Writing experience signals you can actually prove

Why “I have 10 years of experience” doesn’t do anything

I’ve written that sentence. Early on, on an old affiliate site, I put “with over a decade of hands-on experience” in an author bio and moved on. It didn’t help. Nothing measurable changed, because there was nothing behind the sentence for anyone, human or algorithm, to check. A claim of experience and evidence of experience are different things, and search quality raters are specifically trained to tell them apart.

Google’s quality rater guidelines added “Experience” to E-A-T in December 2022, making it E-E-A-T. The guidelines are public, and they say raters should look for first-hand, life experience with the topic, not just credentials. That’s a document raters use to evaluate search quality, not a ranking algorithm, and I’m not going to pretend I know exactly how or whether individual pages get scored against it. What I do know, from running content sites for years, is that the industry conversation shifted hard toward “show, don’t tell” around the same time, and the sites in my own portfolio that lean on actual proof tend to hold up better over time than the ones that just assert authority.

What counts as a claim, not a signal

Before getting into what works, it’s worth being blunt about what doesn’t. These show up constantly in SEO content and they cost nothing to fake, which is exactly the problem:

• Author bios that say “SEO expert” or “10+ years in digital marketing” with no specifics
• Stock photography standing in for the person or the process
• Generic “as someone who has tested dozens of tools” lines with no tools named
• Screenshots that could have been pulled from anywhere, with no identifying detail
• “In my experience” followed by advice that reads identically to every other article on the topic

None of these are lies, necessarily. But none of them are provable either, and that’s the distinction that matters. A rater, a competitor, or a reader can’t verify any of it. If it can’t be verified, it’s just tone.

What actually holds up

The signals that are hard to fake are the ones that take effort to produce and would be awkward to fabricate. A few categories I actually use on my own sites.

Original screenshots with your own interface state showing are one of them. I mean a screenshot with your account name, your data, your specific numbers visible (redacted where it needs to be for privacy, but visibly yours). Anyone can grab a stock screenshot of a tool. Fewer people bother running the tool themselves and showing their own results.

Specific failure points are another. If you’re writing about a tool, a tactic, or a process, name the exact place it broke for you. “The API rate limit kicked in at 47 requests” is a sentence nobody writes unless they hit it. Generic praise or generic criticism is easy to generate. Specific failure detail isn’t, because you need to have actually run into it.

Timestamps and version numbers do real work too. “As of this update” plus a date, or “in version 4.2 of the plugin,” pins the claim to a moment that can be checked against the tool’s own changelog. It also protects you, because tools change and your advice ages. If you tell readers what version you tested, you’re not making a permanent claim you can’t back up later.

Your own data counts, even small data. I’ve run tests on my own sites where the sample size was one site, or three sites, and I say so. “Across three sites in my portfolio, this cut load time from roughly 2.1 seconds to 1.4 seconds” is provable in the sense that it’s a specific, falsifiable claim about a specific set of properties. It’s not a study. I don’t pretend it is. But it’s more than “this will speed up your site,” which is just marketing copy.

Process documentation matters as much as outcomes. Screen recordings, step-by-step screenshots of a setup, a config file with your own values (secrets stripped). This is the part people skip because it’s slower to produce than writing the outcome and calling it a day. It’s also the part that’s hardest to fake, because faking a full process walkthrough basically means doing the process.

Author pages that mean something

A byline with a name and nothing else is decoration. What makes an author page functional as an experience signal:

• A real, checkable link to something the author actually runs or owns (a site, a GitHub, a YouTube channel, a LinkedIn with actual activity, not a shell profile)
• Consistency between what the bio claims and what the content on the site actually demonstrates. If the bio says “runs a portfolio of content sites,” the articles should read like someone who does that, with the specific headaches that come with it, not like generic advice columns
• No stacking of unrelated credentials. If someone is a “certified SEO expert, digital marketing strategist, growth hacker, and content specialist,” that’s a list of titles, not evidence of anything

I keep my own author bio short and specific to what I actually do, run my own site portfolio, handle the SEO and content and link building myself, and make the mistakes myself. It’s not impressive copy. It’s checkable.

Things I stopped doing

For a while I tried “expert quote” sections, reaching out for a one-line quote from someone with a title, to sprinkle in as an authority signal. I stopped. It reads exactly like the automated version of the same tactic, and readers, and probably raters, have seen a lot of it by now. A single generic quote from someone with a job title doesn’t demonstrate experience with the specific thing the article is about. It demonstrates that I sent an email.

I also stopped writing “case study” in the header of anything that wasn’t actually a study. If it’s one site over one month, I call it that. Overselling the format erodes trust faster than underselling it builds trust.

A practical checklist before publishing

Before I push an article live now, I run it through a short list:

1. Is there at least one piece of media (screenshot, screen recording, chart from my own data) that couldn’t have come from a competitor’s article?
2. Does the author bio link to something that still exists and still gets updated?
3. Is there a specific number, date, or version reference somewhere in the piece, not just adjectives?
4. If I removed every instance of “I” and “my,” would the article still sound like it came from someone who did the thing, or would it collapse into generic advice?
5. Would I be comfortable if a reader tried to verify the specific claim I made? If not, I either cut it or soften it to what’s actually true.

None of this guarantees anything about rankings or timelines, and anyone telling you it does is selling something. What it does is make the content harder to dismiss and harder to copy, because copying a claim is free and copying a documented process isn’t.

If you want more breakdowns like this on how to build content and site signals that actually hold up, The SEO Desk covers keyword research, link building, and technical SEO from the operator side, not the theory side.