Why one number?

Open your favorite SEO tool and count the numbers on the first screen. Domain rating. Health score. Crawl errors. Core Web Vitals — four more numbers right there. Keyword positions, backlink counts, traffic estimates, "content grade."

Forty-seven metrics, most of them green-ish, and you still can't answer the only question you actually have: is my site getting healthier or worse?

That question is why TrustGrowth is built around a single score.

For us, this wasn't hypothetical — before there was a TrustGrowth Score, TrustGrowth was that dashboard. We had an SEO score coming out of site audits. A set of E-E-A-T signals. A visibility index built from keywords, competitors, and search rankings. Authority metrics from backlinks, carrying their own scores. Every one of them measured something real, and none of them could answer what any of it actually contributed to a site's trustworthiness or growth — or how they were supposed to work together.

The breaking point was ranking. The moment we tried to put sites side by side, the ordering changed depending on which metrics a site happened to have data for. And underneath that sat a fairness problem we couldn't unsee: a three-month-old site and a ten-year-old publication don't just score differently — the same metric means something different at each stage, and different again depending on the marketing strategy behind the site. So how do you make all of it make sense? That question is what the TrustGrowth Score is: one number, built deliberately, as the only honest way to make those metrics mean something together.

A score is a decision instrument, not a report

Dashboards report. Scores decide.

When a metric exists to be looked at, more of them feels like more insight. But when you need to act — fix this before that, keep going or change course — a pile of disconnected metrics quietly transfers the hardest job to you: weighing them against each other. Every site owner staring at a dashboard is doing amateur composite scoring in their head, badly, differently each time.

So the honest version of "we show you everything" is: we made the hard part your problem.

A composite score does that weighing explicitly, consistently, the same way every day. That's the entire point. Not that one number captures everything — it can't — but that the weighing happens in a system that can be inspected, questioned, and improved, instead of in whoever's head is looking at the dashboard that morning.

The confession

Compressing a website's health into one number destroys information. There is no way around this. Two sites can hold the same score for different reasons; a site can improve in one dimension while declining in another and the number barely moves.

We did it anyway — because the alternative destroys more information, just invisibly. The unweighted dashboard doesn't avoid compression; it outsources it to your gut.

But doing it anyway creates a debt. If you ask people to trust a single number, the number has to earn it. That obligation shapes almost every design decision in the TrustGrowth Score, and it comes down to three demands.

What a score must get right to deserve trust

1. You should know what it looks at, and why.
Not the arithmetic — the reasoning. Every input to the TrustGrowth Score is there because it answers a question a real site owner would recognize: Can search engines actually use my site? Is it fast enough that people stay? Does the content deserve to rank? Would a stranger trust who wrote it? If we can't explain why a signal belongs, it doesn't go in. This series will walk through every pillar of the score and make exactly that case, one by one.

2. It has to be honest about what it doesn't know.
Most scores treat missing data as a zero, or worse, quietly fill it with an average. Both are lies with decimals. When the TrustGrowth Score doesn't have enough signal to judge a dimension — a site too new for trend data, a check that hasn't run yet — it says so, and the dimension sits out rather than faking a contribution. This matters double the moment scores get compared: a ranking where position depends on which data happens to exist isn't a ranking, it's an accident of coverage. We learned that one firsthand. A number that pretends to know is worse than no number at all.

3. It has to resist being gamed.
Any score that rewards checkbox compliance gets gamed into meaninglessness — everyone learns the checklist, everyone scores 90, the number means nothing. So the score leans on earned signals over declared ones: what your site actually does in search results, references other sites actually give you, experience real visitors actually get. You move the score by making the site genuinely better, because the things it measures are downstream of the site being genuinely better.

The first two were the hardest to commit to. Explaining your reasoning in public invites people to argue with it — that's the deal. And refusing to fake missing data is exactly what broke our early attempts at ranking sites: it would have been far easier to fill the gaps and present a tidy leaderboard. We took the harder version both times, because the tidy version is the one that eventually gets found out.

Eight ways a site can be healthy

Under the single number, the score evaluates a site across eight dimensions — pillars, each one a different way a site earns trust with search engines, AI systems, and the people they serve:

• Technical health — can crawlers actually get to, read, and understand your pages?
• Performance — is the site fast enough that speed never costs you a visitor's patience?
• Content — do the pages deserve the clicks they're asking for?
• E-E-A-T — experience, expertise, authoritativeness, trust: would a careful stranger believe you?
• Growth — is your search presence expanding or decaying right now?
• Authority — has the rest of the web vouched for you with real references?
• Visibility — how much of search does your site actually occupy today?
• GEO — when AI systems answer questions, do they see you?

Each of these gets its own article in this series: what it measures, why it earned its way in, and what "good" looks like — without the formula, because the formula isn't the point and (as we'll cover in the next piece) it's designed to keep evolving. The reasoning is the part we're committing to in public.

We take our own medicine

One more thing, because "trust us" is exactly the kind of claim this whole post argues against: our own sites are scored by the same system, and the results are public — history, dips, and all. You can inspect our live score page and see precisely the instrument we're describing, running on us.

Next in the series: how we keep the score honest — the champion/challenger engine that lets the score improve without ever silently moving under your feet.

The TrustGrowth Score series explains what our score measures and why — every pillar, every reason. New pieces weekly.