Why CPC is the least bad proxy for commercial intent
Search volume tells you how many people ask a question. It says nothing about whether those people buy. The cost per click does: it is the price at which an auction of competitors, each measuring their own conversions, has settled. A keyword with 200 searches a month and a 45 CPC is worth more to most businesses than one with 20,000 searches and a 0.15 CPC, and the second is where inexperienced content plans go to die.
Use CPC as a signal, not as an equivalence. Paid and organic clicks on the same query behave differently: the ad is above the fold and gets the impatient click, the organic result gets the researcher. Conversion rates diverge accordingly, usually in favour of organic for considered purchases and in favour of paid for immediate ones. That is why this simulator asks for your conversion rate separately rather than assuming paid and organic behave alike.
When a keyword shows a high CPC and no ads on the live SERP, look closer: it often means the auction is seasonal, or that the term is expensive precisely because it converts and your competitors bid on it only during campaigns. Both are useful intelligence for the content calendar.
The position ladder: where the value actually sits
The ladder in the results makes a point most keyword plans miss. Moving from position ten to position five roughly triples the clicks; moving from five to one roughly quadruples them again — but the effort curve is the inverse. Positions eight to four are usually won with better content, correct internal linking and fixing the technical issues on the template. Positions three to one are usually won with authority, which is slow and expensive.
For most sites the rational target is therefore top five on many keywords rather than position one on a few. The exception is when the SERP is heavily truncated: on a page with an AI Overview, ads and a carousel, the difference between position one and position four can be the difference between visible and invisible, and the ladder collapses. That is precisely what the SERP context selector models.
One more nuance the ladder makes visible: below position ten, value falls off a cliff. A page on the second page of results is, for practical purposes, not ranking. If your keyword sits at fifteen, the decision is not "how do we get to one" but "is this page even the right asset for this intent" — and quite often it is not.
- Prefer many top-five positions over a few number ones on most sites
- Treat position eleven and below as not ranking at all
- Recheck the SERP layout before setting a target — it changes the ladder
- Value the keyword on annual customer value, not the first order
- Discount branded keywords: you would win most of that traffic anyway
From one keyword to a portfolio
Valuing keywords one at a time is useful for arbitration — should we build this page or that one — and misleading as a way to size a programme, because it double-counts demand. Ten keywords in the same cluster are largely the same searchers phrasing the same need differently, and one page will capture most of them. To size a programme, use the traffic forecast simulator on the whole cluster instead.
Where per-keyword valuation earns its keep is in deciding what not to do. Run it on the three terms a stakeholder keeps asking for, and the conversation changes: a head term with 30,000 searches, a 0.20 CPC and a 0.3 % conversion rate is worth less per month than a boring long-tail page nobody was excited about. Numbers settle these arguments faster than opinions.
Once you know which intents carry the value, the delivery question follows: who builds the pages, and with what authority behind them. If that capacity is not in-house, compare agencies and freelancers against the workload, and read how to choose an SEO agency before signing anything.