Open a keyword research tool with a blank mind and it will hand back a list ranked by search volume, sorted by a metric that has nothing to do with whether that phrase is how a real buyer actually talks. Volume tells you how many people search a term. It says nothing about whether that term belongs to someone early in the process comparing categories, or someone ready to buy who just hasn't found the right vendor yet. Sales calls already answer that question, and most businesses have dozens of them sitting untouched.
Why calls beat tools as a starting point
A keyword tool suggests phrases based on what other websites already rank for and what volume of searches a phrase gets. A sales call contains the actual words a real prospect used to describe their problem, in their own vocabulary, before a salesperson reframed it into industry language. That gap matters: a prospect might say "my invoices keep getting stuck in QuickBooks" where the industry term is "invoice reconciliation software." Whichever phrase potential customers actually use is the one worth targeting first, and a keyword tool has no way to know which one that is without being told.
What to listen for
The useful language usually shows up in three places on a call: how the prospect first described the problem before being asked any qualifying questions, the objections they raised about alternatives they'd already looked at, and the comparisons they made unprompted ("is this like [competitor] but for..."). Discovery calls are richer than closing calls for this purpose, because early in a call people haven't yet adopted the salesperson's terminology.
A simple process
Pull ten to fifteen recent call transcripts or recordings, whichever the CRM or call system already stores. Read or skim them specifically for how the problem gets named, not for what was sold. Write down the recurring phrases exactly as said, including the imprecise or informal ones — those are frequently the underused, lower-competition versions of a keyword that a generic tool would never suggest because they don't fit standard keyword syntax. Only after this list exists does it make sense to check those exact phrases in a keyword tool, to confirm they carry real search volume rather than being one prospect's idiosyncratic phrasing.
Where this fits into the rest of the plan
This is not a replacement for structured keyword research — it's the input that makes structured research aim at the right target instead of a plausible-looking one. The resulting phrases feed directly into the page-per-question approach covered in the SEO strategy for businesses with a specific offer, because a page answering the exact words a real buyer used converts differently than a page answering an industry term nobody outside the business actually types.
The ongoing habit
Call-derived keyword research isn't a one-time exercise. Language shifts as a market matures, competitors enter, and customers' own vocabulary evolves. Revisiting a fresh batch of calls every few months catches phrasing that wouldn't have existed the last time this exercise was run — a habit our marketing team builds into ongoing content work precisely because a keyword list from a year ago quietly goes stale.
