The 2023 SaaS CEO Survey

All Acquisitions: Across the entire data set, searchers were most likely to acquire their companies for a multiple between 3x – 4x ARR (29% of total). The second most likely range was 2x – 3x ARR (24% of total).

Comparing the Cohorts: Almost all of the of the 3x – 4x acquisitions took place after 2019. Prior to 2019, deals were much more likely to trade within the 2x – 3x ARR range. Indeed, prior to 2019, 56% of searchers acquired their business for a multiple of 3x ARR or lower. After 2019, only 30% of searchers paid 3x or lower.

All Acquisitions: Across the entire data set, the most frequent response to our question of the EBITDA acquisition multiple was “not a meaningful number”, which suggests that the multiple was so high that it did not convey any meaningful insights. However:

Comparing the Cohorts: There appears to be a reasonably large disparity between the two cohorts: Prior to 2019, 55% of transactions took place at an EBITDA multiple of 8x or less. After 2019, only 30% of transactions fit that description. What’s more, after 2019, ~40% of all transactions took place at an EBITDA multiple that “was not meaningful”. That same figure for the pre-2019 cohort was only 11%.

Conclusion: The basis for valuation has shifted from predominantly EBITDA-based (in 2019 and prior years) to predominantly revenue-based in 2020 and beyond. Some of our analysis below digs deeper into why and how this shift might have occurred.

All Acquisitions: Across the entire data set, I was actually somewhat surprised to see that the top of the bell curve appeared between 10-30% EBITDA margins (24% of companies purchased were generating EBITDA margins between 10-20%, and a further 24% of companies purchased were generating EBITDA margins between 20-30%). This profitability profile is actually much higher than I would have initially suspected.

Comparing the Cohorts: The data across the two cohorts appears to be somewhat mixed:

• On one hand, search funds after 2019 were much more likely to purchase companies that were either barely profitable or not profitable at all: In the “2019 or prior” cohort, only 6% of targets reported EBITDA margins of 0% or less. After 2019 however, 22% of targets reported EBITDA margins of 0% or less.
• On the other hand however, for those companies that were profitable post-2019 targets appeared to be more profitable than their pre-2019 counterparts

All Acquisitions: Across the entire data set, target companies were most likely to be growing ARR at a rate between 10% – 20% annually (29% of total responses), followed closely by 40%+ annual growth (24% of total responses).

Comparing the Cohorts: After 2019, target companies appear to be growing more rapidly: In the “2019 or prior” cohort, only ~28% of all companies purchased were growing ARR at 30% or more. After 2019, 43% of all companies purchased were growing ARR at 30% or more.

Conclusions:

• After 2019, within the “rule of 40” equation, more weight has seemingly been ascribed to growth than to profitability;
• The increase in the growth rate of target companies may, at least in part, explain the higher valuations that have been paid since 2019.

Definition: EBITDA margin + % revenue growth rate

All Acquisitions: Across the entire data set, 46% of all companies acquired were 40% or more on the “rule of 40” scale, representing (by a large margin) the most frequent response

Comparing the Cohorts: Though 40%+ was the most frequent answer across both cohorts, companies purchased after 2019 were slightly more likely to be 40% or more on the rule of 40 scale (52% of total responses) than those that were purchased prior to 2019 (39% of total responses). As we saw above however, this appears to be largely attributable to the revenue component of the rule-of-40 equation, not the profitability component.

All Acquisitions: Across the entire data set, both cohorts exhibited a wide range of EBITDA values, with some target companies as small as $0-200K, and others as large as $2M+

Comparing the Cohorts: Consistent with our findings above, it appears that a greater proportion of sub-$1M EBITDA businesses were purchased after 2019:

• In the “2019 or earlier” cohort, only 39% of businesses had $1M of EBITDA or less.
• After 2019, 59% of businesses had $1M of EBITDA or less.

All Acquisitions: Across the entire data set, the greatest number of companies were purchased while generating $2 – $3M in total ARR.

