Mind the (Pipeline) Gap: Using Cohort Analysis to Forecast Bookings and Identify Risks

A bottom-up, deal-by-deal approach to forecasting serves SaaS companies well in the immediate term, but effective financial planning requires longer-term forecasting and by extension, real math. Cohort analysis is a powerful and straightforward technique for predicting the future behaviors of customers and prospects.

Consumer businesses regularly leverage cohort analysis frameworks for both analyzing and forecasting customer retention, yet somehow the practice is far less discussed in the B2B world. Understanding pipeline cohort behavior is not only mission-critical for forecasting bookings, but also for understanding gaps in pipeline coverage and identifying risks in your go-to-market plan.

A refresher on cohort analysis

Before we dive into the nuts and bolts of pipeline cohort analysis, it’s important to first understand what cohort analysis is in the first place: It is a method for examining how a population behaves over time, ultimately so that you can make an apples-to-apples comparison of behaviors at certain points along the way.

A very pragmatic way to understand cohort analysis is to think about a drug trial. If Person A started a drug trial for Condition X on September 1, 2024 and Person B started the same trial for the very same condition on January 1, 2025, we cannot equitably compare the wellbeing of those two individuals on February 15, 2025, as Person A has had significantly longer exposure to the drug. However, we can compare each of their reactions and behaviors in the first 45 days of the trial.?

Cohort analysis can be used for a host of different populations—everything from people to pipeline opportunities—and can be cut across myriad time dimensions (weeks, months, quarters, etc.).

From pharmaceutical trials to SaaS pipeline math

In the example below, we look at the number of qualified sales opportunities created in each quarter and what portion of those opportunities close in the same quarter, +1 quarter later, +2 quarters later, etc.

As more quarters pass, we can begin to build an average in-quarter win rate, a +1 quarter out win rate, and so forth. In-quarter win rate is calculated as follows:

• Q1’2023: 2/55 opportunities close in-quarter = 3.64%?
• Q2’2023: 5/50 opportunities close in-quarter = 10.0%?
• Q3’2023: 4/60 opportunities close in-quarter = 6.67%?
• Q4’2023: 2/45 opportunities close in-quarter = 4.44%
• Average in-quarter win rate = 6.19% or (2+5+4+2)/(55+50+60+45)?

The same math applies for +1 quarter out win rate, +2 quarters out, and so forth, ultimately yielding the following cohorted win rates:

The sum of the cohorted win rates totals to 25% (and is probably a number that’s talked about frequently), but focusing on total win rate without considering this “intake” curve will mask important nuances about exactly when deals will close.

While the aforementioned example considered cohort behavior by opportunity count, it is equally helpful to examine cohort behavior by dollars piped.

When we run the very same in-quarter, +1 quarter out, etc. calculations for the dollars closed view (vs. the opps closed view), the win rate intake not-so-coincidentally lands us in the very same place:

Leveraging cohort behavior to forecast bookings

Now that we understand what our closing dynamics look like over time, we can leverage our win rate intake curve to produce a bookings forecast.

Let’s assume that we’re nearing the end of this quarter (Q1’2025) and have built the following pipeline over the past year:

• Q1’25 - $6MM
• Q4’24 - $5MM
• Q3’24 - $4MM
• Q2’24 - $3MM

We can multiply these historical pipeline numbers by our win rate intake breakdown to forecast our Q1 bookings. Q1 forecast is thereby the sum of:

• Q1’25: $6MM * 6.19% in-quarter win rate = $371K
• Q4’24: $5MM * 7.59% +1 quarter win rate = $379K
• Q3’24: $4MM * 6.70% +2 quarter win rate = $268K
• Q2’24: $3MM * 4.73% +3 quarter win rate = $142K
• Q1’25 expected bookings forecast = $1.16MM

Using cohort forecasts to identify pipeline gaps

Now let’s assume that we are at the start of Q2’25, and we need to unpack 1) the feasibility of achieving a $2MM bookings target for the quarter and 2) what it will take pipeline-wise to get us there (since we’re hypothetically at the start of the quarter, let’s assume we’ve generated $0 pipeline in Q2 thus far).

Our Q2 bookings forecast based on past pipeline generated and our win rate intake is the sum of:

• Q1’25: $6MM * 7.59% +1 quarter win rate = $455K
• Q4’24: $5MM * 6.70% +2 quarter win rate = $335K
• Q3’24: $4MM * 4.73% +3 quarter win rate = $189K
• Q2’25 start-of-quarter bookings forecast = $979K

Knowing our Q2 bookings target is $2MM and that we can count on $979K to close from our previously generated pipeline, we calculate that we have a $1.02MM bookings gap we need to close to hit our plan.

