Zero to $50000 in six months: growing Ritza, a technical publishing company, as a productized service
For more background, you can find previous retrospectives for 2019, Q1 2020, April 2020, and May 2020.
In July, I founded Ritza - a technical publishing company that offers (very) technical content marketing, developer advocacy as a service, and a bunch of related publishing services. What does that mean exactly? We publish ebooks, documentation, and blog articles, usually with the goal of helping developers or technical managers in some way. For example, we did https://codewithrepl.it as a companion website for Repl.it. We offer this content on a subscription basis.
The title is clickbait: I’ve been working on Ritza for quite a bit longer than 6 months, but it only became ‘real’ in July this year as a registered business which I was devoting 100% of my time to, and – from what I’ve seen in Maker circles – writing about how you got your initial revenue is the best way to 2x your revenue, so the title is also an experiment in some ways.
I’ve gained a lot by reading about other people’s transparent revenues. That said, talking about money is still weird for me, and I think it’s harder to be fully transparent when offering services instead of a SaaS - there is more customized pricing both in charging for work and paying contractors and talking money is a sure way to get emotions up and make people feel like they got a bad deal.
Here are some high level figures for the last 6 months to get it out of the way.
Revenue
Ritza made between $6000 and $14000 revenue each month between July and December, with a total revenue of around $58000 for the last two quarters of 2020.
This was below my target of hitting $20000/month in 2020, but – especially given circumstances – this was way better than my worst case predictions. There was no clear overall trend, but a definite drop off in November and December, with July and October being our best performing months.
All of our contracts are month-to-month and one client went quiet in October. While most of our revenue comes from recurring contracts, we also did some one-off work in July through October, but none in November or December.
Nearly all of our costs are from paying writers, editors, and designers. We additionally spent money on office space that we don’t use (over $500/month for coworking space that is tied to my ‘visa facilitation’, which is a requirement for my visa in the Netherlands), and accounting costs. We pay a few dollars a month for various online services, such as GSuite and various domains, and we have credits on DigitalOcean and AWS which take care of all of our hosting for now. We also recently picked up a subscription to SemRush - a tool that I think is overpriced and slightly shady, but which has proven its use in better understanding what people are searching for.
Overall, this revenue allows me to pay myself a ‘salary’ that pays rent in Europe, with more cash than my previous job as Head of Technology for a South African EdTech startup.
That said, my plan B after quitting my previous role was to find a full time job remotely or in Europe. I shopped around a bit and did some interviews for roles that were offering $150k-$200k/year + benefits so in terms of opportunity cost I am still ‘losing’ for now and it remains to be seen how far Ritza can scale with its current business model.
Experiments and content
Our bread-and-butter service is producing technical content: writing tutorials and blog content. While our mission is to unlock marketing budgets “for good” by making this content available for free, we have also helped companies produce and improve internal proprietary content that they then sell.
On the side, I have been playing around with the idea of building online tools to help with producing and publishing content. None of these got the attention they needed to get off the ground, but they remain on the back burner as ideas I’d still like to work on. They included
-
‘vs’ Graphs - a tool inspired by this article. I built this because I needed it - especially for writing articles about ‘hot’ spaces like MLOps and DevOps, it’s easy to get overwhelmed by the sheer number of tools, products, and services. This visualisation makes it easy to explore a new area and find the most popular products, as well as to find out how a specific product ‘fits in’ (“Oh X plays in the same space as SageMaker”). This is also useful to write ‘x vs y vs z’ articles, which turn out to be very low hanging fruit in terms of ranking well on Google. If you’ve ever tried to find out how two products or services compare and found yourself frustrated at all the low quality ‘alternativeTo-like’ pages you’ll know why these articles are useful, and I ended up also building a related tool to outline these articles automatically. The fun part about this was that I built it in a single train ride using only my iPad and repl.it, which felt like a big step in the direction of being able to code as a nomad.
-
Rate my Copy - Also while researching products and services, I got annoyed at how cliched most landing page copy is, and even more annoyed by how uninformative it is. I scraped over 20000 landing pages for online products and services and built a database of the most commonly used cliches.
-
Opinionated tutorial Publisher - while most of our clients have their own systems already in place to host the content we produce, I was frustrated at how hard it was to generate a lightweight, good looking page with writing, code samples, and images. There’s a long way to go on the design still, but some of the other pieces are in place.
I won’t link to all the content we produced in 2020, but some highlights are
-
CodeWithRepl.it - a set of tutorials, also in book form. This was very well received on reddit and continues to rank well for a variety of search terms. I am strongly against most kinds of tracking, but caved at the end of the year and installed Plausible Analytics on it, which also makes it easy to share publicly everything we track: https://plausible.io/codewithrepl.it. Traffic has died off a lot at the end of the year (based on some server logs analytics I did), but rankings continue to improve and I think the content is in pretty good shape now (we’ve done several sets of updates since initially publishing it).
-
DataRevenue Blog - This is the project I’ve personally learned the most from this year, writing about everything from general machine learning through BioTech (I didn’t even know what Metabolomics was in 2019, so there was a steep learning curve to write some of the articles about that).
Growing a team
While we initially mainly worked ad-hoc with contractors on various projects, towards the end of the year Ritza started feeling like a real company with a core team. Eugene Dorfling joined in October for a full time internship and is continuing in 2021 as an Associate Developer Advocate. Several other freelancers have regularly worked on projects throughout the year. I’m still trying to figure out if the next full time hire should be a senior writer (to take over some of the writing I am doing), or a managing editor (to help out with client feedback and all the nitty-gritty but super important aspects of getting from ‘first draft’ to ‘production’), but I’m leaning more towards the latter.
What IS Ritza?
Friends and family still don’t really understand what Ritza is. I get a lot of “uh, so you make IT manuals, right? Isn’t that just telling people to turn stuff off and on again until it works?”
Ritza is probably closest to a ‘content marketing agency’ at this stage. But I really dislike most content marketing agencies. Not to name and shame, but there’s definitely decent demand for very low cost content such as https://contentfly.com/blog/tag/content-sample/, and I honestly can’t see how that adds any value to the world at all, so I want to avoid being pulled in that direction or associated with content like that.
My initial response to this was to try to move towards “Developer Advocacy as a Service”, but it doesn’t really roll off the tongue, and I’m not sure that “Developer Advocacy” is actually a job that will survive the Tech Bubble I’m convinced we’re in.
Towards the end of the year, it became clear that Ritza is simply a publishing company (or at least the very beginnings of one). We are slowly becoming experts in the entire publishing pipeline, from sourcing high quality writing, through editing, designing, publishing, and distributing. While we have started out with a strong focus on technical content, there’s no reason that that has to remain a core focus forever. I’ve had pretty mediocre experiences with technical publishing companies like PacktPub and I believe non-tech publishers that have existed for decades or centuries are even more in need of a bit of modernization.
I previously helped my Dad publish his first book of (hilarious) memoirs as a Doctor in Africa and during the slower period between Christmas and New Year now I spent some time tweaking the landing page, making a sample available, and doing some marketing. So far, all I’ve done is burn $50 on Reddit ads that led to zero sales, but I think with some experimentation with different distribution channels there’s more potential. Stay tuned.