Sample layout. Not a real client.
Trade, city: the problem in one line
- Starting point
- Rank grid before any change
- What changed
- The one or two changes that moved it
- Result
- Rank grid after, and the change in calls
The Scientific SEO Audit: every finding is a test.
Most audits tell you what is wrong and stop there. This one records where you stand and states a suspected cause for each problem, with the single change that would prove or disprove it.
No long contracts. You talk to Simon.
Map pack: AC installation Poway
A
B
C
- 17 years in SEO
- Engineering degrees
- 85% of clients in San Diego County
- No long contracts
What an untested audit leaves you with
- Recommendations arrive with no evidence behind them
- Ten changes made at once, so nothing is learned
- No stated way to tell whether a fix worked
How the method runs
Scientific SEO
- 1 Measure You see a dated baseline: rank grids and crawl results, plus the searches that show your pages.
- 2 Change one thing I pick one hypothesis and name the single change that tests it.
- 3 Measure again About two weeks later, I take the same measurements the same way.
- 4 Keep what works You repeat what was confirmed on similar pages, and I cross disproved ideas off the list.
What the method is applied to
- Technical health of the site
- Map visibility by neighborhood
- Existing pages, one by one
- Competitors who outrank you
- Rank grid baselines
- A written hypothesis per problem
How the Scientific SEO Audit reasons from measurement to test
The Scientific SEO Audit is my audit method: I measure where a business stands in search first, then write a hypothesis for each problem the measurements show and recommend changes that can be tested one at a time. Every part of an SEO audit I carry out uses this reasoning.
Measurement comes before opinion
I diagnose nothing until a baseline exists. For a local company the baseline starts with rank grids: the same search run from many points across your service area, so I can see where you appear and where you drop out. That’s the basis of local rank tracking. Alongside the grids go a crawl of the site and, if you give me view access, Google’s own record of which searches showed your pages.
The baseline carries a date. Without it, later improvement is a matter of opinion.
Each problem gets a written hypothesis
A suspected cause becomes a hypothesis once it’s stated clearly enough to be proved wrong. I write every problem in the audit in the same form. Take an air conditioning company based in Poway.
| Part | What it records | Example |
|---|---|---|
| Observation | What the measurement shows | Appears for AC installation near its shop, absent on the grid points in Rancho Bernardo |
| Suspected cause | The most likely reason, given the evidence | No page mentions Rancho Bernardo, and the companies that appear there each have one |
| Change | One specific edit | Publish one page about AC installation in Rancho Bernardo |
| Measurement | What is checked again, and when | The same grid, about two weeks later |
| Disproof | The result that would rule the idea out | The page is indexed and the Rancho Bernardo points don’t move |
Most audits leave out the last row. Without it, you have no way to tell whether the work was worth paying for.
Why changes are tested one at a time
If a new title and a new page go live in the same week as a category change on the Google Business Profile and rankings rise, nobody can say which one to repeat. A change that hurt can also hide behind one that helped. So I list changes singly, and where two can’t be separated the audit says so. This is slower than doing everything at once, and what is learned carries over to the next page.
How the tests are ordered
A problem that stops Google from indexing a service page comes first, because nothing about that page can be measured until it’s in the index. After that I order tests by how much they could matter and how cleanly I can read the result, weighing the size of the likely effect against the work involved and the risk of making things worse. Changes that are easy to reverse go early.
Where the method runs out
This is disciplined before-and-after testing, and it falls short of a laboratory experiment. I can’t hold Google still while a test runs. A Google update, a competitor’s new reviews or a heat wave that changes what people search for can all move the numbers. A small site may also receive too few searches for a result to be clear. When that happens, the audit records the test as inconclusive and says why.
Where the method is applied
In the technical SEO audit the hypotheses are about how Google reaches your pages. In the local SEO audit they are about the map. To start either, use the SEO audit request form.
Also in Scientific SEO Audits
Questions owners ask
What a result looks like
Every engagement starts and ends with the same measurement.
Scientific SEO Audits across San Diego County
- San Diego
- Poway
- Escondido
- Carlsbad
- El Cajon
- Chula Vista
What if the first test shows nothing?
Then that cause is ruled out, and the audit plans for it. Each hypothesis states in advance what result would disprove it. If the measurement about two weeks later shows no movement, I cross that cause off, in plain words, and move to the next most likely one. You have spent one small change to rule something out, which costs less than a year of work built on a wrong idea.
Is this the same as A/B testing?
No. An A/B test shows two versions of a page to different visitors at the same moment. Google ranks one version of a page, so for a single page that can't be done with rankings. I run a before-and-after test: one change, with the same measurement taken before it and again afterward. It's less rigorous than a true split test, because a Google update or a competitor's new reviews can move the numbers in the meantime, which is why the audit notes what else changed during the test window.
Can I run the tests myself after the audit?
I write each recommendation so that someone other than me can carry it out, so yes. It names the page or setting and the exact change to make there, along with the measurement to repeat. The part that needs care is discipline. Make one change and write down the date. Then leave everything related alone until the re-measurement. If you'd prefer that I run them, I can, with no long contract.
Does every recommendation have to be tested separately?
Repairs with a known cause, such as a broken redirect or a page accidentally blocked from Google, are simply fixed. I test where the outcome is uncertain, as with a new title or primary category on the Google Business Profile.
Ask for an audit built on measurements.
Fill in your trade and the areas you serve. My reply says what I'd measure first.
- A rank grid showing where you appear, and where you don't
- The technical and Google Business Profile issues that matter most
- What I'd change first, in plain English
Prefer to talk? Call (619) 675-7678, 9am to 6pm.