Comparing the Cohorts: It appears as if target companies have been getting larger since 2019:

• In the “2019 or prior” cohort, only 39% of all companies purchased were generating more than $3M in ARR at closing.
• After 2019, a full 70% of all companies purchased were generating more than $3M in ARR at closing.

Conclusion: The increase in the size of target companies (as well as the higher growth rates) may, at least in part, explain the higher valuations that have been paid since 2019.

All Acquisitions: Across the entire data set, having at least 60% of total revenue coming from truly recurring sources appears to be table stakes (61% of all companies made this claim).

Comparing the Cohorts: In the “2019 or prior” cohort, only 17% of all acquisitions generated at least 80% of their revenue from recurring sources.
After 2019, however, a full 47% of all acquisitions generated at least 80% of their revenue from recurring sources. Searchers thus appear to be increasingly focused on buying businesses with a higher proportion of recurring revenue relative to total (this trend may also loosely coincide with the broader market’s gradual shift from perpetual-use pricing models to subscription-based pricing models)

Conclusion: Alongside larger revenue bases and faster growth rates, the higher proportion of total revenue coming from recurring sources may, at least in part, explain the higher valuations that have been paid since 2019.

All Acquisitions: Unlike many of the variables that we’ve presented thus far, the two cohorts appears to be more similar than different with respect to gross margin: In both cohorts, one-third of all companies acquired had a gross margin of at least 80%. Further, in both cohorts, ~70% of all companies acquired had a gross margin of at least 60%.

Comparing the Cohorts: The major difference between the two cohorts revolves around the percentage of companies purchased that were “only” generating a gross margin between 40-60%: 33% of all purchases in 2019 or earlier fit this description, but only 17% in the post-2019 cohort did. It’s difficult to accurately attribute this to anything in particular, as there is more variability than most people would think with respect to how software companies even calculate gross margin, specifically within the lower-middle market.

Definition: “For every customer who was a recurring revenue customer last year, what % were still recurring revenue customers this year?”

All Acquisitions: Across the entire data set, gross logo retention rates above 90% appear to be more or less “table stakes” (71% of all companies boasted a GLR rate above 90%). As I’ve said in the past, I believe GLR to be among the most informative metrics for any software business, and in a hypothetical world where I could only see a single metric to evaluate the health of a given software business, gross logo retention would likely be my choice.

Comparing the Cohorts: There does appear to be a difference in retention rates between the two cohorts:

• Prior to 2019, 89% of all companies acquired had a gross logo retention rate of at least 90%.
• After 2019, only 56% of all companies acquired had a gross logo retention rate of at least 90%.

Definition: “What percentage of my recurring revenue dollars from last year are still recurring revenue dollars this year, including the dollar value of customer upsells and downgrades?”

All Acquisitions: Across the entire data set, net revenue retention rates of at least 95% appear to be more or less table stakes (88% of all companies had NRR of 95% or more).

Comparing the Cohorts: Though the two cohorts appear to be reasonably similar, there does appear to be a slight bias towards acquiring higher NRR businesses after 2019: 35% of all companies acquired after 2019 boasted NRR of 105%+, while only 17% of companies acquired in 2019 or earlier can say the same

Conclusions: It is interesting to note that pre-2019 targets featured better GLR rates but worse NRR rates relative to their post-2019 peers. One possible conclusion is that post-2019 targets tend to grow or maintain their recurring revenues through more upsells (or fewer downgrades) from existing customers, as opposed to maintaining a higher a retention rate of the absolute number of customers served.

Definition: % of total revenue represented by the company’s largest customer at the time of acquisition

All Acquisitions: Across the entire data set, target companies were most likely to have negligible levels of customer concentration, with 0%-5% of total revenue coming from their largest single customer (~29% of all responses). Concentration isn’t unheard of however, as the second most frequent response (at ~17% of all responses) was that the company’s largest customer accounted for 15% – 20% of total revenue

Comparing the Cohorts: The data across the cohorts is somewhat mixed: On one hand, pre-2019 targets were less likely to mild have concentration than post-2019 targets, however they were also more likely to have meaningful concentration (defined as 35% of sales or more) than their post-2019 peers.

