AI and inclusive metrics are forcing marketers to get a lot more specific, a lot more often.
Letās not pretend this is just a trend. Between AI content tools and inclusive marketing strategies, the bar for what counts as effective has moved. Itās not enough to crank out more content; it has to land with the right people, and we need to know, with real data, whether it actually did.
So, hereās the thing: scale and personalization used to be at odds. Now theyāre tangled together. AI lets us produce faster, sure, but that speed doesnāt mean much if weāre still measuring success with one-size-fits-all KPIs. And thatās exactly where inclusive marketing is shining a light.
What your averages arenāt telling you
HubSpot just published a piece that digs into a blind spot most marketers didnāt know they had: measuring campaign performance by averages. It sounds harmless, right? But when you zoom out too far, you miss the details that matter.
The example they use is a campaign for the film Sinners. One scene, which resonated deeply with Black audiences, almost got axed because general market testing didnāt pick up its emotional weight. The data said āmehā; the actual audience said ākeep it.ā That disconnect is costing brands more than they think.
Marketers are being pushed to triangulate data now, looking at how different subgroups respond to the same campaign across metrics like brand lift, engagement, and retention. Not just because itās the right thing to do, but because ignoring it is expensive. U.S. minority consumer spending is projected to hit $7 trillion. Thatās not a rounding error; thatās a market.
AI can write, but can it connect?
Now letās talk about the content itself. According to HubSpotās State of AI Marketing Report, more than half of marketers are using AI to draft everything from emails to full blog posts. No surprise there. The tools are fast, cheap, and increasingly
decent.
But hereās the rub: 46% of those same marketers say AI-generated content has problems. Accuracy, bias, toneāitās all on the table. The content might be technically correct, but emotionally flat. Or worse, it might be subtly wrong in a way that erodes trust.
The takeaway isnāt to stop using AI. Itās to stop treating it like an autopilot. AI works best as a collaborator. You prompt it with purpose, edit it with intent, and train it on data that reflects your actual audience. Otherwise, youāre not just using a tool; youāre scaling mediocrity.
Zoom out, then zoom way in
Put these two threads together and the message is pretty clear: scale is great, but precision is mandatory. Whether youāre segmenting KPIs or training AI to reflect your brand voice, the work now lives in the details.
And if youāre not paying attention to those details, someone else will. Someone whoās willing to ask: who is this for, really? And how do we know it worked?
One more thingāGoogleās getting weird
While weāre on the subject of AI and precision, hereās something that should be on your radar: Googleās AI Overviews are acting a little… erratic.
According to Ahrefs, 76% of AI Overview citations are pulled from the top 10 organic results. So far, so good. But in 8.64% of cases, those Overviews show up outside of position #1. Thatās a shift. And it gets weirder. Pages written in āAI modeā are consistently failing to rank, even when they come from high-authority domains.
So if you thought you could just hit āgenerateā and coast on domain strength, think again. Googleās not buying it.
That’s it for today, folks.
Catch you in the next post.
Until then, keep building.
ā Perfect Sites Blog
