How to Measure Pipeline Coverage —and See Why It Fails
- Bill Kantor
- 2 hours ago
- 7 min read
I recently posted about how pipeline coverage is such an unreliable indicator of future sales. I commented that if you were to measure how predictive pipeline coverage is, that you would be disappointed. And I encouraged sales leaders not to trust me. Just measure your actual coverage ratio yourself.
Some people have asked: “How do you do this?” It’s a great question because there are a lot of ways to interpret what I said. Here’s what I meant.
Most businesses use pipeline coverage to understand if they have enough pipeline to meet their goal. What they mean is that at some point in a forecast or planning period, the ratio of pipeline value to the goal is high enough to suggest they can meet or beat it.
Now I really advise against doing this to predict what will happen because there are far better ways to do that. (We have blogged extensively on why coverage is such a bad approach.) But, if you want to do this exercise to see how bad coverage is, here’s how I advise you measure your actual coverage ratio.
Actual Coverage Ratio = Open Pipeline for a period / Actual Sales in that period
That formula is only useful if you define “Open Pipeline” and "Actual Sales" consistently.
Definitions
We need to define Open Pipeline. There are two dimensions to consider:
How advanced in your sales process does a deal need to be to be included?
How should the estimated close date affect whether a deal is included?
The guiding principle is to include only deals that had a realistic chance of closing in the planning period—then compare that pipeline to the deals actually won during the same period.
Assume you are planning for the next 90 days.
Should you include MQL opportunities? Probably not. Most will not close within the quarter. Unless you have a very short sales cycle.
You may need a stage threshold. For example, Sales Qualified may be appropriate. In that case, only Sales Qualified opportunities or later—excluding Closed Won opportunities—would count toward open pipeline.
Close dates require judgment too. Should you include opportunities whose close date was a year ago? Probably not. Thirty days ago? Perhaps. What about opportunities with a close date more than 90 days in the future?
That flexibility is part of the problem. Small changes in the definition can produce very different coverage ratios.
For a 90-day planning period, we suggest defining Open Pipeline on each date as opportunities that:
Were at the Sales Qualified stage or later on that date
Had not previously been won
Had a close date between 14 days overdue and 90 days ahead
The 14-day allowance keeps common close-date overruns in scope for up to two weeks. If the close date is not updated by then, the opportunity drops out of Open Pipeline. The 90-day window matches the planning horizon.
Now you have a clean definition of the Open Pipeline. You have an amount field which you can total to get the Open Pipeline value. Your CRM records the history of these key fields (stage, close date, and amount) so you can reconstruct this Open Pipeline value for any date in the past.
Next you want to define and compute Actual Sales in the planning period. This is the denominator of the equation above. Note that this will include deals that were not in the Open Pipeline. These may be previously lost deals that were rejuvenated, close date overrun deals that got updated during the planning interval, future close date deals that got pulled in to close earlier than anticipated, MQL excluded deals, and even deals that were not in the CRM at all at the start of the period. That’s the way coverage is used. You don’t care where the deals you will close come from.
Actual Coverage Ratio on each date is computed as follows:
The Open Pipeline value as of 90 days before a date, divided by the total value of won deals in the 90 days after the Open Pipeline measurement date.
So this gives you a 90-day sliding-window Actual Coverage Ratio on every day. Create this ratio say every week for the past and you can see the trend in and distribution of your historic 90-day Actual Coverage ratio. Note that this has nothing to do with the prior goals. We are simply showing the Actual Coverage ratio achieved on each date.
Examples
We’ve done this analysis countless times and consistently see similar conclusions. Actual coverage ratios vary widely over time within a business. Here’s one example:
In this example, the actual coverage ratio was relatively “stable” for the last six quarters—yet it still ranged from 2.5x to 5.0x; a 2:1 spread. But in the three quarters prior to that, it ranged between 2.1 and 7.5 (a 3.6:1 spread). Within a company, we typically see peak-to-trough coverage variation of roughly 2:1 to 3:1. That’s a big range and this is one reason (of many) why coverage ratios are so bad at predicting future sales. Admittedly, there are periods where for some businesses, coverage is more stable than shown here. But for every one that we have seen, that pattern falls apart at some point.
