Tech debt is your new AI epic

Recently, a rather audacious cold pitch landed in my inbox. Your competitor has 78 structured data items on their page while yours only has 34! I'd like to take a moment to respond to that email now.

Dear sir, madam, or gentlethem,

That's not how this works.

Sincerely,

Code bloat

At Google Search Central Live APAC, Google Gary Ilyes shared that excessive or redundant schema only adds page bloat and is not used as part of the ranking process. He likely saw the inbound wave of useless structured data inbound from GEO efforts. Still, studies like those from Ahref where they added schema to 1.9K pages and saw results "Statistically indistinguishable from zero" aren't as catchy as not-another-purple-circle-logo's latest PR stunt.

Businesses are tumbling product-over-tea-kettle to shove AI into every nook and cranny, schema into every paragraph, and llms.txt into the loneliness of unrequested file types. Build faster! Release more! Uh oh, what's going on with our servers?

Code bloat. Tech debt. CMSes battling like territorial beta fish. All of that can be ignored for a little while. At least until it's time to make a big pivot or move. Then your steering a busted shopping cart overfilled with cars instead of that Maserati promised with all those rapid-fire released.

Jono Alderson penned an excellent look into why unglamorous work of reducing complexity and shoring up systems is key for excelling in AI– and everywhere else.