Buttondown's engineers, writers, and biggest fans write about useful tips and tricks, new features, and more.

Behind the scenes

, we're updating our approach to open source funding.

 to the open source projects that power Buttondown. (We technically actually give 

 than this, at the moment, but as we start to hire a few more full-time folks this will be a useful ratchet to keep us pushing forward our commitments.)

You can see an itemized list of our commitments 

, which additionally lists all of the open source projects that we use and are 

Our logic for identifying which projects to fund was fairly anodyne:

Investing in the software that makes Buttondown possible

Our new Open Source Funding Pledge

Not only would Buttondown not exist without Python, but I — your humble founder — would likely not even be a programmer, having washed out of the industry after one too many invocations of 

 and honored to be a Participating Sponsor of 

, the largest and longest-running US-based Python conference — sponsored by the Python Software Foundation. This will be the first of many such commitments this year, in a way to pay back some of our own recent growth into the technologies that made it possible (all under the auspices of marketing!)

. We donate 10% of our profits to the open source technologies we use — a number that, while small, in 

 is still larger than most companies one hundred times our size — and aim to open source parts of our codebase that we think could be broadly useful to the community.

If there's one message I'd like to proselytize, it's that you don't need to be a behemoth to pledge financial support to great open source causes (though perhaps a close second might be "the behemoths should certainly do so too.")

Some great writing I recommend on the subject:

Investing in the Python community responsible for Buttondown's success

We're sponsoring PyCon

list of downstream services and apps we use

, and just published a big update to backdate some of the changes from the past few months. You can view the full list there, of course, but I thought it'd be fun to chat a bit more about the specific changes:

Some recent updates to our downstream services.

Changes to our stack in 2024

By now, you've likely received the email from Mailchimp informing you that 

 after over ten years of service. This is — there's no other way to put it — a huge bummer, and all too common in a tech landscape that seems increasingly interested in the "billions or bust" mentality.

Before building Buttondown, I was a happy and faithful TinyLetter user for many years. I loved its core mission — to be an app for people who weren't trying to make millions of dollars or spend all day customizing things — but wanted something that wasn't as buggy and had just a couple more quality-of-life features (image resizing, link checking, that sort of thing.)

We run Buttondown a little bit differently than other tech companies. Unlike Mailchimp, our goal is not to funnel every user into a complicated, expensive tool meant for S&P 500 companies; unlike some of our venture-capital-backed competitors, our goal is not to lose hundreds of thousands of dollars every month until we either pivot to a new strategy or shut down.

We're running Buttondown for the long haul: we provide a service, we charge a small fee for providing that service, and we remain profitable and independent so we can put your needs first for decades to come. This means we can spend all of our time improving and refining a simple, pleasant experience just for people like you. (That's why 

 trust us with their emails, from Pulitzer Prize winners to high-school poetry classes.)

If you're looking for a new tool to fulfill TinyLetter's old promise — 

 — you can give us a shot today. As a special offer in light of the news, I'm happy to migrate 

 for you: your subscribers, your old emails, the works. Just shoot me an email at 

read more about how TinyLetter and Buttondown compare.

Buttondown is free if you've got less than a hundred subscribers. (If you're, like, 

 subscribers — that's probably fine.) If you've got more than a hundred subscribers, you'll need to sign up for one of Buttondown's 

. The reason we charge for our services is the very reason TinyLetter is shutting down — it costs money to send emails.

Will subscribers have to re-confirm their subscription when I move them over to Buttondown?

Do you offer online archives, like TinyLetter?

Can you migrate my archives from TinyLetter?

Even then. (We just need the CSV export that TinyLetter gives you.)

Dear Tinyletter Users

This year — really great, we grew a lot, shipped a bunch of stuff, sent many emails, had little downtime.

Next year — localization, better API/programmatic access, prettier archives/subscription pages, better writing/sending experience, better in-app experience.

It's been a while since I've shared a "peek-behind-the-curtain" style of essay — the last one was technically 

, published in the first week of January. It was meant to be a year-in-review blog post, and of course it ended up being published a little too late for that.

