Preventable Churn vs. Structural Churn
I recently reflected on my experiences at FeedbackPanda and noticed that it's important to understand and differentiate two kinds of churn: preventable churn and structural churn. Preventable churn happens when someone chooses to cancel because they are moving to a competitor or don't use the product enough to warrant the price. Structural churn occurs when your customers go out of business, take a paternal break, or move on in their careers.
Imagine you're building a SaaS solution for up-and-coming actors to find their first gigs in the movie industry. There is a clear "happy path" churn here that you can't avoid: when your wannabe actor actually succeeds and becomes an actor that gets signed by an agent. They won't need your business anymore, or you elevated them into a new career category.
But there is a hidden kind of structural churn here, too. Every month, a certain percentage of actors-to-be decides to call it quits and "get a real job." They hang up their actor hat and venture out to interview for other occupations. They cancel because they don't need your SaaS anymore, as well.
Both are kinds of churn you can't prevent, success or failure. And it's pretty likely that the actors calling it quits outnumber the ones that make it big in the industry. That's a statistically traceable part of reality for every sector. Structural churn like this can be measured and should definitely be a part of your decision-making process on whether to take a shot at solving problems in this particular industry.
Churn itself is unavoidable —just as you will get new customers, some will leave your business over time— but you can learn as much as possible about how much there is to be expected for your field. Mind you, there will always be unexpected things that will cause your customers to cancel: sudden policy shifts or technological breakthroughs could trigger mass cancellations that you might struggle to prevent. But structural churn is visible from miles away. Make sure to look out for it.
This is a great framework. One addition: involuntary churn (payment failures) is probably the most preventable category and often gets lumped in with structural churn incorrectly.
When a subscription fails because of an expired card or bank decline, founders often assume the customer chose to leave. But these customers didn't make an active decision — they're often unaware the payment failed. Most would stay if you reached them within 24 hours.
That's why it deserves its own category: recoverable churn. The intervention is simple (a Day 1/3/7 email sequence after invoice.payment_failed), the recovery rate is 30-60%, and the customer didn't leave intentionally.
In your framework, I'd place it between "preventable" and "structural" — it looks structural, but it's actually preventable with the right tooling.
Hmm, wouldn't "don't use your product enough" fall under structural churn?
It just means your targeting wasn't narrowed down enough, and you somehow acquired a customer that isn't really your ideal buyer persona.
The idea is that it's them, not you. Meaning that, even though your product was excellent, it didn't matter to them, because they're just not the target market.
It's like trying to sell a Louis Vuitton or Coco Chanel handbag to me. To many, it's worth thousands of dollars. To me, it's worth exactly $0.
Doesn't make them poorly-made handbags.
That kind of churn is preventable — you said it yourself, just stopping to try to sell you the bad would be a BUSINESS CHOICE that reduces churn. The same goes for "not using it enough": build a better product or sell your existing product to different people. Anything you can affect will likely result in preventable churn (or, best case, prevented churn).
The difference between preventable and structural churn is agency: people canceling because of your product lacking features happens because YOU failed to build the right thing. People canceling because THEY gave up their acting aspirations has nothing to do with your business and everything to do with their choices.
You can't really do much about the unemployment rate in your country. That is systemic, structural.
At scale, structural churn should be okay.
Not a SaaS example though. For example, consider baby products industry - Their products are usually required by parents who has kids from 0-3 years. After 3 years, it's almost that the customer churns and the parent would never buy again after baby grows up (assuming a single baby). But this works at scale. If the brand is well established like Toys R Us, the business works.
SaaS example - 'Build Your Own Resume' products obviously has structural churn once the candidate gets a job, but if the cost to acquire a customer is low, it should be still okay for first time product builders if there are able to get an acquisition channel.
Thank you so much for bringing this up.
That's why I pointed out the "good" and the "bad" structural churn. And choosing a market where good structural churn is higher than bad structural churn is something self-funded entrepreneurs should aim for.
Let's make up a few numbers.
Your example is great because it has high good structural churn and low bad structural churn. Let's say for resume-generators, the good structural churn is 30% while bad churn —people stopping looking for jobs altogether— is as low as the unemployment rate of their particular country (which is a systemic kind of churn), maybe 10%.
Now let's take my example with the aspiring actors. I would argue that there are way fewer actors who "make it big" —maybe 5% if we're lucky— than those who bury their acting dream every month. I'd assume that the bad structural churn in this industry is somewhere around 20% or more.
From the looks of this, the ratio between these two might actually be a very important metric for your business and the industry as a whole.
For resume-generators, that would be a structural churn ratio of 3, while the aspiring-actor business has a structural churn ratio of 0.25. That's a huge difference, and that will be even more significant at scale.
What do you think about that?