To find an additional $1.02MM in bookings for the quarter, we can use our in-quarter win rate (6.19%) to calculate the amount of new pipeline required to find the $1.02MM to close. In other words, we need to generate $1.02MM ÷ 6.19% ($16.5MM) of new pipeline this quarter.?

While nothing is impossible, $16.5MM of in-quarter pipeline is quite a stretch from the $6MM generated the prior quarter, so we will likely need to either re-forecast our plan or get very aggressive and creative about our top of funnel activities for Q2! But as always, sunlight is the best disinfectant: It’s better that we understand the pipeline shortfall so that we can work to rectify it.

Important nuances to consider

The aforementioned examples are deliberately high-level to convey a general construct for using cohort analysis for pipeline and bookings math, but companies can (and should) absolutely layer on added complexity—by month, by customer segment—as they grow and scale.

Before you go whole hog on cohort analysis, consider the following:

• Ignore the edge cases in your close rate intake curves. Every company has outlier deals that may take years to close; you should not include those deals in your intake curve. In our examples, my win rate math stopped at the +3 quarters out mark because for hypothetical Company X, most deals would close within a year if they stood a chance of getting done.
• What are the dynamics of your GTM motion / sales cycles? Your sales cycle will ultimately dictate the time horizon you use for cohort analysis. A quarterly assessment makes sense for deals with longer sales cycles, but companies who can close deals in two weeks may conduct cohort analysis at a level as granular as weekly. A weekly view is useless for 90-day sales cycles, and a quarterly review is too late for a 14-day sales cycle.
• Cohort analysis highlights risks in your ability to achieve your plan. As we summarized with the aforementioned Q2 bookings work-up, cohort analysis reveals risks in your plan. Risks are perfectly acceptable so long as they were identified and discussed before they come to fruition. It baffles me that revenue leaders will regularly sign their names to ambitious GTM plans that quite literally may not be possible. Data-driven leaders will do this math and push back on it where needed, but of course they need to have a solution to offer beyond the “this plan is impossible” statement. If the example where we calculated a $16.5MM pipeline need for Q2 assumed a 90-day sales cycle, the sales leader would be thinking about how they accelerate as much pipeline build as possible to the start of Q2 so that it has some chance of closing in-quarter, so perhaps they launch a SPIF to incentivize opportunities pipelined by April 15. They may also push the business to think about opportunities to tap the installed base for ARR expansion if the new logo plan feels untenable. Most importantly, that leader must be thinking about how to avoid a similar situation coming into Q3.
• It’s always easier to plug a pipeline gap when dealing with shorter sales cycles (read: more frequent cohort periods). This truth is self-evident, but still worth highlighting. SaaS companies with higher velocity sales cycles have more ability to “turn the dials” on their business quickly and effectively. If they can close deals with a few meetings and see a pipeline gap a few weeks out, there is typically a clear path to close those gaps. In a 90-day sales cycle business, there is very little you can do by way of near-term course correction, which is why cohort analysis is so important.?
• Effective forecasting includes both top-down projections from cohort analysis as well as a bottom-up consideration of all of the deals in play for that period. Cohort analysis is powerful directionally, but ultimately your forecast needs to consider the deals that are actually being worked in your pipeline. For slower sales cycles, cohort analysis tends to be more useful for projecting your out periods, as your in-quarter view will likely be more of a bottoms-up forecast of the deals nearing the finish line. But when your finance team is thinking about the feasibility of Q3’25 today, cohort analysis of the current pipeline is a helpful leading indicator.
• You must include seasonality and the influence of macro factors in your plan. Once you get the outputs of your cohort analysis, it is critical to either dampen or accelerate the assumption based on seasonal or macro factors. Many years ago I worked in a business that sold job board subscriptions to job seekers—and we loved cohort analysis! But importantly, once we had the results of the cohort analysis, we applied an X% “dampening” factor in the summer months (when people are less likely to be looking for jobs) and conversely, applied a bit of a “bonus” lift to the January cohort results knowing that jobseekers were kicking their activities into high gear that time of the year.

—

While this post focuses on the powerful role cohort analysis can play with pipeline math specifically, the practice has myriad applications beyond that—everything from churn analysis to customer adoption curves to marketing unsubscribe rates. I encourage every operator out there to learn it, love it and live it!