish rankings to the storefront, a budget queue can execute in ad platforms — each action waiting for a named approval.
Apps are shared like documents, with roles deciding who can view, edit, or approve. Anyone with access can refine it in chat — add a column, change a threshold, reshape a view — without breaking it for others.
A dashboard shows numbers. An app carries the workflow: the suggested decision, the review step, and the action into the connected system. It is the difference between seeing a markdown is overdue and shipping it.
Bring one commercial job