All Acquisitions: Across the entire data set, transactions were most likely to be funded with a minimal amount of debt (defined at 10% or less of total capitalization). This represented approximately one-third of all responses, and was largely consistent with my expectations

Comparing the Cohorts: I was however quite surprised to see that transactions that took place after 2019 were actually more likely to use more leverage than their pre-2019 counterparts:

• In the “2019 or earlier” cohort, 50% of all transactions used 20% leverage or more.
• In the post-2019 cohort, 61% of all acquisitions fit that same description.
• Given that the post-2019 cohort seems to be targeting higher-growth, lower-profitability businesses, I would have expected the post-2019 cohort to be considerably less leveraged than the pre-2019 cohort.

All Acquisitions: Across the entire data set, companies were most likely to have at least 300 customers at the time of acquisition (37% of all responses)

Comparing the Cohorts: Companies purchased after 2019 do indeed seem to be getting larger as measured by the size of their customer bases (this is consistent with our analysis above, which suggests that post-2019 acquisitions were, on average, generating more ARR than their pre-2019 counterparts):

• In the “2019 or earlier” cohort, only 33% of companies had 150 or more customers at the time of acquisition.
• For the post-2019 cohort, 52% of all companies had 150 or more customers.

Conclusion: This consideration may, at least in part, have contributed to the higher valuations that have been paid after 2019 (under the assumption that a greater number of customers produce more ARR). It is also consistent with our observation above that post-2019 acquisitions are less likely to have meaningful customer concentration relative to their pre-2019 counterparts.

Definition: “What is the lifetime value of an “average” customer, expressed as a multiple of the cost to acquire that same “average” customer? (higher = better)

All Acquisitions: Across the entire data set, I was not particularly surprised to see that 63% of all respondents said that they either didn’t know the company’s LTV/CAC ratio, or couldn’t measure it with any real degree of precision. Though LTV/CAC is generally considered to be a “SaaS 101” type of metric, accurately measuring it in small, privately held companies with low information hygiene is often much more difficult than most might suspect.

Comparing the Cohorts: Though “I don’t know” was indeed the most frequent answer within both cohorts, for those companies where LTV/CAC could be calculated, the post-2019 cohort did seem to demonstrate better unit economics relative to the pre-2019 cohort: 35% of all post-2019 companies had LTV/CAC ratios of 5x or more, while only 11% of pre-2019 acquisitions could make the same claim.

Conclusion: Again, more favorable unit economics could have played a material role in the higher valuations that have been paid since 2019.

All Acquisitions: Across the entire data set, by far the most frequent response (at ~34% of all responses) was “beyond day-to-day working capital requirements, we funded an additional $900K or more” to the balance sheet as part of closing. The frequency with which this response was provided was somewhat surprising to me, as the majority of search fund acquisitions outside of software typically require only a modest amount of cash to the funded to the balance sheet at close beyond day-to-day working capital requirements.

Comparing the Cohorts: By now it shouldn’t surprise you to hear that transactions consummated after 2019 were more likely to require a material amount of cash funded to the balance sheet relative to those consummated prior to 2019:

• 39% of all post-2019 deals funded $900K or more to the balance sheet, while only 28% of pre-2019 deals required the same.
• 39% of all pre-2019 deals funded nothing to the balance sheet beyond day-to-day working capital requirements, while only 22% of post-2019 deals could claim the same.

Idea: Anecdotally, I’ve observed that many new SaaS CEOs comment on how the product that they inherited is in worse shape than they had originally suspected. I wanted to test to see whether this was indeed true, and whether it differed materially between the two cohorts.

Comparing the Cohorts: For those companies purchased in 2019 or earlier, a full 72% of CEOs reported that the state of the product, code base, and technical debt was in worse shape than they had suspected. Only 30% of post-2019 CEOs shared that same sentiment. Indeed, in the post-2019 cohort, CEOs were much more likely to be reasonably satisfied with the state of their products, with ~70% of respondents reporting that the product was “about the same as I had thought” or “better than I thought”.

