Why keyword prioritisation changes the calculator's inputs
The default page asks a single question: if this keyword ranks at this position, how many clicks and conversions might it bring? Prioritisation asks a different one: of several candidates, which should the team work on first? Answering that fairly requires every candidate to be entered on the same honest basis, and three inputs need to change to get there.
Volume needs to describe what a page can win, which is a cluster rather than one keyword. A single page rarely ranks for a head term alone; it ranks for a family of variants that share intent. The keyword clustering guide explains how to build those groups. Position needs to describe what the site can reach, not what it would like to reach. Entering the top position for every candidate turns the estimator into a ranking by volume, which is the habit the exercise is meant to break. And the conversion rate needs to follow intent: a cluster of research questions converts far less often than one of comparison or purchase queries, so one site-wide rate flatters informational topics.
The inputs for keyword prioritisation
| Calculator input | On the default page | For keyword prioritisation | Where to find it |
|---|---|---|---|
| Monthly search volume | Volume of one keyword | Combined volume of the cluster one page can rank for | Keyword tool, grouped by cluster |
| Ranking position | Any position | A realistic target position given the competition on the results page | Review of who ranks now |
| Conversion rate | One rate for the site | The conversion rate for the cluster's intent; informational clusters convert far lower than commercial ones | Analytics conversion rate of pages with similar intent |
For position, review the SERPThe page of results a search engine shows. for the cluster's main query. Note the strength of the sites ranking, whether the results are product pages, guides or tools, and how many places are taken by features such as ads, maps and answer boxes. A site with modest authority facing established leaders should enter a middle-of-page position; one that already ranks for related topics may reasonably enter a higher one.
For conversion rate, label each cluster's intent using the method in the intent classification guide, then use the conversion rate your analytics show for existing pages of the same intent. If the site has no such pages, use a conservative assumption and apply it consistently.
Entering the prioritisation inputs step by step
Group the candidates into clusters
Turn the keyword list into clusters that one page each could answer. Add up the monthly volume of every query in each cluster, leaving out queries that belong to a different page.
Set a realistic position for each cluster
Review the results page for each cluster's main query and choose the position the site could plausibly hold. Apply the same standard to every cluster so the comparison is fair.
Assign a conversion rate by intent
Label each cluster informational, commercial or transactional and enter the conversion rate your analytics show for existing pages with that intent.
Run each cluster and record the outputs
Enter each cluster in turn and record the assumed click-through rate, estimated clicks and estimated conversions in a single table, one row per cluster, sorted by conversions.
A worked prioritisation with illustrative inputs
A pet insurer compares two content clusters
Pawprint Mutual, an invented pet-insurance provider, is choosing between two clusters for its next page. The first is a commercial cluster around comparing dog insurance cover, totalling 4,400 searches a month; a review of the results page suggests position four is achievable, and its comparison pages convert at about 1.5%. The estimator assumes an 8% click-through rate there, giving about 350 clicks and roughly 5 conversions a month. The second is an informational cluster about puppy vaccination schedules with 14,000 searches a month; strong veterinary sites hold the top places, so position seven is realistic, and its guide pages convert at about 0.4%. That gives about 560 clicks but only about 2 conversions. The smaller commercial cluster goes first, while the informational one stays on the list as a supporting page.
SEO Traffic Estimator
Open the SEO Traffic Estimator →Prefilled with the example's inputs: Monthly search volume 4,400 · Ranking position 4 · Conversion rate 1.5% → Assumed CTR 8% · Estimated monthly clicks 352 · Estimated conversions 5.
All figures above belong to the invented insurer and rest on the estimator's illustrative curve. Change the inputs in the embedded estimator to the second cluster's figures to see the comparison directly.
Interpreting the estimates to rank keywords
Read estimated conversions first, because that is the output that ranks the candidates. A cluster with fewer clicks but a commercial intent often outranks a larger informational one once conversion rate is applied. Sort the table by conversions and treat that order as the starting point for the plan.
Then read estimated clicks as context. Clicks show the visibility a page would bring, which matters for clusters whose role is to build topical authority or feed internal links to commercial pages. A high-click, low-conversion cluster can still earn a place, but as a deliberate supporting page rather than because its volume looked impressive. The keyword mapping guide shows how to record each cluster's role against the page it belongs to.
Next, look at the assumed click-through rate and remember it is illustrative. The estimator's curve falls steeply from the top positions, so the gap between a reachable and an unreachable position is large. That is why a smaller cluster at a position the site can reach often beats a large one at a position it cannot. The curve is useful for ranking options on a consistent basis; it is not a promise of traffic, and real CTRShare of impressions that result in a click. varies with ads, features on the results page and how well known the brand is.
Finally, plan to replace the curve. Once the site has pages ranking, export click-through by position from Search Console and use those figures to recheck the estimates. A prioritisation built on your own click-through data is far more reliable than one built on any general curve, and it helps the wider SEOImproving visibility in unpaid search results. plan converge on what the site can actually win. When the chosen clusters need to become a dated plan, the SEO forecasting guide shows how to build that click-through table and add a baseline, ramp-up timing and low, central and high scenarios.
Pitfalls when keyword numbers go into the estimator
- Entering one head term's volume for a whole cluster. This understates what a page can win, and it penalises clusters made up of many long-tail queries with little volume each.
- Assuming position one. Entering the top position for every candidate turns the estimate into a ranking by volume and hides the clusters the site could actually win.
- Using the site-wide conversion rate for every intent. One rate overstates informational clusters and understates commercial ones, which reverses the order the estimator should produce. Intent-based CVRShare of visits or clicks that convert. figures are what make the comparison fair.
- Treating the estimate as a forecast. The curve is illustrative. Present the output as a ranking of options with stated assumptions, not as the traffic the content team will be held to.
