Yes, I know Google exists
A few days ago, I posted a question on Facebook.
Hypothetically, how would one identify something like tennis elbow from having done terrible and irrevocable harm to one's elbow?
Purely academic question, driven by curiosity and pursuit of knowledge only.
A friend replied: “Doesn’t everyone use Google for this?”
Of course I use Google.
I also use AI. Frankly, I spend an unreasonable amount of my life looking at search results. I know where the search box is, and I know how to use it.
But Google wasn’t quite what I was looking for.
I was hoping someone I knew might wander into the comments and say, “Oh, I’ve had tennis elbow. Mine felt like this.”
That’s a different kind of answer.
I wasn’t looking for a definition
If I wanted to know the standard symptoms of tennis elbow, Google would be a perfectly sensible place to start.
Medical websites can tell me where it usually hurts, which movements tend to aggravate it, and when I should stop conducting independent elbow research and consult someone considerably more qualified than Facebook.
That information matters.
But sometimes the question we type, ask, or post isn’t the whole question.
Maybe what I really want to know is whether anyone I know has experienced something similar. What did it actually feel like? Did it happen suddenly or build over time? What did they initially think was wrong? Was there some oddly specific movement that made it obvious something wasn’t right?
None of that replaces medical advice.
It gives me another kind of information.
We have more ways to search than ever
For a long time, “look it up” more or less meant Google.
That isn’t really true anymore.
I can search Google. I can ask ChatGPT to explain something or help me compare several possibilities. I can look for a Reddit thread full of people who have dealt with the same problem. I can go directly to an authoritative source. I can ask a professional.
Or I can post a question on Facebook and see whether someone I know has actually experienced the thing I’m asking about.
Those aren’t simply different interfaces for getting the same answer; they’re different kinds of search.
Google is excellent at finding published information. AI is very good at synthesizing information, comparing it and letting us keep asking questions. Professional sources give us expertise. Communities can give us a much larger pool of lived experience.
And my own social network gives me something else again: people whose judgement I already have some basis for evaluating.
Sometimes I want the official answer.
Sometimes I want the summarized answer.
Sometimes I specifically want the person who can say, “Yes. This happened to me.”
Sometimes the people are the search
Asking people a question is a form of information retrieval too. It’s just a much smaller and considerably stranger dataset.
My Facebook friends aren’t an authoritative medical database.
They are, however, people I know.
I know who likely has relevant experience. I know whose judgement I trust. I also know who is perfectly capable of diagnosing me with gout, scurvy and demonic possession before lunch.
That context matters.
When someone I know answers, I don't evaluate the words in isolation. I have information about the source.
And unlike most search results, I can immediately ask another question.
“Was it more like this?”
“How long did that last?”
“Wait, what do you mean by crunching noise?!”
Human conversation has a credibility layer that doesn’t translate neatly into ten blue links.
Google has an abundance problem
There’s another reason I don’t automatically turn to Google for every question.
For a lot of searches, there isn’t a shortage of information. Instead, there’s an enormous amount of the same information.
Search an ordinary question, and you can find page after page explaining essentially the same handful of points. Sometimes those pages are genuinely useful.
And sometimes you get three paragraphs in and realize somebody identified a keyword opportunity and then worked very hard to turn six useful sentences into 1,500 words.
Read enough search results, and you start recognizing the pattern.
The headings change. The examples get shuffled around. The wording is different, but the information isn’t.
This isn’t entirely Google’s fault. Search engines can only index what we put on the web.
For years, though, an enormous amount of SEO content has been created by looking at what already ranks and producing another version.
That approach becomes harder to justify in the age of generative AI.
If 50 websites all explain the same thing, an AI system is extremely good at pulling those common points together and giving someone one useful answer.
Which raises an uncomfortable but necessary question:
What does page 51 contribute?
AI makes interchangeable content easier to see
AI didn’t create generic web content, but it did make the problem much more obvious.