Conclusions: Though the magnitude of the delta between the two cohorts was higher than I would have thought, I wasn’t necessarily surprised to see that post-2019 targets tended to have better quality products/code bases than their pre-2019 counterparts. This could be due to a combination of a) Better pre-purchase product due diligence performed by more experienced third-party advisors; b) Fewer legacy on-premise businesses with significant technical debt available for purchase; and/or c) the propensity of software companies to update their tech stacks periodically over time (sometimes in a very meaningful way)

All Acquisitions: Across the entire data set, 63% of companies saw their EBITDA margins decline post-acquisition, while only 15% reported them increasing.

Comparing the Cohorts: Based on our “state of the product” observations above, we shouldn’t be surprised to see that the pre-2019 companies were more likely to see their margins decline after purchase than were their post-2019 peers:

• In the pre-2019 cohort, only 28% of companies saw their margins stay flat or increase post-closing.
• In the post-2019 cohort, 43% of companies saw their margins either stay flat or increase.

Conclusions: I have long suspected that search funds don’t model in enough margin contraction post-close in their base cases. Margin contraction in the early years can be common, and is often attributable to a combination of: a) Product-led investments, often of a “defensive” sort (see next section below); b) Sales & Marketing investments (again, see next section below); c) Historical under-investment in tools, people, and processes by the previous company owner; and/or d) replacing the work that had previously been done by the selling shareholder (that can often necessitate 2-4 incremental hires, as the new incoming owner doesn’t have the decades of knowledge and experience that their predecessors often did).

Idea: For those companies who have seen margins decline since the purchase, I was curious to better understand where the opex spend was primarily going.

All Acquisitions: Across the entire data set (only those companies who have seen EBITDA margins decline since purchase), two responses presented themselves with equal frequency: i) Product (including engineers, QA, infrastructure, product management, etc.) (~36% of total responses); & b) Sales & Marketing (also ~36% of total responses)

Comparing the Cohorts: Based on our analysis thus far, we shouldn’t be surprised to see that the pre-2019 acquisitions were more likely to direct their opex to fixing product-related problems than their post-2019 counterparts: In the “2019 or earlier” cohort, 44% of CEOs said that their margin declines were mostly attributable to product considerations. In the post-2019 cohort, only 29% of CEOs reported something similar.

Idea: Because of the importance of revenue growth to the “typical” SaaS investment thesis, I wanted to better understand how the growth of the companies now, under new leadership, compare to the rate of growth at the time of purchase, under the previous leadership group.

All Acquisitions: The cohorts appear to be very similar in this domain:

• 55% of pre-2019 purchases recorded a higher LTM growth rate than the company was producing at the time of purchase
• 52% of post-2019 companies reported the same.
• Across the entire data set, only 12% of companies reported a lower revenue growth rate in comparison to when the company was purchased.

Idea: Although growing the value of a company is an incredibly complex and multi-variable exercise, at the risk of oversimplifying I wanted to better understand the primary components of each CEO’s plan to create equity value, ranked from most to least important

All Acquisitions: Across the entire data set, revenue growth was by far the most important value creation lever, with 80% of all respondents classifying it as the most important component of their value creation plan. Similarly, across both cohorts, leverage was classified as the least important component to the value creation plan. Based on our analysis above, this is unsurprising.

Comparing the Cohorts: Unsurprisingly, the post-2019 acquisitions do seem to have a bit more of a growth orientation relative to their pre-2019 peers: 87% of post-2019 companies ranked revenue growth as their most important consideration, while only 72% of pre-2019 companies made the same claim.

Interestingly, nobody, in either cohort, classified profitability growth or margin expansion as the most foundational component of their plan to create equity value. This is one variable in which the SaaS investment thesis appears to differ materially from the more typical non-SaaS investment thesis within the search fund ecosystem