Export your top 20 queries by impressions and sort each into two piles: learning or hiring. Someone researching how a trust works is doing homework. Someone searching for an attorney in their county is shopping. If your visibility sits in the learning pile, the site is working as built — it was built for the wrong visitor.
"We rank well and the phone doesn't ring" is almost never a rankings problem
It's the complaint we hear most often, and the diagnosis is usually the same: the site is winning attention from people who were never going to hire anyone.
This is a good problem in one sense — it means the technical foundation works. Google can find you, read you, and rank you. The targeting is what's off, and targeting is cheaper to fix than infrastructure.
The two-pile sort
Open Search Console → Performance → Queries. Sort by impressions. Export the top 20 to a spreadsheet. Add one column: Learning or Hiring.
Then go line by line. The test is simple: could this person plausibly hire someone today?
Learning queries look like:
- "what is a revocable trust"
- "how long does probate take in south carolina"
- "difference between a 401k and an IRA"
- "when are quarterly taxes due"
- "how much are closing costs"
Hiring queries look like:
- "estate attorney charleston"
- "fee only financial advisor near me"
- "cpa for small business summerville"
- "probate lawyer dorchester county"
- "[your firm name]"
Some sit in between — "do I need a lawyer for probate" is a learning query with hiring intent underneath it. Put those in a third pile and don't agonize; the ratio is what matters, not the edge cases.
Now count. If fifteen of your top twenty are learning queries, you have your answer, and you can stop wondering whether it's an algorithm problem.
Why this happens to good firms
It's rarely carelessness. It's usually the natural result of two reasonable decisions.
You published helpful content. Someone told you to blog, so you explained the things clients ask. Those explanations rank, because they're genuinely useful and because informational queries are far less competitive than commercial ones.
Your service pages were never really written. They exist — "Estate Planning," three paragraphs, a stock photo — but they were built as a menu rather than an argument. Nothing on them answers why this firm, and nothing targets the phrase someone types when they're ready to call.
The result is a site where your best-performing pages are the ones that lead nowhere and your money pages are invisible.
The second pattern: cannibalization
Sometimes the intent is right and two of your own pages are fighting over it.
In Search Console, filter to a single high-value query and look at the Pages tab. If two URLs both show impressions for it — or if the ranking page alternates week to week — you're splitting the signal. Google isn't sure which page is your answer, so neither performs as well as one would.
The usual culprit is a blog post outranking the service page for a commercial term. That post has more internal links, more words, and more history; the service page has three paragraphs. Google picks the better page, which happens to be the one with no path to contacting you.
What you can and can't conclude
Verified — This is one of the few diagnoses you really can complete alone. The query list is your evidence, the sort is unambiguous, and the ratio is a fact. No interpretation required.
Directional — Why the mix looks that way is a step further. Whether it's thin service pages, cannibalization, weak internal linking, or a site architecture that buries the commercial pages — those look similar from the outside and are distinguished by looking at internal link flow and page-level query mapping together.
What to do with the answer
If you're heavy in the learning pile, the fix isn't to delete the helpful content — it earns trust and, increasingly, it's what answer engines quote. The fix is to build the commercial pages that don't exist yet, and to link the helpful content to them deliberately rather than leaving readers at a dead end.
If you find cannibalization, resist the urge to delete one page. Decide which URL should own the term, then make that decision legible: consolidate the thin one, point internal links at the winner, and give the winner the depth that made the other page competitive.
Ranking for the wrong intent is one of six symptoms that look like a traffic problem and aren't. The full map is at how do I tell if my website is losing me clients.