Most SaaS acquisitions fail because buyers trust the headline retention rate. This guide shows you how to dissect monthly cohort data to uncover the true health of a subscription business.
Deal Alert AI is reader-supported. We earn commissions from affiliate links at no cost to you.
This post is based on a video from our Deal Alert AI YouTube channel. Watch the original or read the full breakdown below.
When you are sitting across the table from a SaaS founder, the first slide they show you is almost always the "Logo Retention" number. It’s usually a glossy, big bold number. "95% Retention," it says. You feel a spark of confidence. The business is sticky, the customers are happy, and the revenue floor is solid. Here is the hard truth: that number is likely lying to you.
In my experience advising buyers at Deal Alert AI, I have seen too many investors walk away from profitable SaaS businesses simply because they focused on the wrong metric. Logo retention is a vanity metric. It tells you that the accounts did not churn, but it tells you nothing about revenue retention. A company can have 95% logo retention but only 60% Monthly Recurring Revenue (MRR) retention if they are forcing customers onto lower tiers or losing the high-value seats within their teams.
To buy a profitable online business, you cannot rely on marketing dashboards. You need to dive into the raw data. Specifically, you need to master the art of analyzing cohort retention charts. This is the single most important data point in any SaaS due diligence process. It reveals the actual behavior of your users over time. If you can read these charts like a surgeon reading X-rays, you will never overpay for a broken product again. The rest of this article is a step-by-step guide to doing exactly that.
We scan Empire Flippers, Flippa, Acquire.com and Quiet Light daily — scoring every listing. Start free.
A standard cohort retention chart is a heat map. The rows represent cohorts of users grouped by the month they signed up. The columns represent the months elapsed since they signed up. The cell at the intersection shows the percentage of that specific cohort that is still active in that specific month. For example, the "Cohort 1, Month 1" cell shows who stayed from their first month to their second. The "Cohort 1, Month 12" cell shows who has stayed for a full year.
Most buyers make the mistake of looking at the average of all months or just the diagonal line. You must look at the shape of the curve. A healthy SaaS business should show a "cliff" in the first month, where perhaps 20% to 30% of users drop off due to onboarding issues or buyer's remorse. After that, the retention curve should flatten out quickly. By month 3 or 4, you want to see the retention percentage stabilize. If the curve continues to drop sharply for six months, you have a product with severe engagement issues that will likely burn out long-term users.
Consider this real-world scenario. I recently reviewed a project on Empire Flippers, a B2B marketing tool that claimed 80% Year 1 retention. It looked impressive on the surface. However, when I zoomed in on the cohort data, I saw that Month 1 retention was 60%, Month 2 was 50%, and it dipped to 40% by Month 6. The "Year 1" number was an average of old cohorts that had higher initial retention from a pre-price-increase era. The newer cohorts were bleeding users much faster. The true retention baseline was closer to 45%, not 80%. This single insight saved our client from buying a business that was effectively shrinking.
This is where the math gets critical. You must distinguish between Gross Revenue Retention (GRR) and Net Revenue Retention (NRR). Gross Retention measures the revenue kept from existing customers, excluding any new sales or upsells. It is the purest measure of how "sticky" your product is. If you lose a customer, it hurts your GRR. If you lose a seat within a customer, it hurts your GRR.
Net Retention includes expansions. If you lose a customer but gain $10,000 from upselling another customer, your Net Retention might stay positive. However, as an acquirer, you care deeply about GRR first. Why? Because NRR can be manufactured. If a product is hard to use, users will churn, but if your sales team is aggressive, you can push extra seats to happy users to mask the churn. This creates a "leaky bucket" scenario where you are constantly filling the bucket and patching leaks, rather than having a stable, capped vessel.
When analyzing the cohort chart, look for the "Churn vs. Expansion" ratio. If the chart shows a stable count of active users, but the MRR is fluctuating wildly, you are likely dealing with price-gouging that is triggering a "land and expand" strategy that isn't scaling. Healthy cohorts show a decline in the number of cohorts (users) but consistency or growth in the value per cohort. If the value per cohort is dropping, your product is becoming less relevant over time. This is a red flag that demands immediate investigation into customer support tickets and NPS scores.
One of the most deceptive patterns in SaaS due diligence is the "Zombie Cohort." This happens when a company had a viral marketing spike in the past, bringing in a massive chunk of low-quality traffic. These users signed up, got a discount, and then disappeared. On the surface, the chart looks huge because of the volume of users in that specific row. However, the retention percentage for that row will crash in Month 2.
Now, compare that to a "Super Cohort." This is a small group of users who signed up during a period of high product maturity or strong brand advocacy. Their retention stays flat and high. Sophisticated buyers at Deal Alert AI ignore the volume of the cohort and focus on the retention percentage of the most recent 3 to 6 cohorts. The history of a company matters, but the trajectory of its newest customers predicts the future. If the last three cohorts show a downward trend in retention curves, the business is deteriorating. The fact that older cohorts are sticky is irrelevant because they are shrinking in number.
You might think, "Well, the average is still good." This is a statistical trap. Averages hide volatility. Imagine a company with two products. Product A has 99% retention but zero growth. Product B has 50% retention but 500% growth. The blended average looks healthy. But if you buy the business, you are betting on Product B succeeding where it has historically failed to retain users. You must segment the cohort charts by product or feature set if possible. If the seller refuses to provide segmented data, assume the worst-case scenario: that the high retention is coming from the stagnant product, and the growing product is leaking users.