. It is cute to re-read this: I was very confident that internationalization was going to ship (it still has not shipped), and I was already fed up with Heroku, which Buttondown uses for infrastructure (I would say my likelihood of churning off of Heroku in the next twelve months has 

Here's, in no apparent order, a bunch of stuff that happened in 2023:

One of the hardest things about running a business is managing your emotions: the thing that made this challenge much easier is hearing from so many of y'all that Buttondown has made your life — in some small, marginal way — better, or easier, or less frustrating. I can easily and confidently say that 2023 has been the most fulfilling work of my career.

Okay! Enough with the navelgazing! What's coming down the pike?

I've spent the past month chatting with users, sifting through the 

, and thinking hard about how best to spend the next year. Here's what I've got:

 is going to be added. Instead, we're doubling down on the parts of the authorial experience that can be improved:

That's a lot of "re"'s, I know. It's intentional! Buttondown has expanded a 

 in the past twelve months, but I'm very mindful that the majority of you trust Buttondown not because of the knobs and whistles but because it does the 

 very well. I think we can do that core work even 

, and that's where time and energy will be spent next year.

2023 (and what's next)

For this coming holiday spendtacular we're offering 

free for your first hundred subscribers and then from $9/month if you've got more than that.

Buttondown is already a pretty affordable product, and we want to earn your trust and money as a customer not through artificial scarcity or time-bomb sales but by making a high-quality product that is a good deal every month of the year.

 with BFCM deals, but if you're pulling the trigger on something that's "50% off" — make sure that the price hasn't been slowly ramped up over the past three months to make the "discount" look more attractive.

Black Friday / Cyber Monday prices

Everything we do comes down to three things:

There's no secret sauce — it's not rocket science. It's just diligence and good engineering. But for the details, keep reading!

Some email platforms treat double opt-in as a "table stakes" feature that authors can turn on and off as they wish, or even treat cold emailing (the practice of sending emails in bulk to addresses who haven't given you permission to do so) as legitimate.

Buttondown... does not. Every single subscriber on Buttondown 

 explicitly opt into your newsletter. This keeps the risk of spam complaints low, your engagement high, and readers 

That being said, we're nothing if not flexible and hospitable to other use cases. If you're confirming double opt-in outside of Buttondown (for instance, asking authors to confirm their interest as part of a registration or checkout flow) you can always 

Even if a subscriber to your newsletter is a real human who explicitly signed up to hear more from you, you might not want to keep them around if they haven't opened a single email from you in 

. This is referred to as a number of different things across the industry ("list cleaning", "email scrubbing", "email hygiene"...) — we call it 

. It's an automatic opt-in process that checks for folks who are suspected to have abandoned their email address or have never engaged with your content. Removing these folks from your list might feel a bit like ripping off a band-aid (it can be sad to see the big subscriber count go down!) but is great for the long-term health of your subscriber base, because it means your engagement rate amongst your 

(You can read more about subscriber clean-up 

Last but not least, every single new Buttondown newsletter who imports subscribers from an external source goes through a manual approval process to make sure everything looks kosher. This is old-school, we know, but it's for the safety of you 

 every other author on the platform. We take our sending reputation and deliverability seriously, and this helps us maintain our best-in-class benchmarks on both fronts.

(If you're worried that this process is Kafka-esque and will take days on end, worry not — we generally complete it in less than eight hours during the week and less than 24 hours during the weekend.)

Offboarding people who don't pass our bar

, but it's not perfect, and circumstances can change — maybe an author starts abusing their API privileges or decides to otherwise violate Buttondown's terms of service by sending emails to folks who haven't signed up for them.

If you're found to have breached our internal threshold for either 

the number of emails you send that bounce

the number of emails you send that are marked as spam

, we'll reach out to talk about your sending strategy. If those conversations don't go anywhere, you'll be removed from the platform.

(Sometimes, when we mention this, we're greeted with a bit of shock and dismay — "so you'll remove me just for sending more emails than usual?!" Buttondown's never off-boarded a well-meaning customer who made a mistake double-sending an email or accidentally imported bad subscribers; it's only ever been malicious actors.)

