Connect the systems and data providers you already use. Lumos reads every source and works out what a customer is, what counts as revenue and how your systems connect. Ask in plain English and get the number, the definition it used and how it was calculated. No data pipeline to build, no SQL to write, no BI tool to maintain.
How it works
Billing, CRM, accounting, your product database, spreadsheets and exports. Read-only, and we set it up with you.
Whatever shape the data comes in, each source is read as it is, down to every field. Nobody writes a data model first.
The same customer in Stripe, HubSpot and your product is recognised as one customer. Revenue, subscriptions and deals are named and linked the way your business runs.
Most AI data tools guess what a field means each time you ask. Lumos works it out once, from your data, and keeps it.
Does an active customer mean paid in the last 90 days, or signed in during the last 30? The data supports both, and they give different numbers. Lumos shows your team both, once. The answer becomes the company's definition.
Other tools pick one and don't tell you.
This is your business as Lumos understands it. Now ask it something.
Ask
Anyone on the team can ask. Each answer shows the definition it used and how it was calculated, so you can check a figure before it goes in the board pack.
A chat tool can write a query that runs cleanly and still counts the wrong thing. Here the definition sits on the answer, where you can see it.
| Customer | Plan | Q3 2025 | Q3 2026 |
|---|---|---|---|
| Northwind Labs | Enterprise | $48,000 | $61,200 |
| Oakridge Supply | Enterprise | $36,000 | $42,800 |
| Halcyon Foods | Growth | $14,400 | $17,100 |
| Brightline | Growth | $12,400 | $14,100 |
| Fieldnote | Starter | $3,600 | $3,240 |
Example company, illustrative figures.
Missing data
Other tools answer anyway, or stop.
None of your connected tools records why a customer cancelled. To answer this, Lumos proposes:
Example company, illustrative figures.
What you get
Between your data and an answer there is usually a data engineer, an analyst and a BI tool. Lumos does all three, so your team goes straight from what the business wants to know to the number.
Lumos loads every source you connect and builds the tables that answer for your customers, revenue and the rest. Data you do not have yet is planned, approved by your team, and built.
Anyone on the team asks. Lumos writes and runs the query, and every answer shows its definition and how it was calculated. No SQL, and no waiting in an analyst's queue.
Pin an answer and it becomes a dashboard tile that refreshes as your data does. No BI tool to buy, set up or maintain.
Retention, margin and revenue, each traced to its definition and how it was calculated. When finance and sales count customers differently, the gap is a definition you settle once.
Why does finance count 1,284 customers and sales count 1,537?
Ask about the business directly, across every tool you connected, without building a dashboard first.
Which accounts grew the most since they signed?
Questions nobody built a dashboard for are answered without a ticket. After an acquisition or a migration, Lumos works out what each new source holds and matches the same customers across systems.
Which of the acquired company's customers already buy from us?
Lumos only reads your sources. It never writes back to them.
People only see the data their role allows, in answers and in dashboards.
Open any figure to see its sources, the definition used and how it was calculated.
Each definition and link shows who made it: Lumos, or the person who approved it.
We are onboarding a small number of companies and set up each one with them. Leave your work email to request a place.