Revenue Intelligence vs data warehouse

A data warehouse is a central store where data from all your systems comes together, so you can query it in one place. It is infrastructure: it carries no revenue logic, no checks and no follow-up. Revenue Intelligence is an application that brings exactly that logic: it connects CRM, contracts, orders and invoices, looks for where they disagree and puts an amount on each difference. The data warehouse is the storeroom. Revenue Intelligence is the person who counts the stock and notices what is missing.

What is a data warehouse?

A data warehouse, such as Snowflake, Google BigQuery, Microsoft Fabric or a comparable environment, is a database built to store large volumes of data from different sources and query it quickly. The work around it has three parts:

1. Loading. Data from CRM, ERP, billing, marketing and other systems is copied into the warehouse on a schedule. That takes separate tools or scripts.
2. Modelling. The raw data is turned into tables that fit together: one customer table, one order table, one invoice table, with keys linking them. This is the heaviest part of the work.
3. Using. BI tools, analysts and other applications query the warehouse.

The advantage is that all the data sits in one place, in a form you can combine. The drawback is that the warehouse only does what somebody builds into it. It is an empty room with very good shelving.

What is Revenue Intelligence?

Revenue Intelligence starts from a purpose: checking revenue. It knows which data that takes, how the data has to be linked and which differences matter. It puts deals next to orders and invoices, contracts next to billed prices, usage next to subscriptions. Discrepancies become findings with an amount, evidence and a proposed action. The full explanation is in What is Revenue Intelligence?.

What is the difference between a data warehouse and Revenue Intelligence?

Where the warehouse stops

Suppose you have a good warehouse. CRM, ERP and billing are loaded every night, and there is a customer table that links the systems. What do you still not have?

The comparison. Someone has to decide which deals belong to which orders, on which field, and what happens when they cannot be matched unambiguously. That is logic that has to be built into the warehouse.

The agreements. Indexation clauses, volume tiers, minimum commitments and rates for additional work sit in contracts. A warehouse loads tables, not PDFs. Unless someone records those agreements as data, they are not in the warehouse.

The definition of an error. Is a deal with no invoice after 20 days a problem? After 45? Is a difference of EUR 12 rounding or a mistake? Is a credit note a correction or a leak? Every check needs choices that have to be recorded somewhere.

The follow-up. A table of discrepancies in the warehouse does nothing. Someone has to read it, investigate and resolve. There has to be an owner and a way to record what happened to each discrepancy.

All of that can be built. It is simply a project of months, followed by ongoing maintenance. The general structure of such an environment is covered in Revenue data architecture for B2B.

Do you need a data warehouse for Revenue Intelligence?

Not necessarily. Revenue Intelligence can get its data in two ways:

• Directly from the source systems. Through connections to the CRM, the ERP and billing. That is the usual route for companies without a warehouse.
• From the warehouse. If a company already has a warehouse holding the relevant data, Revenue Intelligence can read from it instead of from each system separately.

For a company with EUR 2M to EUR 20M in revenue and no data team, building a warehouse only to check revenue is usually a detour. A warehouse makes sense when there are far more questions than revenue control, and when there are people to run it. Which connections matter for Revenue Intelligence is set out in Which systems need to be connected for Revenue Intelligence?.

When a data warehouse is the right choice

• You have many different sources and many different questions, not only about revenue.
• You have a data team or a regular partner who maintains pipelines and models.
• You want to keep historical data for a long time, including when source systems are replaced.
• You use BI tools that need to run on one central source.
• You have your own analyses and models that demand more than standard reporting.

In that situation a warehouse is a sound foundation, and Revenue Intelligence can sit on top of it. The two do not compete: the warehouse supplies the data, Revenue Intelligence supplies the revenue control.

Worked example: build or use

Worked example: suppose a wholesaler with EUR 18M in revenue is considering building revenue control in its own data warehouse. A rough estimate of the work:

• Loading CRM, ERP and billing and building a customer and order model: several weeks of a data engineer's time.
• Building checks for deal without order, order without invoice, price differing from contract and indexation not applied: several more weeks, including agreeing with finance on what counts as an error.
• Recording contract terms as data: depending on the number of contracts, days to weeks of manual work, plus a process to keep up with new contracts.
• Building a report or alert and agreeing on follow-up.
• After that, ongoing maintenance: every change in a source system can break a pipeline or a check.

What that comes to in euros depends on rates and on whether a warehouse already exists. The point is the shape: a one-off project plus a permanent maintenance burden. Against those costs stands what the control returns. The commonly cited estimate for revenue leakage is 1 to 5 percent of revenue; at EUR 18M that would be EUR 180,000 to EUR 900,000. Whether this company falls within that range, you only know after measuring. How to weigh this up is covered in How do you calculate the ROI of Revenue Intelligence?.

Checklist: what is in your warehouse, and what does it do?

If you already have a data warehouse, work through these questions:

1. Are CRM, ERP and billing all three in it, and are they updated daily?
2. Is there a key that links a deal unambiguously to an order and an invoice?
3. Are contract terms with a price impact held in the warehouse as data?
4. Have checks been built that show discrepancies between the systems?
5. Who reads those discrepancies, how often, and what happens next?
6. What happens to the checks when a source system changes a field?

If you have no answer to questions 4 and 5, you have a well-stocked storeroom where nobody counts the stock.

The wider comparison with BI, CRM, ERP, dashboards and other software is in Revenue Intelligence vs Business Intelligence.

Frequently asked questions

Is a data warehouse the same as BI?

No. The warehouse stores the data. BI tools present that data in reports and dashboards. They often work together, but they are different layers.

Can Revenue Intelligence run on Snowflake?

Revenue Intelligence can read data from a warehouse if the relevant tables are there. Whether a specific product can do so depends on the connections it supports.

We have no data team. Do we need a warehouse?

Not for revenue control. Revenue Intelligence can connect directly to your CRM, ERP and billing. A warehouse without people to maintain it goes out of date quickly.

What is the difference between a data warehouse and a data lake?

A warehouse stores structured, modelled data, ready to query. A data lake stores raw data in all sorts of forms, including unstructured data. For revenue control the distinction matters less than whether someone builds the checks and follows them up.