With the rise of SaaS businesses, retention rate is often discussed and followed by investors.
Here are some of the notes I took from an academic paper discussing illusions/misconceptions when it comes to retention rates.
2/9 Issue #1: Reported retention rate may not be indicative of realized renewal rate. Let me give an example.
Let’s say a business reports retention rate is 95% which is, of course, awesome. What is less discussed, however, is the duration of the customer contracts.
3/9 To illustrate why it is important, imagine Company A, B, and C all report 95% retention rates, but customers only renew the contracts in every 1, 3, and 5 year respectively.
Here’s how the reported and underlying retention rate differs for these companies.
4/9 “without knowing the duration of the contract, it is very hard to indicate whether these high retention rates are the result of a high renewal rate or driven by the inability of customers to cancel their contract.”
5/9 Issue #2: Companies can *temporarily* increase retention rates by providing discounts
If a business has high customer retention rates but low revenue retention rates, it may be indicative of steep discount the company is providing its customers to window dress.
6/9 Issue #3: If customers are not homogenous, historical retention rate may not be indicative of future retention rate.
Retention rate is likely to be different when you acquire customers organically vs advertisement or promotional offers.
7/9 Issue #4: When a company takes an ecosystem/platform approach and creates a diversified product portfolio, switching cost can incrementally increase.
In this case, future retention rate can be higher than historical one.
8/9 Issue #5: Important to understand whether retention rate is reported on an annual basis or monthly basis.
When a company reports 90% retention, if they report annual retention, it means 10% churn, but if it means monthly retention, it may mean 72% churn on an annual basis!
ADSK had a pretty decent 1Q, comfortably beating high end of the revenue guidance. 98% of revenue are now recurring, and net revenue retention was in the range of 100-110%.
Topline guidance was raised by ~$40 mn. Here are my notes from the call
2/ Q1 is expected to be trough from growth standpoint and the rest of the year is likely to have some acceleration post-pandemic. ~75% of FCF of this year will be generated in the 2H.
3/ Billings from converting noncompliant users doubled YoY in Q1. In fact, a noncompliant customer converted into one of the largest premium customers.
But don’t expect hockey stick growth from conversion of noncompliant users. ADSK wants to gradually and naturally convert.
It's easy to read Keynes' quote, "When the facts change, I change my mind - what do you do, sir?" and nod your head; but it doesn't make it any less difficult for anyone to change their opinions.
Why is that?
2/8 Murakami had a great quote that I try to remind myself every time I disagree with someone:
“Always remember that to argue, and win, is to break down the reality of the person you are arguing against. It is painful to lose your reality, so be kind, even if you are right.”
3/8 Unfortunately, one of the downsides of arguments/debates on twitter or any social media is it's mostly performative in nature.
You not only lose arguments that shatter your "reality", you go through the experience publicly which makes it even harder to accept and change.
@mjmauboussin and Callahan recently published a piece highlighting Customer-based Corporate Valuation (CBCV) which works particularly well for subscription businesses.
Here are my notes.
2/ Subscription based businesses’ topline grew 17.8% in 2012-2020 period whereas S&P 500 grew sales ~2% during the same time.
Digital subscription has also expanded TAM significantly for some businesses.
$NYT print subscription in 1996: 1.1 mn vs digital subs in 2021: 5.3 mn
3/ The CBCV framework appears simple: customer value comes from existing and future customers. Of course, the devil is the details.
One of the questions most investors ask themselves at least at some point is whether they are indeed good investors, or all their past success are just random luck which by definition may not persist.
2/ My basic assumption is I am probably not a great investor. Even to be average, it will require a lot of work for me.
A common retort is why bother investing then? If I am so unsure of myself as an investor, shouldn’t I just index?
3/ This feels like equivalent of telling kids there’s no point in playing basketball because they’re never gonna make it to NBA.
I doubt most NBA players knew before touching the basketball that they are probably very good at it and it might be possible to make it a profession.
1/ I have thought about it a little more and now I'm a bit confused whether FactSet's methodology is actually better than Grant's. Let me explain what's giving me second thoughts.
2/ if we imagine overall earnings of today's market is a stable pool of total earnings that grows at a historically similar rate, it perhaps makes sense to just the total earnings of the market rather than adjusting it by their mcap weights.
3/ Companies within today's index will die or be left out and new companies will join, and overall profit pool in the economy will just shift around to companies that are creating the most value.
1/6 Thread: What is the market's valuation multiple?
One of the reasons I was becoming bullish on some of the big tech stocks ~2-3 years ago is some of the them, especially $AAPL was trading at market multiple.
2/6 That didn't make quite sense to me since I thought Apple was clearly a better business than the broader "market".
I now think there was a flaw in that argument.
If I remember correctly, I first came to know about this flaw from one of the pieces by Jim Grant.
3/6 Grant's argument was really simple.
The way market valuation multiple is calculated by FactSet/CapIQ can understate the "true" multiple of the market. Let me explain.