SEO Intelligence Without the BS
Notes on search: what changed, what it means, and what to do about it. 608 of them so far, going back to August 2024.
Start here
- Local SEO in 2026What actually moves local rankings, minus the forty-seven-tips listicles.
- Technical SEO checklistCore Web Vitals, crawlability, indexation, site architecture.
- Google Business Profile changesThree 2026 changes to how profiles feed local rankings.
- E-commerce SEOProduct pages, category structure, and the technical layer underneath.
- SEO for startupsWhere to spend first when the budget is small.
- The metrics that matterWhich numbers track revenue, and which ones most people track wrong.
The archive
How Long Do Core Web Vitals Take to Improve After a Fix?
Field data moves on a rolling window, so a correct fix looks like it failed for weeks. Here is what is happening and when to expect the number.
Which Time Period Should You Compare Organic Traffic Against?
The window you compare against decides whether a month looks like growth or collapse. Here is how to choose one that answers a real question.
How Much Structured Data Is Too Much on One Page?
There is no published cap on schema types per page. The real limits are accuracy, eligibility and whether you can keep the markup true over time.
Should a Landing Page Ever Be Left Out of Your Navigation?
Keeping a page out of the menu is normal. Leaving it linked from nowhere is a different decision, and the two get confused constantly.
Why Click-Through Rate Rises While Conversion Rate Falls
The same title edit that wins more clicks often wins the wrong ones. Here is the mechanism behind the trade, and the number that settles it.
How to Tell a UX Problem From a Search Problem
Traffic is down and two teams disagree about why. Three numbers separate a ranking problem from a presentation problem from a page problem.
Which Third-Party Scripts Are Costing You Core Web Vitals
Most failing Core Web Vitals scores are caused by code the site owner did not write. Here is how to find which tag is doing the damage and what to do next.
Why Two Analytics Tools Report Different Organic Traffic Numbers
Search Console and your analytics platform will never agree, and the gap is structural rather than a bug. Here is what each one counts and which to trust.
When Structured Data Should Be Removed Rather Than Fixed
Markup that makes claims nobody maintains is a liability, not an asset. Here is how to decide which schema types to keep, repair, or delete outright.
How Many Landing Pages Should One Service Have?
One page per service, or one per keyword? The answer comes from the search results themselves, not from your service list. Here is the test to run.
Is Bounce Rate Still Worth Reporting?
Bounce rate now means the inverse of engagement rate, which is a different metric wearing an old name. Here is when it earns a place on the report.
How Long Should You Keep a Landing Page That Never Ranked?
Before deleting a page that never ranked, check whether it failed or was never given a chance. The two look identical in a traffic report.
How to Tell Which Core Web Vitals Failures Are Worth Fixing
A red Core Web Vitals report can mean four thousand URLs or one template. Here is how to triage the failures by traffic, distance and shared cause.
How Many Conversions Before the Rate Means Anything?
A page-level conversion rate built on four conversions moves 25% when one more person converts. Here is how to tell a signal from arithmetic noise.
How to Measure Whether Your Structured Data Is Actually Working
Valid markup is not evidence that schema is doing anything. Here is what you can genuinely measure, and what a validator will never tell you.
Which UX Problems to Fix First When Search Traffic Is the Goal
Not every usability problem costs you search traffic. Here is the order to work in, from the ones that lose visitors outright to the ones that are taste.
How to Separate Branded and Non-Branded Organic Search in Reporting
A combined organic traffic number hides whether search is winning you new customers or just catching people who already knew your name. Here is the split.
Briefing a Writer Who Is Not an SEO
A good brief gets search results from a writer who knows nothing about search. Here is what belongs in it, and the three instructions to leave out.
Setting a Realistic Timeline for Organic Results
Search results arrive in stages, and each stage has its own signal. Knowing the sequence tells you whether month three is on track or already failing.
How to Test an SEO Change Before Rolling It Out Sitewide
Sitewide changes are hard to reverse and slow to evaluate. Testing on a subset first costs a few weeks and prevents the changes that quietly cost a quarter.
Recovering Search Coverage After a Redesign
Rankings dropped after the new site launched. Here is the order to check things in, and which of the usual suspects is actually responsible most often.
Keeping a Large Content Archive Accurate Over Time
An archive that grows faster than it is maintained turns into a liability. Here is a maintenance model that scales without rereading everything you published.
Deciding What to Prune and What to Keep
Deleting old content is sometimes the highest-return work available and sometimes an expensive mistake. Four questions separate the two.
Impressions Went Up and Clicks Did Not. What That Means
Rising impressions with flat clicks has four common causes and they call for opposite responses. Here is how to tell which one you are looking at.
The Queries You Rank For But Never Targeted
Every site ranks for questions nobody planned. Those accidental queries are the cheapest content research you own, and most teams never open the report.
How to Audit NAP Consistency Without a Paid Tool
Citation tools charge monthly to find inconsistent listings. The checks that matter most take an afternoon and no subscription. Here is the sequence.
How Much Faster Does a Page Need to Be to Matter?
Speed work has a threshold and a ceiling. Below the threshold you are losing visitors; above the ceiling you are optimizing something nobody experiences.
When Adding Internal Links Makes Things Worse
Internal linking is close to free and hard to overdo badly. There are four ways to get it wrong, and automated linking plugins produce three of them.
Domain Authority Is a Vendor Score, Not a Google Signal
Google does not use Domain Authority. It is one vendor modeling another company system. That does not make it useless, but it changes what it can decide.
What Search Console Cannot Tell You
Search Console is the most trustworthy free data you have about your search performance, and four of its limits routinely produce confident wrong conclusions.
How Wrong Search Volume Estimates Are, and How to Use Them
Every keyword tool reports volume as a precise number. It is a bucketed, modeled estimate of a moving target. Here is how to plan around that honestly.
How to Tell Whether Your Schema Markup Is Doing Anything
Valid structured data is not the same as useful structured data. Here is how to check which of your markup earns anything, and which is decoration.
Nothing filed under this topic yet.