Most businesses don’t use coverage ratios to predict as much as to understand if they have enough coverage to make a goal. That’s a little more reasonable way to use coverage ratios. The question then becomes what coverage ratio do you need to be relatively certain that you will meet your goal?
It’s tempting to use the historical distribution of actual coverage ratios to estimate how much coverage a business would need today to meet its goal with a given level of confidence.
For instance, In this example, a coverage-to-goal ratio of 5.0x corresponds to the 90th percentile. So, if the historical relationship held, the business would need roughly 5.0x coverage to goal ratio to have a 90% chance of meeting its goal.
Unfortunately, a 5.0x threshold is usually so high that the analysis becomes unhelpful.
And the larger problem remains: coverage ratios are often so loosely related to future sales that even a historically derived threshold may not be reliable enough to manage to.
Another twist on this is that you may not want a sliding window ratio. You may only care about the ratio at the start of your quarter. You can do that analysis. But you only get one measurement per quarter. You can see that in the charts. Grey bars are the start of quarters. In the example here, we have 10 quarters. Ratios are: 4.8, 2.7, 7.3, 4.8, 4.2, 3.8, 4.2, 4.1, 4.0, 3.1. Can you make sense of that? There are fewer observations than the sliding window, and a lot of it is old. It’s hard to make decisions with so few data points—particularly given the observed inherent variability at other times.
You also may rightfully argue that this is not the way you use coverage. You only care about the deals you will close before the end of the quarter. As you march up to that date, the planning period gets reduced every day. We’ve done that analysis too. Here’s what that looks like (a different company). Each line represents the actual coverage ratio measured throughout the quarter, with a fixed end point of the end of each quarter.
When calculated this way, coverage ratios are similarly volatile (ranging from 5 to 15 here) quarter-to-quarter for the first half of the seven quarters shown; and wildly volatile for the second half of the quarters. They become meaningless at the end of the quarters. This is one business, but the pattern is typical of what we see: high quarter-to-quarter variability early in the quarter, followed by increasingly distorted ratios as the end date approaches.
What coverage can—and cannot—tell you
I encourage you to run this analysis on your own data. Not because I expect pipeline coverage to become useful. Because the quickest way to stop debating it is to measure it.
You may find periods when coverage appears stable. You may even find a historical ratio that looks reassuring.
But do not confuse your average historical coverage ratio with a reliable forecast. If you were to manage your business to the median coverage ratio in Figure 1 above, you would—assuming coverage were actually predictive—fail to meet your goal half the time!
The question is not whether you can find a ratio that looked reasonable in the past. It's whether that ratio reliably tells you what you will sell next.
In most businesses, it does not.
Coverage ratios vary too much over time, become increasingly distorted as the quarter progresses, and are actively gamed as soon as you try to manage to a ratio.
Coverage does not quantify your probability of hitting the number.
It is not a measure of risk.
It does not tell you what your team should work on.
At best, it is a distraction—a rough tally that creates the appearance of insight without reliably improving a sales decision.
At worst, it is harmful and misleading. Coverage becomes a management target—encouraging teams to add deals at unrealistically high values and retain questionable deals because removing them makes the dashboard look worse.
That is not productive sales management. It is forecast theater.
The goal of measuring coverage ratios is not to find a figure that makes the plan feel safe. That’s not possible.
A far better approach is to understand your odds of reaching the sales goal—and identify the actions that can improve them.
That requires separate models for open pipeline and prospective new pipeline, a distribution of possible outcomes, and a way to prioritize the deals and segments most likely to change the result.*
Measure coverage if you need to.
Then stop managing to it.
* I can recommend a good vendor.
