The WSJ published an article in which a research outfit called eMarketer claimed that Google was projected to drop below 50% of search traffic by 2024.

Perhaps because of the WSJ paywall, many people seem not to have read the article itself. I admit that I too did not read it as soon as it came out, assuming that the attention-grabbing headline was clickbait that would turn out to be dramatically qualified somehow.

Sure enough, in order to achieve their eye-catching conclusion, eMarketer had to resort to some tricks; one might even call them shenanigans. The most natural assumption would be to compare Google to its closest direct competitors, like Microsoft’s Bing. Unsurprisingly, nobody is projecting that Bing will overtake Google, nor even that Google would drop behind all other web search engines together. Instead, this dramatic reduction in Google’s share of search is claimed to be due to the rise of vertical search services, notably a combination of “new developments in artificial intelligence and social media search”. The specific examples they cite are TikTok and Perplexity, although the SEO-friendliness of those two does pre-warm my BS alert.

Horizontal versus vertical search

Less SEO-friendly but to my eye more significant competitors to Google’s dominance are Amazon’s internal product search or Apple’s App Store search. The inclusion of these services is less preposterous than it might seem at first blush. After all, these days both Apple and Amazon sell ads within their search functions, making them into something different from a simple in-product search box — different and worse.

I am instinctively in favour of specialist vertical search over generalist search; I am a diehard user of Kayak over Google’s built-in flight search, for instance. The argument in favour of a generalist search box for everything is precisely that it can find anything: users can go to Google and search for a flight, a book to read on the plane, and a gift for the person they are travelling to see. And of course Google is not providing all of these capabilities out of the goodness of its corporate heart; each new search domain is providing rich facets of data on the user who made that search, and that user profile can then be marketed to advertisers trying to reach people with those particular characteristics and assumed level of disposable income.

The argument in favour of vertical search is that specialists can provide novel and differentiated capabilities, whether that is specific domain knowledge, or something as simple as a native app that can do things like hold my travel search context for longer than a browser tab does. This positive argument in favour of vertical specialists quickly becomes a negative argument against the inclusion of competing features in generalist systems, because functionality built into a tool like Google that is already a popular starting point will naturally draw off a certain proportion of the potential user base, who will never go beyond the confines of the Google search box.

Why now?

This is why the timing of the publication of this article is so interesting. Google is in court right now, defending itself against accusations of monopolistic behaviour. Specifically, the prosecution found fault with the payment of traffic-acquisition fees to the likes of Apple and Mozilla. The early judgment, which is certain to be appealed ad nauseam, suggested “structural remedies”, a somewhat vague phrase which is generally understood to be a recommendation to break Google up into smaller constituent parts. The idea is that each of those “baby Googles” would then have to compete on a more level playing field with other vertical specialists in each domain, without the bulk of the Google parent company behind it.

I am somewhat sceptical of how that breakup might work, and even more so that it will take effect in anything like useful time. By the time the EU ruled that Google’s Shopping aggregator service represented unfair competition, most of its would-be vertical competitors had gone out of business.

It is however very convenient — a cynic might even say, suspiciously so — for an article to come out at this specific time, describing Google as a beleaguered competitor in a crowded field rather than an all-conquering dominant monopolist.

Google search

Who stands to win?

I also suspect that this is yet another instance of “heads they win, tails we lose”. The vertical search examples cited in the WSJ article are terrible, enshittified user experiences that force users to fight the algorithms and advertising markets distorting the search results to find what they are actually looking for. Far from specialist experiences that are willingly chosen because of the better quality of results that they provide, these are walled gardens operated by exploitative landlords for a captive audience that has no real choice.

For a while it is still possible to get what you want by pitting one corporate feudal lord against another — adding “:reddit” to Google search queries to get higher-quality results audited by human Redditors, or using Google to search Amazon results to try to hone in on the specific thing you are looking for, rather than a myriad rip-offs and sound-alikes. But there is no question that the operators of the walled gardens would very much like to build the walls higher to block these interlopers, giving them free rein to shape their users’ experience — and not in the users’ favour.

What is to be done?

I don’t have a structural solution; all I can do on an individual basis is try to construct and maintain my own bubble of services I have selected for myself. I have no illusion that DuckDuckGo will ever take a meaningful share of search from Google, let alone overtake it. For one thing, the system underpinning DuckDuckGo’s results is actually Bing, and Microsoft is only able to present itself as the nice guy in this story because of its failure to do exactly what Google has done. If Bill Gates had had his way, we would all be using Internet Explorer, Hotmail, and MSN, instead of Chrome, Gmail, and Google — and I doubt we users would be any better off.

All I can do in the meantime is cheer on any antitrust or anti-monopoly regulatory effort, whether in the US or the EU. I find the carping about the tactical specifics and likely outcomes of any one particular action annoying, because a lot of the point is to do something, just to rein in these behemoths a tiny bit. Even if a particular remedy is not going to solve the entire problem immediately, it may at least force one more compliance check on whatever these companies were going to do next, and delay the inevitable just a bit. The Lilliputians were able to immobilise Gulliver by tying him down with hundreds of individually tiny ropes which they presumably already had lying around, rather than trying to construct one gigantic rope in the hope of getting it right first time and having it ready before the big G woke up and squashed them all.

🖼️ Google photo by Christian Wiediger on Unsplash; the “Extra” icon at the top is an Easter egg for the old heads.