The Google Privacy Sandbox debacle could be the straw that breaks the company's back
It was supposed to be a cookie replacement we could all enjoy. Privacy Sandbox has turned into a dumpster fire.
Anguish in the world of website monetisation this week, after Google decided its recipe for a post-cookie future was about as appealing as a cactus lollipop, and reverse-ferreted its Privacy Sandbox initiative.
For those unsure, Google's Privacy Sandbox was the company's attempt to get ahead of global moves to ban "third-party" cookies which can work out who you are and follow you round your web travels like some creepy cyberstalker reporting back on what sites you visit.
Such cookies let advertisers and sites, including those run by publishers of course, send you the most relevant ads. And since more relevant ads are worth more money to everyone involved in placing them, the collective web has become addicted to them - even though everyone knows they suck from a privacy perspective.
Third party cookies (TPC) are almost certainly going to be banned, which is good for individual privacy. But no-one has convincingly worked out how this will happen without hammering monetisation across the web, at least in the short term.
Because it is assumed online commerce will be significantly disrupted, legislators - although keen to see the back of those naughty cookies - are being cautious in how quickly they push to ban them. They initially welcomed Google's 2019 plan to look into how it could be done smoothly, although initial warmth did not mean consent.
The Google alternative, the Privacy Sandbox, would have supposedly replaced third-party cookies with a system in the company's dominant Chrome browser that wouldn't let on to anyone the specifics of who you were or what you were looking at. Instead, it would anonymously clump you into groups by interest, and flag the interest of the group to advertisers who could place an ad. Not much more sophistication than a TV ad break.
Unfortunately, as time wore on it became increasingly clear that the cookie-free alternative proposed by Google was being forged in the fires of naked commercial opportunity: while it cut off user data and real insight from site owners and ad markets, it made even more of those precious commodities available to Google itself who could sit on the info, self-deal across its wide portfolio, and use it to cut the legs off anyone who one day dreamed of chipping away at its absurd percentage of global ad revenues and market share.
Quelle surprise. It turned out the Privacy Sandbox is more like being locked in the cyberstalker's basement.
Confusion reins
So in a case of frying pan or fire, Privacy Sandbox will continue but now in an opt-in fashion (which would doubtless become the default), while TPC continue in Chrome for those who don’t opt-in to Sandbox. While failing to solve one problem, they have conjured a new problem.
Some people are relieved by this continuance of TPC - with no alternative in place, stability is welcomed - while privacy advocates, regulators, and site owners are not.
Why not site owners? Shouldn't they be happy that the secret to better ad income is being retained? Well, when Apple rolled out a feature similar to Privacy Sandbox, ad revenues from iPhones dropped like a stone and evidence suggests users check the "privacy" opt-in and just leave it turned on. Meanwhile, research into Privacy Sandbox indicated it would divert something close to 80% away from current channels into Google's big chubby tummy, even more than it does now.
The horse is galloping without a rider. Not only has this wasted years of effort and expenditure from across all online industries to prep for what the Privacy Sandbox was meant to offer, but we are again collectively unaware as an industry - by which I mean, anyone doing business on the internet - of what comes next.
Publishers are like countless others trying to work out how to navigate a world where TPC don't exist. Not because they don't know how in principle - publishers being all-time greats at understanding what to give their audiences - but rather because the whole ad industry is led around by Google's decisions like a bull with a ring in its nose. There may be some great ideas and alternate tech, but it's Google's decisions that shift markets.
Inadvertent success?
However, there is a possible silver lining.
Everything Google does at the moment is to be viewed through two lenses:
The impact of AI on its search business
The forthcoming anti-trust trial looking at its stranglehold on global ad spend
This is all about the second one.
While making Privacy Sandbox an opt-in was clearly designed to help defuse regulatory anger and industry shock at the obvious info grab and strengthening of market dominance, it may in fact do the opposite.
The single greatest threat to Google right now is more evidence that the company really is as dominant in the ad market and search business as critics and prosecutors say it is.
What the Privacy Sandbox whiplash has done is reiterate at the worst possible time that Google can turn the entire ad industry, and all who rely on it, from one direction to another with a single decision.
As the lawyers limber up for the re-commencement of the Trial of the Century, this won't have gone unnoticed.
Mum's the word
OpenAI may have had its hands full with the New York Times case that pitches content makers against content takers, but it has a new enemy: 9 million British mums. After an attempt to play nice, UK site Mumsnet are going to law after finding convincing signs that OpenAI has ripped off some or all of its billions of words of content. Click the panel below as Ricky Sutton re-dons his journo hat to interview Mumsnet CEO Justine Roberts.
Haute couture vs hot manure
Conde Nast have slapped content-users Perplexity.ai with a cease-and-desist to stop using CN content to generate responses for their AI search engine, accusing them of plagiarism. The practice drew criticism from content creators, but also shed light on a broader issue in the AI industry about scraping content without permission.
Read
Paywalls for robots
As media firms struggle to work out just who is pinching their content and who isn't, this video podcast looks at the rise of new tools to act as paywalls against AI scrapers - handy since we know they can happily ignore your robots.txt instructions. See the video below.
The simplest is often the best
And how else do you find out who's taken your work? Sometimes, it can be as simple as searching for it in AI itself.
Read
Time for the good AI
INMA's Sonali Verma highlights AI's widespread use in newsrooms for tedium-reducing tasks, stressing the need for human oversight will continue. Her advice is to start small, show value quickly, and establish guidelines and teams. Alternatively of course - speak to us! Glide can tell you about our approach to Augmentative Intelligence and how it is having immediate, positive impact.
Read
Have you Reddit about this SEO trick?
Reddit seems today to have blocked all crawlers via their robots.txt file. If you search for site:reddit.com on Bing and order by the last 24 hours, nothing appears. If you do the same thing on Google, results aplenty. This either means Google is ignoring robots.txt, or they are populating their search results directly as part of their $60m deal with reddit for AI training. One to keep an eye on, monopoly hounds.
G, thanks
Use Google shorteners for URLs and links? The goo.gl URL shortening service is being shut in just over a year, meaning billions of URLs will turn into 404s. Start planning now if you lean on the service.
Read
Great headline…
No puns this week? Never!
Read
I Will not use Chrome anymore
Good points 👌