I released my first paperback book, Building Mobile Apps at Scale: 39 Engineering Challenges, a month and a half ago, which book is free until 31st May as a PDF.
I wrote this book with an unusual distribution model. I reached out to sponsors - vendors I mentioned in the book already - offering them extended coverage if they decided to support the book. I hypothesized that this model would be a win-win, as much of the book is about tooling and process challenges: and for tooling-related challenges, buying a vendor solution is often more pragmatic than building and maintaining it yourself.
Eight sponsors said yes: which meant the book was funded. I was not only able to dedicate working on the book full-time, but had budget for editing and marketing.
As an experiment, I decided to allocate $5,000 to sponsor well-known iOS and Android newsletters and websites and collect the results. How expensive is it to get a website visitor from these sites? How many free PDF downloads would this $5,000 result in? What newsletters and websites deliver the most traffic, the "cheapest" traffic, or the one with the highest conversion?
I ended up paying up to $40 per download for this free book, in the end, in a somewhat surprising set of results. But before we get there, why did I even bother advertising a book, where being free should have been enough advertising?
Why Spend Money on Advertising a Free Book
I did this experiment for a few reasons:
- I guaranteed a minimum number of downloads for the sponsors: 5,000, to be exact. Although I was hopeful that some newsletters would cover the book even without sponsorship, I did not want to leave this to chance.
- To give back to the prominent community members who write excellent newsletters and in-depth articles, week after week.
- To understand how expensive it is to reach mobile engineers. As I'm planning to launch a startup focused on platform engineering, one of my challenges will be to connect with software engineers. The fact that the developer community builder is the hottest new job in tech shows how it's becoming more and more difficult to reach software engineers, who are bombarded with more and more developer products.
- To evaluate if it would be worth advertising my next book, The Software Engineer's Guidebook, on email lists or via website sponsorships. That book will be traditional, with no free-to-download option.Because of this, I'd probably have to multiply the cost of a "download" by at least 10x.
Finally, I had more sponsorship revenue booked than I expected. I figured, what other time would I have the opportunity to run an experiment like this, advertising a quality book, and also directly supporting some of my favorite newsletters and sites?
Sponsored Newsletters and Websites
I booked the sponsor slots in January and February 2021 to go live in May 2021. I was surprised to learn how some of the popular newsletters, like iOS Dev Weekly had no availability for months - and since then, it has sold out for the year, with a $1,500/issue sponsor price tag.
I mostly sponsored newsletters and websites I personally read and like, as well as a few indie publications that I came across. I booked sponsorship on the following publications:
- iOS Dev Weekly: the #1 iOS Newsletter, going out to over 46,000 iOS engineers. $1,500/issue.
- Android Weekly: the #1 Android newsletter, going out to more than 66,000 Android folks. $750/issue.
- Swift by Sundell: one of the best-known Swift publications with in-depth articles. About 100,000/week in visitors, and $750/week (prices increased since)
- SwiftLee: an excellent Swift publication written by a fellow Amsterdamer, Antoine, who is a staff engineer at WeTransfer. Around 60,000 page views/week, and I got a discounted rate of $350 (prices have increased since).
- Code with Andrea: a well-known Flutter publication with around 7,500 visitors/week, and a cost of $60/week.
- Programming Digest and Tech Lead Digest: simple and to-the-point newsletters with 13,000 and 2,600 subscribers each, a cost of $250 and $100 per issue.
- Sarunw.com: another in-depth iOS publication I came across and liked the content I read. $200/week.
- A leadership newsletter and an indie mobile blog I read, at $600/week and $200/2 weeks.
I booked a total of $4,840 with these various sponsorships. I had to say no to a few offers - like a leadership newsletter I really like that has a rate of $7,500 for 4 issues. I also got inbound offers from iOS newsletters that claimed to have 50-100,000 visitors/month, but they refused to share analytics data, and my investigation showed they probably don't even have a few thousand per month.
The adverts were similar enough in their description, but I did tweak them to each site a little bit. Here's how a few of them looked:
From $1 to $40 per free PDF Download
I collected analytics results with Plausible - a privacy-focused and bootstrapped analytics solution that I use for my site as well.
It cost $1.23 on average to get a visitor to the landing page and $2.66 by the time someone downloaded a free PDF. It's hard to tell if these are good or bad results: however, I assume they must be very good, compared to e.g., trying to get people to click on a paid product or another offering that includes paying for it. I was, after all, offering months' worth of writing and years of experience building large apps, in a book form, for free.
On the above photo, I marked values that performed at or better than the average with green, and with yellow or red those who did above this average.
For some newsletters and websites, each free PDF download cost up to $40. This result was surprising. The indie mobile blog delivered very few clicks and below-average conversion. This was a blog where I did not get any traffic estimation, and I went ahead with the sponsorship as the blog looked good. For the tech leadership newsletter: it's one I really like reading. However, I almost always don't notice the advert, and the newsletter has over 10 articles per issue. In comparison, Tech Lead Digest - another leadership newsletter - has five links, which probably makes the sponsor stand out better.
The best price-for-value websites were:
- SwiftLee for the lowest cost per click
- Code with Andrea for in the lowest cost per download
- Swift by Sundell also delivered good results for the money - including a bonus shoutout article
Sarunw cost a bit above average both for visitors as well as for downloads.
The best price-for-value newsletters were:
- Android Weekly ($750): this was the only newsletter that cost "below average" in book downloads.
- iOS Dev Weekly resulted in more engagement, bringing almost double the downloads with a 50% smaller subscription base.
- Both Programming Digest and Tech Lead Digest did surprisingly well, considering the readers are not mostly iOS and Android engineers.
And the number of successful, free PDF downloads:
Reaching Mobile Engineers is Very Expensive
My biggest takeaway from this experiment is just how expensive it is to reach mobile engineers. $5,000 of spend came out to 4,000 visitors, of which 1,800 went through the trouble of entering an email to claim the PDF. If I was trying to sell a book or a service of value for mobile engineers, this $5,000 of spending would have meant far less than 1,800 customers: I'd guess 5-10%, or 90-180 people at most.
Knowing just how expensive it is to reach software engineers is an advantage both for technical bloggers and newsletter creators, and for startups. If you run an engineering blog or newsletter, you can probably make a nice side income by offering sponsorships for products that your readers might be interested in. If I would have been a company aiming to get 100 people to sign up for a trial for my software, I would probably need to get ready to spend in the realm of thousands of dollars through sponsorships or ads.
It's also a wild west when it comes to newsletter and website sponsorships, with little data points to start off with. There are no projections or refunds in this space: and the newsletters and sites tend to charge based on what has worked in the past. I shared the data back with all the publications, and several of them told me it was the first time they got information on how their performance compares to other sites. Several of them told me they were wary of raising prices, while some sites increased their pricing since the writing of this article. The only way to understand which sites work is to spend money on them.
For anyone building a software business, knowing that it's expensive to reach engineers is an opportunity to start doing this from day one. I'm seeing an increasing number of developer tooling startups hiring a developer advocate or developer community manager very early on. These are the teams that are very much aware that it's becoming more and more difficult to reach potential customers who are software engineers: the earlier you start, the better off you'll be.