Growing the support function
(If you haven't yet: read my 2025 blog post!)
One of the tricky contradictions in growing a support function is that even when things are going well, it can feel Sisyphean. To wit: we grew our customer base by north of 60% last year, and our support burden—as measured by overall incoming ticket volume—grew by 30%. This is objectively a victory — it means our ticket-per-user metric went down — but it still feels like the water level is rising.
Intellectualizing doesn't help the fact that waking up every day and always seeing 30 items in the queue takes a toll on you. Earnestly, there were parts of this past year where I felt the same kind of support burnout that I did when I was running support entirely by myself, even though the material circumstances are way different and way better. It can be exhausting to have a part of your brain that knows at any given point in time there's a ticket you can answer.
There's a surprising lack of literature around scaling support functions. I suspect this is largely because companies don't really think about it or try that hard. Even companies with otherwise sterling reputations—like Stripe—have objectively bad support by almost any metric you can throw at it, including the qualitative ones.
Swim lanes
But there are obvious throughlines and things we can do. The first is that we're moving out of a free-for-all style system into dedicated swim lanes organized by module. There will be a couple shared modules for things that everyone has (or needs to have) a common understanding on. But for the rest, you can merely assign a ticket to the right label, and then it'll get routed to the subject matter expert.
The goal here is to be able to close the inbox and shift into working on proactive things—such as improving documentation or fixing paper cuts—without feeling like you're leaving food on your plate.
The most interesting part of this exercise was bikeshedding the list of modules. In general, Buttondown is too small of a place to think about modularity or swim lanes too seriously. To the extent that it's happened at all, it's mostly happened organically in the realm of Linear tickets and people who just so happen to spend an inordinate amount of time in a given part of the codebase, such as Mary with the design system and the author-facing UI.
The challenge of dividing all of our surface areas and touch points into a discrete list was a fun one. Here's where we landed:
- Subscribers
- Archives
- Paid subscriptions
- Outbound emails
- Developer tools
- Author-facing app
- Onboarding
- Automations
- Top of funnel (doc site, demo site, growth, marketing site, etc.)
You could credibly criticize that list for being too focused on features and code-based implementation rather than jobs-to-be-done or actual verticals. A different version of this list might have looked simply like: monetization, subscriber growth, attach, etc. The problem with going purely vertical is that you introduce a lot of cross-cutting concerns.
SLAs and expectations
We're also going to be focusing on SLAs for paid versus free versus enterprise users, and rigorously communicating and context-setting these expectations. This goes both ways. It means neglected tickets will get escalated and our P99 will go down, but we're also giving ourselves license to not be in full reactive mode if we happen to see a ticket come in for a prospect or new user. It also gives us the grace to not have to micro-decide between responding capably or quickly: I in particular am guilty nacking a response because I want to just fix the thing.
Why we take this stuff so seriously
I've said this before and I'll say it again: people choose Buttondown because they hear good things about our support. People often leave Buttondown if their support experience is bad. And most of all, people stay with Buttondown because the support experience is so good.
When I sent out our annual letter, I got 63 replies—which is just kind of awesome. And this is not an exaggeration: I went through email by email, and 47 of those 63 replies explicitly and specifically called out how much they loved our support, why it was the reason they were sticking with us, and why they were so happy to recommend us to their friends.
We take this stuff seriously because it's our competitive advantage. And we want to keep it that way.
