The Sample publisher report

• I redesigned the publisher console. It's split up into several different pages now instead of having almost everything on one page. I think it's nicer to use now, and also this will let me add more things easily without having to stuff it all on one page.

• For cross-promotion, I've made it so your number of referrals includes people who submit newsletters. Previously it only included people who subscribed as readers. If someone both submits a newsletter and subscribes as a reader after clicking your referral link, it'll count as two referrals. So if you've shared your referral link in the past, you may get more cross-promotion forwards over the next several weeks as the algorithm adjusts.

Some user-facing things I'm planning to do this week:

• Make an auto-updated page for the data in this spreadsheet, to help people get an idea of how many subscribers they can expect to get from paid forwards. (I mentioned this last week).

• Maybe: include a little calculator along with that paid forwards data page, so you can plug in a max bid price and it'll estimate how many subscribers per week you'd get. It shouldn't be too hard to make this; I'm just skeptical that it would actually work very well since we don't have much scale yet and there's a lot of variance from week to week.

• Add a table to the stats page in the publisher console which will show forwards by day. Kind of like how you can currently see forwards by issue.

• Add an integration for Ghost to import 1-click subscribers automatically. Other integrations might follow—I'll check which newsletter platforms are used the most and prioritize those. Now that I've stopped doing 1-click subscribes for Substack newsletters, I've found I actually really enjoy not having to do them manually.

Let me know if you think of any other changes we can do to make your life more pleasant.

Speaking of which—For people on Substack, how are the please-export-your-subscribers alerts going? Not too annoying hopefully? I randomly had a short conversation a few days ago with someone at Substack, and I mentioned that it would be very nice for us if Substack would add an API for importing subscribers so that we wouldn't have to do it manually like this. She said she'd pass this along to the product team. If you'd like to join the petition, you can submit a request to Substack and ask them to do it.

---

Once that's wrapped up, I need to do one last batch of work to finish up a freelancing gig, and then there are a few different things I might work on.

To give some background: one of our main business metrics is the number of 1-click subscribes that happen each week. That number has floated around between 600 and 800 since last summer, i.e. instead of growing steadily, it's been a plateau. The reason is churn. People are most likely to hit subscribe-in-1-click right when they sign up for The Sample, and then they gradually become less engaged. Very few people continue doing 1-click subscribes after they've been on The Sample for 200 days (although some of them keep opening emails, as far as we can tell). So our organic growth basically just replaces the people who churn.

As I've mentioned before (a while ago), this isn't necessarily a show stopper. Some businesses, like mobile games especially, are naturally just high churn. For a while I've figured we'll just roll with it for now and try to get lots of new users all the time. Then when the business is financially stable we can decide if we want to keep scaling the same model or if we want to branch out to something new.

Our plan A for that is Facebook ads. We started scaling up our ad spend around the end of November. For most of December it was working: for example, between Dec 1 and Dec 20 we spent $628 on Facebook ads, and since then, the users we received have resulted in $719 in revenue. (Hooray!) However things went downhill around Christmas and still haven't fully recovered. Long story short, we think we have a good shot at fixing it, after which we could scale up our ad spend and then go frollicking through the meadows and stuff.

(Also, as always, my cofounder continues to work on the recommendation algorithm, which will improve our ability to recapture our ad spend from FB users.)

So we're definitely going to keep working on that. But if it doesn't work, we need to try something new, and this is where we'd be at a crossroads. One possibility is cross-promotion. If we can get lots of people to cross-promote with us, maybe that'd be a reliable acquisition strategy. In fact, the cross-promotion system is the whole reason I got the idea for The Sample! Last week I had an idea which I believe might do the trick. It's a cross-promotion directory with a few important twists. The idea is that it'd help lots of newsletter publishers do more cross-promotion with each other, and The Sample would simply be one of the participants.

Another possibility is that we could, you know, just try to fix our churn problem. Continuing to improve the recommendation algorithm should help with that, but the change will likely be quite gradual.

Someone suggested the other day that instead of forwarding people a single newsletter every day, we could send a weekly email with a list of newsletters. Each one would include a little snippet and a "read more" button, which would lead to a page on The Sample's website where you could read the full issue. I think that's worth trying. While pondering on that over the weekend, I also had a bit of an epiphany in which I thought of a way we could smoothly introduce some big, fundamental product changes that might result in a huge reduction in churn.

And I'll tell you what it is next week! This email is already long, and I'm exhausted. I'll also give more details about the cross-promotion directory thing. Maybe throw in a survey to see which one people think is more interesting. If we do the big-fundamental-changes thing, we'll do it via a beta program, so we can see if it actually does make the product more awesome before we roll it out to everyone. Then if it doesn't work we can nix it easily.