Before sending your email, Buttondown runs a battery of 

 pre-flight verifications to make sure everything's in ship-shape. We check everything you can think of, like:

This validation suite makes sure you're sending out the best possible email, and the best possible email gets delivered at the best possible rate.

Buttondown works very hard to make sure all of the subscribers joining your newsletter are 

, so that they're not giving your publication a bad reputation. One of the first and most effective lines of defense is checking incoming subscriptions' IP addresses against a combination of both our internal datasets and those of trusted vendors like 

. (Note that we do this by pulling in omnibus data from external sources, rather than polling them, so as to protect the privacy of you and your subscribers.) This lets us immediately filter out automated or malicious bots subscribing to your newsletter.

One of the bigger product decisions Buttondown has made is to not push cross-promotion — this is the fancy industry term for "when someone subscribes to your newsletter, we suggest three similar newsletters for them to also subscribe to."

There are a number of reasons we do this — mostly, we don't think it's our place to get in between you and your audience — but the biggest one is 

 Studies that Buttondown's run show that subscribers who sign up for other newsletters through cross-promotion end up reporting emails as spam and unsubscribing at a 

 higher rate — even for the newsletter that they originally signed up for!

Other platforms taut growth through cross-promotion to juice your subscriber base; while you're of course free to run your own manual promotional efforts, it's not worth the risk to deliverability for us to push them on your behalf.

 to know what these clumsy-sounding acronyms are, but they're protocols that we (along with many other reputable senders) use to ensure that the emails we send are coming from 

Setting up SPF and DKIM are a bit of a hassle to handle yourself, but if you're sending from Buttondown it's automatically handled for you (even if you're sending from a custom domain!) And starting in 2024, Google 

 to send their emails with these authentication protocols in place.

We keep it easy for subscribers to unsubscribe

Yeah — that drives us crazy too. Every email coming from Buttondown is one-click unsubscribe compliant, meaning that there's no labyrinthine set of hoops to jump through to unsubscribe. You might not love this as an author, but it's good for you in the long run — unhappy subscribers turn into frustrated subscribers, frustrated subscribers report your newsletter as spammy, and spam reports tank deliverability.

Buttondown's default email templates are very plain-text forward. This is nice for aesthetic reasons, but also is 

 positive on deliverability. Text-heavy, less rococo emails (as compared to their image-heavy counterparts) work great on inboxes of all sizes and shapes and score extremely well in content scanning, meaning that you'll be reaching more of your subscribers more reliably.

Last but not least, Buttondown lets you send from own domain (if you own one!) This improves the overall engagement rate for your emails, as they appear more personal and have a higher domain reputation. More importantly, that domain reputation carries with you — even if you leave Buttondown for another service, so long as that service also allows you to send from a custom domain.

What we do to make sure your emails make it to your readers' inbox.

How we keep deliverability high

By now, you're probably aware that in three weeks' time, Mailchimp will be automatically using your data — your customers, your content, your 

 — to power their generative AI products:

We may use your Inputs and Outputs, including Customer Data, for machine learning purposes in order to develop and improve the AI Model, the Services, and similar products and features, and you instruct us to process Customer Data for such purposes.

(There's no opt-out, of course; the only opt-out is to stop using Mailchimp, and to delete your data.)

I'm no Luddite: I think whether you want to call it an "LLM" or "generative AI" or "GPT", the products and technologies that seem to be taking over the world are legitimately useful and cool.

But "useful" and "cool" do not outweigh privacy concerns. Nor do they outweigh your right as a user to decide what companies you trust get to do, not just with 

Buttondown takes that trust seriously, and considers it a sacred responsibility to be a good steward of your data.

That's why we're the only newsletter tool that offers 

. We don't gather any data unless you explicitly ask us to, and we certainly won't be feeding it into any other newsletters or black-box AIs without you knowing it.

If you're a happy Mailchimp user, no hard feelings! But if you've been frustrated by some of their changes over the years — the price increases, the complicated UI, the sale to Intuit — and this is the straw that broke the camel's back, 

. I'd be happy to personally onboard you onto a platform that respects your privacy.

Isn't it time for software that respects you?

