The Pop-Up Paradox

A few days ago I read some of Jeremy's thoughts on how websites can slide from beautifully designed tools for content delivery to frustrating user experience nightmares. Intrusive cookie pop-ups, distracting ads, privacy-eroding trackers, irritating marketing overlays. That post hit a nerve – or possibly a whole bundle of nerves – and I've been wrestling with a response to it ever since[1].

To be clear, Jeremy has a point. An extremely valid point. There are many sites out there I've simply closed after a few seconds in either disgust or despair. I routinely land on news websites that take so long to load on my high-speed home WiFi that I just give up. Hell, I've become so accustomed to browsing a site through the partially-opaque grey overlay of a pop-up window as to barely notice it any more. I've actively changed how I use the web as a result of these irritations, switching to Firefox and using multiple plugins to automatically block some of these behaviours. I really, really wish the web wasn't like this.

But. I've spent the last two years effectively being the bridge between the two opposing points of view that I believe drive these web patterns. My job has been as both a Content Manager and a Frontend Developer for over a year now and that has given me some insights which I feel might be useful to rationalise (for myself if nothing else). In trying to write this response, I've come to realise that those insights have forced me into a kind of double-think around user experience and web performance. There's a lot I've loved about working such a hybrid role, but this subject of debate is not one of them.

Here's the thing. When I'm wearing my web developer hat I care about accessibility, performance, and the users above all else. That doesn't necessarily change when I'm thinking about content (it is, after all, still being made for users and must remain accessible), but it does get muddied with the additional requirements of marketing goals and sales targets. A balance is clearly needed and it's one I feel I've achieved in most of my work, but it requires a huge amount of ongoing mediation. In that sense, Jeremy's focus on communication channels is bang on the mark, but simple understanding each others point of view isn't always enough.

The reality is that a company website, in most instances, is a sales tool. It exists to drive lead generation and create new custom. That is not always the case – this blog, for example, doesn't operate for any commercial reason and many web apps are build on subscription or advertising models – but I'd wager that many of the design patterns on trial here are predominantly used within these kinds of commercially-minded sites.

Now just because a website is a sales or marketing tool does not mean it should ignore user experience; far from it, that must be considered a core part of any brand strategy. For my own part, I'm immensely proud of the amount of time and effort my team and colleagues have poured into making our sites informative, interesting, and genuinely useful. However, as with any tool, all decisions ultimately drive back to the core mission of the product and for these kinds of websites that central factor is sales – and any sales tool needs to prove ROI.

The result is that websites often have specific sales metrics to deliver: newsletter subscriptions, successful checkouts, promotional coupon usage, digital resource downloads. All of these can be done in a completely unobtrusive, accessible, and pleasant manner. But the horrible reality is that those terrible, annoying, irritating techniques... they work. It can be incredibly hard to argue that something is genuinely unethical whilst it continues to achieve positive results.

Now I need to pause here and be explicitly clear: just because something works, doesn't mean it should be done. Nor am I place the blame on users. Dark patterns are a real issue and – to be blunt about it – human psychology is easily manipulated. Abusing these knowingly is inexcusable and not something I am advocating for in any way. I'm also well aware that the "but it works" excuse is precisely that: an excuse. It's right up there with the "argument fatigue" reasoning; the thought process that you've made this point a thousand times and you really don't want to get into that whole thing again today, for whatever reason. I've succumbed to these excuses on occasion and I think part of why I've rewritten this post so many times is that I genuinely feel bad about having done so.

What I'm trying to say, instead, is that it's hard to argue with trend lines, particularly when people's livelihoods are on the line. If a company fails to see decent returns on marketing strategies or sales targets miss their mark, projects can get scrapped. Huge amounts of personal work can be lost. In the worst-case scenario, jobs can disappear. There's a very human (and humane) desire to protect yourself and those around you, so when it comes down to a question of protecting the user or protecting your friend, the user tends to lose.

Therein, ultimately, lies the rub. As users (and as ethical developers) we bemoan the pop-ups, modals, toasts, etc. But they work. That irritating newsletter subscription pop-up increases newsletter subscription rates. Those pesky data requests actively garner better insights and provide the underlying analysis for more successful marketing campaigns. Tracking cookies? They help determine where/when ad spend is necessary and (can) save companies a huge amount of money. If you're the one sitting in the corner arguing to not use them, the repercussions fall on your shoulders. I imagine most people would rather avoid that weight entirely.

Now I have heard a lot of people respond to that line of reasoning with suggestions to run A/B tests or perform user research. The logic goes that if data and statistics keep losing you the argument, then arm yourself with the some of your own. That's a nice line but it falls foul of reality. First of all, A/B tests aren't really viable for most companies, as you just don't get the traffic levels needed to derive meaningful conclusions. On top of which, both A/B testing and user research take time, money, and effort. If you're a small company and a stretched team, you don't have any of those commodities.

Another argument is to point out that these kinds of sales figures are scoped to the short-term and can directly result in long-term brand damage, but that's even harder to track and prove. We just don't have meaningful metrics for brand loyalty in most industries. If anything, all that does is cause a double-down from leadership in ensuring that the short-term goals are met, any means necessary. And that's before you consider that some of these patterns are unavoidable if you want to make use of other, major business platforms.

Without getting too bogged down in the weeds, big companies consistently strongarm smaller ones into using these techniques. As an example, my team needs to advertise on the LinkedIn platform and that means we have to agree to serving their tracking cookies. Our business' success is directly tied to our ability to access their users and that puts them at one hell of a negotiating advantage.

When I first read Jeremy's post, my take away conclusion was that he's right to call this a Sisyphean task. I'm generally optimistic that his views around having simple, clear, two-way conversations can improve this situation for all parties involved. In my experience, that works well and helps you find a middle ground. However – also in my experience – the reality is that a lot of what we hate about the modern web is here because it works, and it's really hard to argue with that. What developers really need is some kind of trump card, something that is unarguably bad for business

Well, in the time between reading and writing, Google have possibly handed us something along those lines. Their announcement that user experience may soon factor into search ranking could be a golden bullet that kills off a lot of these patterns. If I could categorically state that introducing a newsletter modal would negatively impact our Google traffic, I'd win every time. Now I may not be Google's biggest fanboy and they've made some pretty damaging moves to the web in the past, but they've also helped drive forward responsive design to become a de facto baseline thanks to mobile search prioritisation. A one-two punch combination of web performance and user experience metrics being openly pointed to as a reason why you're SEO is failing could be huge for settling this debate once and for all. Only time will tell, but personally I'm holding thumbs 🤞

Explore Other Articles


Trunks and Masters

There are several strong arguments for moving away from using "master branch" as default terminology, but what should it be replace with? Personally, I like the idea of extending the tree abstraction that we use when talking about branches, so have started using "trunk".


Jamstack Conf 2020

Notes from the 2020 Jamstack Conf. Some interesting dives in the Jamstack community and various applications of Jamstack technologies, with tweet threads as usual.

Further Reading & Sources


Want to take part?

Comments are powered by Webmentions; if you know what that means, do your thing 👍


  • <p>Marketing needs versus user experience is a topic that I have some deep misgivings over, but a recent post made me want to try and boil some of those thoughts down into their underlying rationale. I'm not sure I totally succeeded, but there we go.</p>
  • Murray Adcock.
Article permalink

Made By Me, But Made Possible By:


Build: Gatsby

Deployment: GitHub

Hosting: Netlify

Connect With Me:

Twitter Twitter

Instagram Instragram

500px 500px

GitHub GitHub

Keep Up To Date:

All Posts RSS feed.

Articles RSS feed.

Journal RSS feed.

Notes RSS feed.