← Home

Eight questions to ask before choosing a web publishing platform

Anyone keen to write on the web faces a bewildering array of choices: They could use Medium or Wordpress, post directly on LinkedIn or Facebook, or use Squarespace or Wix to help them build a website. Or, if they want to dig deeper, they could explore this list of web content management systems.

When I started this site, I decided to build www.robinkwong.com using a heavily-modified Jekyll template, and to host it on Github pages. I also chose not to use any audience analytics or tracking cookies.

I arrived at these decisions through asking myself a series of questions that are important not just for neophyte bloggers, but for any content producer having to deal with increasingly powerful web platforms in an age of distributed content.

What’s the site for?

Why does your site exist? Knowing the answer to that question makes the subsequent questions a lot easier to answer.

For me, this blog is a space for rumination. It is my calling card; an extended curriculum vitae. It tells the world what I stand for and what interests me professionally, and allows me to use as much (or as little) space as I need to explain my thoughts and ideas. It is not a way for me to directly generate income.

It is also a bridge connecting me back to the skills I learnt while on the FT’s interactive desk. It is where:

How much does getting a big readership matter?

So much online content production is supported by advertising that audience size has become a default measure of success. Platforms like Medium or Facebook either use their ability to control distribution as a selling point, and as a lever of control. But before you start chasing pageviews, consider why it matters to you — I suspect most people write for reasons that have little to do with audience size.

Like most people, I would of course prefer to have more readers rather than fewer. But because some of the reasons why I started this blog could be met even if I have no readers at all, ‘reach’ is not so important for me.

I am also aware that my writing addresses an audience with very particular interests: digital journalism/news nerds, and those interested organisational change. I therefore much more about specifically reaching those people, rather than getting a bigger overall audience.

If you collect analytics data, what will you do with it?

Since the internet makes it so easy to collect audience data, it is now nearly unthinkable for websites or blogs to not have any analytics built into them. But in order for the data to be useful, you need to know both the questions it can answer, and be able to act on those answers. Furthermore, most people under-appreciate the drawbacks of a ‘collect it because you can’ approach.

As Peter Drucker said:

What gets measured gets managed, even when it’s pointless to measure and manage it, and even if it harms the purpose of the organisation to do so.

This is why I don’t use any tracking or collect analytics on my site. I believe the benefits of collecting analytics data is minimal, while the costs are high.

I don’t expect to gain a lot by looking at audience analytics for several reasons:

Meanwhile, my time is limited. I can either spend it on writing, or on analysing the audience data on this site. I also know that, if I collect the data, it would be impossible for me not to seek out meaningless patterns in the data, or to chase that hit of dopamine that comes from seeing the pageview counter tick upwards. That makes it harder for me to focus on writing well.

What investment are you prepared to make?

This question can be split into a) the initial, one-off investment in setting up the site, and b) the ongoing investment of adding new content and updating the site.

The value proposition of most blogging platforms and services is that they make it really easy for you to just start posting, without having to think very hard about why you are doing it and what that implies in terms of making your own design choices. But if you are going to make writing a habit, then your ongoing investment will likely far outstrip whatever time or money you put into the initial set-up.

I was prepared to put in a significant initial effort to explore my options, determine my best choice, formulate a plan, and to set up the site. But after that, I wanted posting new content to be easy and maintenance to be very low, because I wanted the bulk of my ongoing time investment to be spent on writing.

How much does freedom of choice and customisation matter?

All publishing platforms make a trade-off between ease of use and freedom in presentation choice: Twitter and its 140-character limit is at one end of the spectrum, while Wordpress’s multiplicity of templates and plug-ins are at the other.

Part of the service that most publishing platforms provide is templating: It makes design choices for you so you could just focus on the writing. There is nothing wrong with this, excpet when the same set of design choices are made regardless of the content: The Awl, The Ringer, and the Pacific Standard are three very different publications, yet their articles look like the same thing: A piece on Medium.

This is the core of my philosophical objections to Medium: That it is a homogenising force in web publishing. In this, I am very heavily influenced by Matthew Butterick, who lays out the full argument in The Billionaire’s Typewriter

The freedome to make my own design choices matter a lot to me. This site is as much about expressing my self identity, as it is about conveying information to readers.

How much does ownership and control over the material matter?

Choice of presentation is just one form of control you give up when you write on a platform. In order for the platform to optimise distribution it must also restrict your control over how your writing is distributed.

Facebook, in particular, optimises distribution based on criteria it deems important to viewers, rather than what’s important to the writer. This may be good systems design for a personalised social media network, but it could have political ramifications. It certainly has journalistic implications.

I noted early on that I do not intend to generate direct income from this site. Yet that doesn’t mean that it has no value and I don’t care about my ownership of it. On the contrary, because I’m spending time and using my expertise and experience to write these posts, my ownership and ability to exercise control over them matter a lot.

This site is a tangible expression of my intangible assets (skills and knowledge I spent time and effort acquiring). I may be giving them to the world for free, but it is important that I can freely choose to give them away.

Do you need a single service for both publication and distribution, or can you find separate solutions?

When Wordpress was first released 13 years ago, web publishing platforms were just that: publishing platforms. Now, most web platforms — Facebook, Medium, LinkedIn, etc — bundle at least three distinct functions together: hosting, presentation and distribution.

There are good reasons to use a one-stop shop. After all, those functions matter only insofar as they contribute to the final product, and there are benefits to be had by tackling the problem as a whole.

But it’s also worth asking whether you can do better by tackling the problem separately, and what you are giving up by diving head-first into someone else’s ecosystem.

I use Github Pages to host, and Jekyll to present my posts. I only need ‘good enough’ hosting, and those two services were designed to work well together. I don’t expect, however, that either will help me distribute my work. Instead, I use a mix of other methods to help me spread my ideas:

Is there any reason why your blog posts should have bigger file sizes than major works of Russian literature?

This rhetorical question comes from a talk by Maciej Cegłowski about the growing website obesity crisis, particularly among sites that primarily serve text. Weighed down by large, unnecessary ‘hero’ images and adtech javascripts, many web articles are several megabytes in size — while the text of, say, Dostoyevsky’s Crime and Punishment is only 532 kilobytes long.

When you write on most web publishing platforms, you have no control over what else they layer on top your text. This one-sentence ‘essay’ that Maciej posted onto Medium to prove his point is 2.4 megabyte in size.

I found it really useful to ask myself these questions from the outset for several reasons. They clarified my thinking and forced me to be very specific about what I was making this site for. This, in turn, made it easy to prioritise and set boundaries on what I would and would not do.

Above all, it helped me become conscious of what my default choices would be. It’s not about being dogmatic with them, but default choices and settings are powerful: I want to make writing on here a habit and part of what means is that I can just do it without having to think through everything from first prinicples every time.

What are your default choices?

What to read next: Wicked problems in the business of journalism