Dear Mailchimp Users

Earlier this year, I came across an article entitled 

Sniper Links: How To Increase Your Email Confirmation Rates

 from the folks over at growth.design. It sounds a little clickbait-y, but it's both benign and brilliant: when asking folks to confirm their email, try and figure out what email provider they're using (gmail, outlook, and so on) and give them a magic link that opens up the search functionality in those apps.

A few months ago, we shipped this functionality as a little test to see if it was in fact useful.

It was! We saw statistically significant uplift — around 6% — in double opt-in for folks shown the sniper link.

We were so happy with this that we decided to release it as a free tool: 

 (I was a sucker for the clever domain name). It's totally free.

You can use it either as an API call to get a magic URL:

fetch(`https://sniperl.ink/v1/render?recipient=${
    recipient
  }&sender=justin@buttondown.email`)

<sniper-link
  recipient="you@example.com"
  from="justin@buttondown.email"
/>
<script src="https://sniperl.ink/v1/sniper-link.js" defer></script>

We support Gmail, Yahoo, Proton, iCloud, Outlook, HEY, AOL, and Mail.ru, which counts for over 85% of the traffic we've seen thus far.

That's it! I hope you like it, and shoot us an 

 if you have any questions or feature requests.

Our free tool can help you get more subscribers. Did we mention that it's free?

sniperl.ink, our free tool for improving subscription rate

How Buttondown uses your content to power generative AI

 to announce that Buttondown has pledged $5,000 as an 

, a not for profit organization that provides internships in the open source world for engineers impacted by systemic bias.

Outreachy's past work is deeply impressive; the accomplishments that their interns have gone on to achieve in the industry is a track record that speaks for itself. I am honored to be able to, however fractionally, help folks fall in love with the kind of work that has shaped my life for the past decade.

This will be the first of many years of contributing to an initiative that I think does 

 long-term good in building the kind of progressive, equitable future that our industry deserves.

We're excited to support the open source and science communities.

Buttondown pledges $5,000 to Outreachy

After six years of not touching the registration page, I added "Login with Google" and "Login with Apple". (You can see exactly what that looks like 

Here are the results, after a week of testing:

The entire project took me around five hours, start to finish — I used 

, as the prospect of relying on a third-party vendor for registration seems fairly terrifying.

Probably the single highest-ROI thing I've done this year (and I've done a lot of high-ROI work this year!) Couldn't recommend it enough.

We added new log in options, and we can't recommend it enough.

You should add SSO

You are perhaps familiar with the concept of 

: writing your Readme first and using it to backtest the interface for your CLI.

I'd like to propose a similar mental exercise for thinking about your pricing model: 

. As in: when thinking about your monetization strategy, write a pricing page first and work your way backwards into something that makes sense.

Buttondown is API-first software for sending emails and managing subscriptions. (Think Mailchimp, except for people who don't want to spend all day inside Mailchimp.) Sending emails cost money; running servers costs money; paying a mortgage costs money. So I charge for the service, albeit with a free tier.

? More than anything else, you want to think about two numbers:

When we launched in 2018, Buttondown had a hybrid model: you pay ~at-cost for the amount of emails you send, and then you also pay an additional $29/mo for additional functionality such as custom domains or whitelabeling. I considered this a very clever approach to align incentives — I "give away" the rails for free and then charge for the fancy stuff!

. The vast majority of new customers would have to begin their journey by emailing me some variation of "wait, so how much do I need to pay?" — rarely a good sign.

Last year, I did what many founders do — admit defeat and swear newfound fealty to the almighty three-column SaaS grid. 

 goes the clever combination of usage-and-feature-based pricing, 

 comes the cascading waterfall of checkboxes and the various price tiers:

 cliff (you have too many subscribers, time to upgrade!) and a 

 cliff (you want to use Zapier, time to upgrade!) proved efficacious: ARPU and expansion revenue steadily rose as new customers coalesced around the $29/mo plan.

 of the new pricing model were solid, I found myself running into two core issues:

Once again, the pricing page presaged the outcome: 

 folks weren't voluntarily expanding because they had no reason to! 

 enterprise customers were harder to convert because the messaging for them was terrible!