Customer Lifetime Value (LTV) is the holy grail metric. But most sellers hand you an LTV calculated as "Average Month 1 Revenue multiplied by Average Churn Rate." This is useless and often wildly inaccurate. Instead, you need to calculate a "Real LTV" based on the observed behavior of cohorts that have reached at least 12 to 18 months of age.
Take a cohort that is 18 months old. Sum up all the recurring revenue that cohort generated in Month 1, Month 2, through Month 18. Then, project that forward to infinity using the flattening rate of the curve. If the curve has flattened at 40% retention, you assume the remaining 40% will stay for a long time. This gives you a true, data-backed LTV. Compare this number to the Average Customer Acquisition Cost (CAC). The ratio should ideally be at least 3:1, meaning the lifetime value is three times the cost to acquire the customer.
If your calculated Real LTV is lower than the seller’s claimed LTV, you have a serious discrepancy. Often, sellers inflate LTV by assuming customers stay for years. For niche tools, customers might stay for 6 months on average. If you buy based on a 3-year average life, you are paying for revenue that doesn't exist for most of your customer base. I have seen deals fall through because the buyer realized the CAC:LTV ratio was actually 1:1.2, but the seller claimed it was 1:5. The difference determines whether the business is an asset or a liability. Always run the numbers yourself using a spreadsheet imported from the raw data. Do not trust the dashboard.
Sellers are smart. They know which metrics matter. Consequently, some engage in subtle data manipulation to make their retention charts look better. The most common trick is "cohort shifting." This occurs when a company changes its definition of a "new user." For example, if a user upgrades from Free to Pro, do they count as a new cohort or stay in the old one? If they count as "new" in the Pro cohort, the start of the Pro retention curve starts high. If they stay in the old cohort, the "Pro" revenue is diluted. Ask specifically how their software defines cohort entry. If the definition is vague or inconsistent, request a raw SQL query of the user logs.
Another tactic is "trial-to-paid leakage." A buyer might see high retention for paid customers. But if that cohort was formed by a 30-day free trial, the "Month 0" is actually the end of the trial. The churn that happens during the trial is never seen on the "Retention Chart" because the chart only tracks users who converted to paid. The real churn is happening before the chart starts. You must look at the "Trial-to-Paid Conversion Rate" for each cohort. If the conversion rate drops for newer cohorts, your future retention will also drop, even if the past paid cohorts look great. This is a leading indicator, not a lagging one.
Furthermore, watch out for "seasonal skew." If you are looking at a B2B company that targets financial services, their cohorts will be weird. They might have huge sign-ups in January (new budgets) and zero sign-ups in August. The retention chart will look jagged. Do not panic. Normalize the data by looking at "cohort age" rather than "calendar month." A user who signed up in January 2023 compared to a user who signed up in August 2023 at "Month 3" are comparable. Aligning by age removes the seasonal noise. This is a simple but crucial adjustment that saves many buyers from misinterpreting the data.
To streamline this process, I have compiled the essential steps I use for every SaaS acquisition. This is not just a tech audit; it is a behavioral audit. You need to verify that the data you see matches the reality of the business. Use this checklist to interrogate the seller's data room. If you cannot get the data, the deal is dead. Transparency in data is a proxy for transparency in ethics.
Now that you have the skills to read the data, where do you find businesses with transparent data? Not every listing on the open market provides raw SQL access or detailed cohort breakdowns. This is why vetting platforms and brokerages matter. You want to work with sources that have already conducted a level of technical audit.
Platforms like Flippa offer a wide variety of opportunities, but you must filter aggressively for SaaS with at least 12 months of history and, ideally, those that have already screened on their specific SaaS metrics. On larger platforms, you will find everything from one-person side projects to enterprise-scale platforms. The key is to ignore the ads and focus on the seller's ability to answer your data requests. If a seller is hesitant to share raw data, they are hiding a problem.
For higher-end SaaS opportunities, established brokers often handle the initial vetting. However, always maintain your own independence. Deal Alert AI connects you with vetted opportunities where the financials have been cleared for potential issues. Our platform helps you filter out the noise so you can focus on the deals that match your retention thresholds. Remember, the goal is not just to buy a business; it is to buy a predictable stream of cash flow. Predictability comes from stable cohorts. Master the chart, and you master the deal.
Buying a SaaS business is not about feeling good about a cool product. It is about analyzing historical behavior to predict future cash flow. The cohort retention chart is the single most powerful tool in your arsenal for this task. It strips away the marketing fluff and shows you the truth: are people staying, are they paying for ongoing value, or are they slowly drifting away?
As you continue your journey as an online business acquirer, make this analysis your non-negotiable standard. If a seller cannot produce the data, walk away. If the data shows a "cliff" that doesn't flatten, walk away. If the GRR is low, negotiate the price down to reflect the high churn risk. The market is full of mediocre SaaS businesses being sold at premium valuations because buyers don't know how to read the data. Be the buyer who does. That edge will save you hundreds of thousands of dollars over the next few years.
Start small. Take a public company’s SaaS metrics, or find a niche SaaS you use, and try to model their retention based on your own experience. Compare it with their reported metrics. This practice will sharpen your eye. The more deals you review using this framework, the faster you can spot the red flags. Accuracy is your best defense. Use these insights to build a durable portfolio of online assets that pay dividends, not ones that drain your capital. Your future self will thank you for digging deep into the cohorts today.
We scan Empire Flippers, Acquire, Flippa, and Quiet Light daily. The best sub-$500K businesses are gone within 48 hours.
We scan Empire Flippers, Flippa & Acquire every morning. The best deals sell in 48 hours.