If the main value of an article is that it gathers information already published in a dozen other places, AI can increasingly perform that job for the person searching.
That doesn’t mean websites, search engines or original articles are going away.
It means that simply having information is less of a differentiator than it used to be.
Google’s own guidance points in a similar direction. It asks whether content provides original information or analysis, goes beyond the obvious, adds substantial value compared with other search results, and demonstrates first-hand expertise where that matters.
Google has even told site owners thinking about its AI search experiences to focus on creating “unique, non-commodity content” that adds value rather than trying to reverse-engineer some special AI-search formula.
That phrase, non-commodity content, gets at the issue pretty well.
If your page could have been published by almost any company in your industry with the logo swapped out at the top, how much value is it actually adding?
Human experience has become more important, not less
This is where the Facebook question starts wandering back into SEO.
Someone who actually does the work usually knows things that aren’t captured by the generic explanation.
They know what customers consistently misunderstand.
They know which (supposedly) minor problem tends to become a much bigger one.
They know which part of a process sounds easy until someone actually tries it.
They know when the textbook answer works and when reality gets inconvenient.
They know the question people ask after they’ve already read the introductory article.
That kind of knowledge has provenance. It came from somewhere.
Someone built the house. Treated the patient. Ran the program. Installed the software. Handled the insurance claim. Repaired the furnace. Answered the same client question 200 times.
That doesn’t automatically make every personal experience accurate or useful. Experience and expertise aren’t the same thing, and neither one removes the need for evidence.
But experience can contribute something that a rewritten summary of everybody else’s content can’t.
Google explicitly recognized that distinction when it added the second E to E-E-A-T: experience. Its quality guidance asks whether content demonstrates first-hand or life experience, when that kind of experience is useful to the person searching.
Google also makes clear that E-E-A-T itself isn’t one specific ranking factor. Its ranking systems use many signals intended to identify content with those qualities.
So I wouldn’t reduce this to “Google rewards personal stories now.”
The point is broader:
Search is increasingly trying to distinguish between content that merely exists and content that deserves to exist.
Maybe we need to ask better questions
Keyword research still matters.
Understanding what people search for still matters.
Making sure search engines can crawl a site, understand its pages and connect them with the right searches still matters.
But “What are people searching for?” shouldn’t be the only question behind a content plan.
There should be another one:
What do we actually have to add?
Maybe your organization has data that nobody else has.
Maybe your staff can explain the trade-offs customers rarely hear about.
Maybe you’ve seen the same problem enough times to know the early warning signs.
Maybe you have a real example that shows how the work unfolds.
Maybe your industry repeats a piece of conventional wisdom that deserves a little more scrutiny.
Or maybe there’s simply something you know because you’ve been doing the work for 20 years.
That’s the material worth bringing to the page.
You can still structure it properly. You can still research search intent, use the terminology people actually look for and make the information easy for search engines and AI systems to interpret.
But the SEO should help useful knowledge travel.
It shouldn’t be the reason the knowledge was manufactured in the first place.
The answer isn’t to stop Googling
Having more ways to search is useful precisely because they aren’t interchangeable.
I haven’t stopped googling answers. Or using AI. Or reddit. Or credible professional sources.
I use all of them.
Sometimes I need a fast factual answer. Sometimes I want the primary source. Sometimes I want someone qualified to tell me what the evidence says. Sometimes I want to compare several perspectives.
And sometimes I want to know whether anyone I personally know has catastrophically damaged an elbow in a way that sounds suspiciously similar to mine.
The skill isn’t knowing which search tool is best, but knowing what kind of information you’re looking for and choosing accordingly.
As generic information gets easier to retrieve, summarize and reproduce, the information rooted in actual knowledge, experience and credible human perspective will become more valuable, not less.
For people searching.
For Google.
And for organizations wondering what they should bother adding to the internet in the first place.
So yes, I knew I could Google it.
That just wasn’t the search I was conducting.