Expanding Buttondown's pricing model (2023)

Earlier this year, I shipped two changes to address both problems:

First, I added and shift functionality to incentivize moving towards Professional — gating whitelabeling behind a higher tier and adding support for teams + dedicated IPs, two long-standing asks:

Second, I added obvious enterprise pricing based on the same value metric that the existing prices were on:

(I dithered between having the public-facing enterprise pricing be based on "emails sent", which is more common in the industry, vs. "number of subscribers" — ultimately landing on the latter because it feels more 

 — for instance, now I need to think about levers to incentivize upselling into the $139/mo plan — but it's much better than it was 24 months ago. (Much better being defined as ARPU is up 

An underrated part of pricing is "does this make sense to users?" Users think about your pricing primarily through your pricing page (and secondarily through your competitor's pricing page.) If rationale or price can't be concisely explained on a page devoted to the subject, then something is probably wrong.

Thoughts on how to use the pricing page to mold and refine your pricing position

Pricing page driven development

I am a sucker for a good list of running costs, dating back to seeing 

, back in 2015. I think it lends an earnest and unglamorous look as to what it 

 takes to a business in the way that "hustle porn"-type essays rarely do.

I've done a poor job of keeping my own such running cost document (though I'd like to do a better job!), but in lieu of a beautiful chart like Cushion's I'll try and offer something sloppier: a screenshot of my finances table!

(As context, Buttondown posts around low five figures of monthly revenue, so this is a pretty healthy margin.)

Beyond that, nothing much to report. There's a slew of things that cost somewhere between $5—$30/month for which I use the following heuristic: does this provide at least fifteen minutes' worth of value? If so, great. That's perhaps a low bar, but it's certainly a higher one than at big technical companies (where the bar might be "has someone used this tool in the past year? okay, we won't churn from our four-digit plan".)

A rundown of Buttondown's software costs for the year.

What it costs to run a software business in 2023

I took a break from the mistletoe chaos of the dead zone between Christmas and New Year's to do two things, both portrayed below:

Thank you, Haley, for cutting up all of these into tiny slivers for me.

I might not work at a big company any more, but I can't stop myself from over-stylizing my Google Sheets.

It is...hard for me to take planning processes too seriously. Sometimes it feels sacrilegious to worry too much about event sequencing when I know I have more than enough high-urgency, high-importance items to tackle for the foreseeable future. As my old colleague 

There’s obvious value in knowing what you want to be doing in a year. Having a vision is important. Laying out all of the individual steps to get there isn’t valuable or important. Besides the planning process for an extended roadmap being tedious and slow, it forces teams to sign up for a plan that is disrupted by any participant in the plan facing disruptions.

(By the way — you should subscribe to his newsletter. I do not think I am exaggerating when I say he was the single most interesting technical writer within the company and I am 

 that he's taken his talents to the public domain, so to speak.)

 thing is good. I think I write every two months about struggling to translate tactics into strategy; this is one of those times where I am trying to be a bit more conscious about having a horizon larger than a month and an appetite larger than a pull request.

All of which is to say: the vision I'm carrying forward into 2023 is (and will be enshrined into some somewhat hackneyed principles, I'm sure):

 these as KPIs; I'm not sure that's useful or interesting. But they will be a good compass, and a good forcing function to push me to say no to work that doesn't move any of these hypothetical needles.

We're excited for what's in store for 2023!

Q1 planning

First off — I've read a countless number of these posts before and they're all couched in evasive, annoying language. It's very annoying! So let me get a few things out of the way:

Okay! That's the spark notes version. Now let's get on to the meat of the matter, which I'll express — out of equal parts rhetorical laziness and a distaste for hackneyed marketing speak — in Q&A form.

Okay, so, what's the new pricing structure?

If you've paid for software on the internet before, it might look familiar: this is what we technologists call a "SaaS pricing grid."

Buttondown offers four tiers, based on the features you're looking for and the number of active subscribers you have:

Uh, what was the old pricing structure again?

What does this mean for my paid subscriptions?

Before the pricing change, Buttondown took 

 percent of your paid subscriptions revenue. After the pricing change.... Buttondown takes 

All about the new pricing plans, and I promise to try and take it easy on the marketing speak.