⚠️ Word of caution before you proceed: Please excuse me for the honesty.
As experienced marketers amongst y’all might already have a clue, Stoa Daily is a great top-of-funnel content marketing tactic for us to attract quality leads for the Stoa program. People who like reading long-form business content on a regular basis generally tend to be a good fit for the program, and the best way to attract such people is by writing long-form content ourselves.
But then, being the growth marketers we are, we wish to see fast growth in the number of subscribers, so that we can rest in the safety of numbers.
“Even if 5% of our subscriber base converts, we are good!” is what we tell ourselves.
And yes, we also know Goodhart’s Law: whenever a measure becomes a goal to optimize, it stops being a good measure. But the allure of stacking numbers is too great, and we don’t think of ourselves as people who are beyond employing growth hacks to increase our distribution — although we try to be a bit tasteful and graceful in how we go about it.
Anyway, as you all must know by now, this week we started a Referral Program for Stoa Daily.
Our aim: incentivize people to share the good stuff with their friends by offering them rewards when they hit certain referral count milestones. At the same time, new subscribers might go through our archives and we might be able to get some quality organic user-generated content that way.
Very straightforward, very conventional, nothing non-obvious about it.
But…
We might have not managed to think through all the second-order consequences this Referral Program might have, although most of us had a good hunch around things that could go wrong.
But being the street scientists that we are, we couldn’t keep ourselves from a good experiment, just for the sake of learning what happens when one decides to be optimistic about stuff.
As Goodhart's Law would have predicted, people started gaming the system.
They started subscribing via their referral link using their own alternate email ids.
They started using hacks to subscribe multiple times using the same email id.
They started signing up their parents, grandparents, unborn children, extended family... everyone who possibly had an email id — regardless of whether they were interested in the content or not.
In fact, we suspect that without the double opt-in feature, people might just start uploading their entire email list without their permission, just to stack up those numbers.
There are some short and long-term second order effects of this, that I would like to discuss:
1. Our calculated customer acquisition cost (CAC) for the campaign might be an illusion.
Assuming 50% of the new subscribers are genuine lovers of good business content, our actual CAC is 2x of the calculated CAC. And 50% seems likely to be an extremely healthy and optimistic assumption. If it's really 20%, our actual CAC will be 5x of assumed CAC.
2. We might create a negative brand perception.
We have gotten emails from a few people saying they didn't sign up for the newsletter, which makes us think that their email ids were referred into the system by others, without their permission.
Some of these might just mark our emails as spam and think of us as a brand that indulges in shady marketing practices and spams you with unsolicited emails.
3. Email open rates, our primary metric, may go for a toss
As a lot of uninterested people are subscribing to the newsletter purely for the sake of helping their friends and family members stack up the referral count, they likely won't be reading the newsletter regularly. This will negatively impact and obfuscate our open rate metric — the one genuine metric we have for measuring if the content is good enough to keep people's attention and provide value.
4. Rate of unsubscribes might go up in the short term
People might just subscribe to help their friends out and then unsubscribe, leading to huge number of unsubscribes in the short term.
Now, we are just 2-3 days into the referral campaign and not everything is as dreary and pessimistic as it might appear.
We've already seen some new subscribers talking about a few essays in our archives on their socials, and the subscriber count also has gone up by over 20% in just the past two days.
On an absolute scale, the referral program is working quite well, considering it will get us a nice influx of sincere professionals who might find our essays useful and insightful. But how the campaign fares on relative terms, compared to just pure word-of-mouth (the best and most preferred marketing) remains to be seen.
There are upsides. There are downsides. You just need to make sure there aren't any irreversible downsides.
With any marketing campaign focused on optimizing a single proxy metric — subscriber count in this case — there are always instances where the supposed growth in the short term might just be a kind of pseudo-growth, with only a fraction of the revenue impact that was predicted before the start of the campaign.
As the metric becomes the target, it ceases to be a good metric.
But we just wanted to experiment and see what happens. So, that's what we did. And that's how we will learn.
