Clouded Judgement 3.31.23

Every week I’ll provide updates on the latest trends in cloud software companies. Follow along to stay up to date!

We’re through Q1 ‘23, and just like everyone expected the Nasdaq is up ~15% YTD, and up 20% from it’s low… What a start to the year!

How Will AI Effect Software Business Models?

Like many, I’ve been thinking about how AI and foundation models will effect the world of software. In particular - how AI will effect software business models. There are two main topics I’ve been pondering lately:

1. Margins. Generally application software companies have 80-85% gross margins (infrastructure software companies closer to 70%). The gross margin profile dictates how much you can spend on operating expenses to ultimately generate free cash flow. The majority of COGS (revenue less COGS = gross profit) fall in hosting costs (ie AWS), and some customer support. The challenge with AI is that it introduces another (potentially large) component to COGS. Ultimately, if an application software company wants to incorporate OpenAI (or an alternative like Anthropic, Cohere, etc) into their product, they’ll have to pay to do so. Today, the cost of inference (this is the incremental marginal cost of asking a question of an AI model and getting back an answer) is non trivial. If an application software company wanted to add AI features to their product it could ultimately have a 5-10% hit on their gross margins (I haven’t actually done the math, this is really just a total guess). The only way for software companies to avoid this gross margin compression when incorporating AI features is to raise prices. And herein lies the big question - do the end users get enough incremental value out of AI features to pay more, and if so how much? At the end of the day the end user doesn’t care if fancy AI is used, they just want more problems solved. Let’s use an example: Notion (i’m not an investor in Notion so this example is purely speculative, I don’t know anything about their plans for AI / their financials). If we head to their pricing page we’ll see that the most popular plan is $8 / user / month. If you want to use Notion AI, the cost is $16 / user / month. Double the price! To ask the question again - will Notion users say it’s “worth it” to pay double for AI features? Only time will tell. I imagine they saw quite a boost initially as everyone wanted to try it out, but we’ll have to wait a few months to see where things settle. If they don’t want to pay double the price, then Notion has a decision to make - do they sell the Notion AI SKU at a price that dilutes gross margins if it ends up generating more revenue? If so, then these AI features may end up structurally degrading margins which leaves less room to spend on S&M (and could end up structurally impeding profitability). To dig a level deeper on this analogy, large incumbent competitors to Notion, like Microsoft Loop, have a lot more room to play around with margins. Microsoft generates tens of billions of free cash flow / year, and can certainly afford to “give away” these AI features for free (ie take a margin hit on selling the AI SKU). Especially if it means sucking the oxygen out of the room for companies without the benefit of a massive balance sheet. End customers then have a decision to make - go with the cheaper alternative from Microsoft, or pay more for Notion? This dilemma is why current popular opinion is that incumbents will capture a lot of the value. They have massive distribution (installed customers + sales teams), and margin to play around with. Startups and younger companies don’t have this luxury. The positive spin here is that the cost of inference, and thus what players like OpenAI charge companies, will certainly come down in the future. If it comes down enough, we may not see a structural shift in gross margins in software companies who are not able to pass through the cost of AI to their end paying customers.

2. Consumption Pricing. This one is interesting to me (and a little more abstract than the above discussion on margins). If we believe that AI will ultimately allow us to do “more with less,” we may see headcount growth slow for traditional roles. What do I mean by this? Companies could end up hiring less SDRs but booking more sales demos. Or hire less engineers but write more code with tools like GitHub Copilot. Or hire less data engineers but write more SQL queries. The list goes on. The pattern here: less people but more usage / output. For businesses tied to seat based models this may pose a challenge! Here’s the cycle - Customers start consuming more and more of your product. As we talked about above, the more use of AI features, the more costs from players like OpenAI flow through your COGS. However, as seats aren’t growing as quickly (these AI features allow users to do more with less), you end up with a scenario where COGS are growing faster than revenue. I don’t really have a solution here, other than to say companies that charge on consumption don’t have this problem (or at least this problem to the same extent). As more of the product is used (ie as more SQL queries executed), your revenue goes up (regardless of # of seats). I’ll do some more thinking on this is a future post, just thinking out loud for now…

Quarterly Reports Summary

Top 10 EV / NTM Revenue Multiples

Top 10 Weekly Share Price Movement

Update on Multiples

SaaS businesses are generally valued on a multiple of their revenue - in most cases the projected revenue for the next 12 months. Revenue multiples are a shorthand valuation framework. Given most software companies are not profitable, or not generating meaningful FCF, it’s the only metric to compare the entire industry against. Even a DCF is riddled with long term assumptions. The promise of SaaS is that growth in the early years leads to profits in the mature years. Multiples shown below are calculated by taking the Enterprise Value (market cap + debt - cash) / NTM revenue.

Overall Stats:

• Overall Median: 6.1x

• Top 5 Median: 11.4x

• 10Y: 3.6%

Bucketed by Growth. In the buckets below I consider high growth >30% projected NTM growth, mid growth 15%-30% and low growth <15%

• High Growth Median: 8.8x

• Mid Growth Median: 6.3x

• Low Growth Median: 3.5x

Scatter Plot of EV / NTM Rev Multiple vs NTM Rev Growth

How correlated is growth to valuation multiple?

Growth Adjusted EV / NTM Rev

The below chart shows the EV / NTM revenue multiple divided by NTM consensus growth expectations. The goal of this graph is to show how relatively cheap / expensive each stock is relative to their growth expectations

Operating Metrics

• Median NTM growth rate: 16%

• Median LTM growth rate: 26%

• Median Gross Margin: 74%

• Median Operating Margin (22%)

• Median FCF Margin: 2%

• Median Net Retention: 116%

• Median CAC Payback: 30 months

• Median S&M % Revenue: 47%

• Median R&D % Revenue: 28%

• Median G&A % Revenue: 18%

Comps Output

Rule of 40 shows LTM growth rate + LTM FCF Margin. FCF calculated as Cash Flow from Operations - Capital Expenditures

GM Adjusted Payback is calculated as: (Previous Q S&M) / (Net New ARR in Q x Gross Margin) x 12 . It shows the number of months it takes for a SaaS business to payback their fully burdened CAC on a gross profit basis. Most public companies don’t report net new ARR, so I’m taking an implied ARR metric (quarterly subscription revenue x 4). Net new ARR is simply the ARR of the current quarter, minus the ARR of the previous quarter. Companies that do not disclose subscription rev have been left out of the analysis and are listed as NA.

This post and the information presented are intended for informational purposes only. The views expressed herein are the author’s alone and do not constitute an offer to sell, or a recommendation to purchase, or a solicitation of an offer to buy, any security, nor a recommendation for any investment product or service. While certain information contained herein has been obtained from sources believed to be reliable, neither the author nor any of his employers or their affiliates have independently verified this information, and its accuracy and completeness cannot be guaranteed. Accordingly, no representation or warranty, express or implied, is made as to, and no reliance should be placed on, the fairness, accuracy, timeliness or completeness of this information. The author and all employers and their affiliated persons assume no liability for this information and no obligation to update the information or analysis contained herein in the future.