Revenue Intelligence vs Forecasting Software

Forecasting software predicts how much revenue will come in over the coming months, based on the pipeline, historical revenue and contracts. Revenue Intelligence also looks back and sideways: is the revenue that has already been sold and agreed actually coming in, at the right price? A forecast tells you where you are heading. Revenue Intelligence tells you whether the starting point is right. A forecast built on data with leaks simply predicts those leaks along with everything else.

What does forecasting software do?

Forecasting software produces an expectation of future revenue. Depending on the company and the package, it uses different sources:

• The pipeline. Open deals in the CRM, with an amount, an expected close date and a probability of winning per stage.
• Historical revenue. Seasonal patterns, growth, average revenue per customer.
• Recurring revenue. Running contracts and subscriptions expected to continue.
• Sales judgement. Commit, best case, pipeline: the estimates of salespeople and managers.

Modern forecasting tools, sometimes part of the CRM and sometimes standalone, use statistical models or AI to weigh those sources. They do not show a single number but a range with a probability. A full explanation is in What is revenue forecasting?.

What does Revenue Intelligence do?

Revenue Intelligence puts the systems in the revenue chain side by side and looks for differences. It does not ask "how much is coming?" but "is what exists right?". Has every won deal been invoiced? Does the price match the contract? Has indexation been applied? Is a customer buying less than their contract or history would suggest?

Revenue Intelligence can also include a forecast, but that forecast is then based on data that has been checked first. That is the fundamental difference.

The difference

Why does a forecast predict the leaks too?

A forecast based on historical revenue assumes the past is a good basis. If 3 percent of revenue went uninvoiced in each of the past two years because of forgotten additional work and missed indexations, that leak is in the history. The model learns that this is normal revenue and predicts the coming period with the same leak built in.

A forecast based on the pipeline has a different issue. The pipeline predicts what sales will win. But won is not invoiced. If some of the won deals are never invoiced, or not in full, the forecast is structurally too high. Why this happens so often is explained in Why are sales forecasts so often wrong?.

In both cases the forecasting software does what it is supposed to do. The problem lies in the data that goes into it.

Three ways leaks distort a forecast

Too high because of won but uninvoiced deals. The forecast counts won deals as revenue. Some are never invoiced. Every month the gap between forecast and reality is explained as "timing", while it is structural.

Too low because of silent growth. Customers use more than they pay for. A forecast based on invoicing does not see that growth. Real demand is higher than the model thinks, but revenue does not follow.

The wrong base because of missing indexation. The model uses current prices as its base. If those prices are too low because of missed indexations, every expectation is too low. And the model has no way of seeing that.

How do they work together?

The logical order is: check first, then predict.

1. Revenue Intelligence finds the leaks between agreement and invoice, and the differences between CRM and accounts.
2. The leaks are closed, or at least quantified.
3. The forecast is built on data whose contents you know.
4. Revenue Intelligence measures afterwards where the forecast deviated from reality, and whether that was down to the model or to a new leak.

If you have both in one system, you can show a forecast with and without the open findings. That makes visible how much revenue comes in if everything stays as it is, and how much if the leaks are closed. How to build a forecast you can trust is covered in How do you build a reliable revenue forecast?.

Worked example: a forecast that misses every quarter

Worked example: suppose a project business with EUR 9M in revenue forecasts each quarter based on won deals and the planning schedule. The forecast for Q3 is EUR 2.4M. Actual invoicing is EUR 2.25M. The EUR 150,000 difference is explained as delay.

On closer investigation:

• EUR 50,000 is genuine delay: projects that will be delivered in Q4.
• EUR 60,000 is additional work that the forecast counted as revenue but that was never created as an invoice line.
• EUR 40,000 is deals marked as won for which no project was ever created in the ERP.

If that pattern returns every quarter, the forecast structurally misses EUR 100,000 a quarter that has nothing to do with timing. A better forecasting model does not solve that. Closing the leaks does. The figures are made up to show the mechanism.

What you can expect from forecasting software

This is not an argument against forecasting software. A good forecast is essential for planning: staffing, purchasing, cash flow, investment. What you can expect from it:

• A range, not a single number. A forecast that says EUR 2.4M without a margin suggests a certainty that does not exist. See Forecast confidence explained.
• Insight into how it is built up. Which part comes from running contracts, which part from the pipeline, which part is assumption?
• Feedback. How large was the difference between forecast and reality in previous periods, and in which direction?

What you cannot expect is for it to check whether the data it calculates on is complete and correct. That is not its job. A forecasting package that sees a structural deviation can at most adjust its model, so that next quarter's miss looks smaller. The leak itself remains; only the prediction adapts to it. That is exactly why the question "why do we keep missing?" should not only be put to the forecasting model, but also to the chain from deal to invoice.

Checklist: is the basis of your forecast right?

• Has the difference between forecast and actual invoicing gone the same way for the last four quarters? Then it is probably structural, not chance.
• Is that difference split into timing and genuine misses?
• Has the historical revenue your model uses been corrected for known leaks?
• Do you count won deals as revenue, or only invoiced ones?
• Do you know whether your prices are current, including indexations?
• Are forecast and invoicing compared per customer or per deal, or only in total?

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

Frequently asked questions

Does Revenue Intelligence include a forecast?

It can. The difference is that the forecast then works with data in which the differences between systems are already visible. But the core of Revenue Intelligence is checking, not predicting.

Will AI make my forecast better?

AI can weigh patterns better and give a more realistic range. It does not fix the fact that the underlying data contains a leak. A clever model on skewed data predicts precisely the skew.

Why is our forecast always too high?

A common cause is that won deals are counted as revenue while some of them are never invoiced in full. Compare forecast and invoicing per deal to see whether that applies to you.

What should I do first: improve the forecast or find the leaks?

Find the leaks. A forecast builds on the data that exists. As long as that data contains leaks, a better model mainly improves the precision with which you predict the wrong thing.