CRM forecast vs actual revenue

The forecast in your CRM and the actual revenue in your accounts measure two different things. The CRM predicts which deals will be won and at what value, at the moment of closing. The accounts record what was invoiced in a period. The gap between them has five parts: deals that do not close, deals that close later, value that drops at signing, revenue spread over time and revenue that leaks away after closing. Measure those five separately and you can correct your forecast and find your revenue leakage at the same time.

Why can the CRM figure and the finance figure both be right?

In the management meeting, sales says the quarter produced EUR 1.4 million. Finance reports EUR 1.1 million in revenue. Both figures can be correct. Sales counts the value of deals marked as won during the quarter. Finance counts the invoices sent during the quarter.

The problem arises when the sales forecast is used for decisions that depend on the finance figure: cash flow, workforce planning, expectations towards the bank. Then the difference is no longer a matter of definition, but a planning error. The article why CRM data is not the same as financial data explains the background. This article is about how large the gap is and where it comes from.

What makes up the gap between CRM forecast and revenue?

1. Deals that do not close

The forecast counted on a deal that was eventually lost, or that simply went quiet. This is the part everyone knows. It grows as stage probabilities become more optimistic and stalled deals stay in the pipeline longer. The article why are sales forecasts so often wrong covers this at length.

2. Deals that close later

The deal is won, but one or two quarters later than expected. For the annual forecast that matters less, for the quarterly forecast and cash flow it matters a lot. Measure how often and how far close dates have slipped in the past, and use that as a correction.

3. Value that drops at signing

The deal was in the CRM at EUR 80,000, the signed value is EUR 68,000. An extra discount in the last round of negotiation, a smaller scope, an option the customer dropped. If the CRM is not updated after signing, the old value stays and the forecast keeps using it.

4. Revenue spread over time

A three-year contract worth EUR 120,000 is a won deal of EUR 120,000 in the CRM. In the accounts it is EUR 3,333 a month. A EUR 200,000 project won in March is invoiced in instalments from April to November. This part of the gap is not an error but a difference in timing. It does have to be built into the forecast, otherwise a good sales year looks as if it misses on revenue.

5. Revenue that leaks after closing

The deal was won at the right value, but not everything is invoiced. No order was created. The first billing month was skipped. An agreed expansion never reached invoicing. The indexation is missing. This is the part of the gap that costs real money, and it is the part that is measured least, because sales has done its job and finance has invoiced everything it was given.

How do you measure the gap?

To measure the gap, you put won deals from the CRM next to invoices from the accounting system, per deal or per customer. One way to do it:

1. Choose a period far enough back. Deals won twelve to eighteen months ago have had their first invoices. More recent deals give a distorted picture because of part 4.
2. Link each won deal to the invoices it produced. This needs a key: a deal number on the order, or a customer number with a date window. The article CRM-to-billing reconciliation explained describes how to set that up.
3. Calculate the expected invoicing to date for each deal. Based on the signed value, term and start date.
4. Compare with actual invoicing. The difference per deal is a combination of parts 3, 4 and 5.
5. Put the forecast from the time next to it. Which deals were in the forecast but were not won, or were won later? Those are parts 1 and 2.
6. Split the total gap across the five parts. The result is a table showing where the difference between your CRM forecast and your revenue comes from.

That table is one of the most useful documents you can produce for your forecast. It tells you which correction to apply from now on and which errors to fix instead of correcting for them.

Should you correct for the gap or fix it?

The five parts call for different approaches.

Do not correct for part 5 in your forecast. If you lower the forecast because you know revenue leaks after closing, you are accepting the leak. Fix it, and the forecast becomes more accurate by itself.

Worked example

Worked example: suppose a business services company forecast EUR 3.0 million of new revenue from new deals last year, based on the CRM. Revenue actually invoiced from new deals was EUR 2.2 million. An analysis of the EUR 800,000 gap:

Parts 1 and 2 call for better pipeline weighting. Part 3 calls for discipline in the CRM. Part 4 is not a problem, as long as the forecast accounts for it. Part 5, EUR 60,000, is leakage: work that was sold and never paid for. Without this breakdown, everything would have ended up under "sales was too optimistic", and the EUR 60,000 would never have been found.

Forecasting at the moment of revenue

The structural fix for part 4 is to build your forecast not at the moment of closing, but at the moment of invoicing. That means every deal in the forecast gets an expected billing schedule: start date, instalments or monthly amounts, term. For deals with a standard structure this can be automated. For projects it needs an estimate of the project schedule.

A forecast built on the moment of revenue ties in directly with the accounts. The discussion in the management meeting is then no longer about two numbers that do not agree, but about one number with an explanation. How to build that step by step is described in how do you build a reliable revenue forecast.

Checklist

• Do you know how large the gap between CRM forecast and invoiced revenue was last year?
• Can you split that gap across the five parts?
• Is the deal value in the CRM updated to the signed value after signing?
• Does every won deal have a start date and a billing schedule?
• Is there a check that flags won deals without an order or invoice?
• Does the management meeting report one revenue expectation with an explanation, or two numbers?

The wider context of forecasting, from contracted revenue to ranges, is covered in what is revenue forecasting.

Frequently asked questions

Should the CRM forecast equal revenue?

No. They measure different things. The aim is that you can explain and predict the difference, not that it is zero.

Which part of the gap is usually the largest?

That varies by company. In companies with optimistic pipelines, parts 1 and 2 are often the largest. In companies with multi-year contracts or long projects, part 4. Part 5 is usually smaller, but it is the only part that costs real money.

How often should I do this analysis?

Once thoroughly, to set the corrections. After that, an update every quarter, so you can see whether the corrections still hold and whether the leak in part 5 is getting smaller.

Who should do this, sales or finance?

Finance, with input from sales. Finance has the invoicing data and the neutrality. Sales has to supply the explanations for deals that were not closed or closed later.