Skip to content
Documentation index for AI agents: /llms.txt — every page is also available as markdown by appending .md to its URL.

Reference

Keywords metrics

How each card on the Keywords overview, strategy and detail pages is calculated.

All position-weighted metrics on these pages use the click-through rate curve.

The number of keywords in the current filter scope, across all regions. No date range applies.

The sum of average monthly searches over every keyword in scope: the demand pool you compete in. Per keyword, the value is the most recent search-volume update.

The share of monthly search volume that sits on branded keywords, weighted by volume:

branded share = branded volume / (branded volume + non-branded volume) × 100

Keywords that are not yet classified are left out of both sides.

Example: branded keywords total 3,000 searches a month, non-branded 12,000 → 20.0%.

The same two groups side by side. Each shows its keyword count, search volume, how often your brands appeared in results over the date range, and the average position of those appearances (sum of positions / appearances). Branded is defence: people already search for you. Non-branded is discovery: category demand you can win.

Annual commercial value and Commercial value of rank

Section titled “Annual commercial value and Commercial value of rank”

The yearly cost of buying, as ads, the organic clicks your positions already earn. Per keyword:

realised = search volume × CTR(your best position) × CPC × 12
potential = search volume × CTR(1) × CPC × 12
uplift = potential − realised (never below 0)
  • Your best position is the best rank any of your active brands holds in the latest scan.
  • Keywords with no search volume or no CPC are skipped.
  • Keywords with no scan data get an uplift of 0, so unknowns never inflate the total.
  • The KPI tile shows the sum of realised, with the sum of uplift as potential. Its trend badge is the relative change against the previous range.
  • The card lists the top 8 keywords by potential. The solid part of each bar is realised value, the tinted part is uplift.

Example: “running shoes”, 10,000 searches a month, CPC €1.20, you rank #3.

Calculation Value
Realised 10,000 × 0.110 × 1.20 × 12 €15,840
Potential at #1 10,000 × 0.275 × 1.20 × 12 €39,600
Uplift 39,600 − 15,840 €23,760

Reading: the #3 rank saves you €15,840 a year in ads. Reaching #1 would save €23,760 more.

One row per search intent, always all four. Per row:

Column Calculation
Keywords, Search volume Count and summed volume of keywords with that intent
Your appearances How often your brand appeared in results for those keywords, over every scan in the range
Avg. position sum of positions / your appearances
Share your appearances / appearances of all monitored brands × 100

Example: in the Informational row your brand appears 12 times and all monitored brands 48 times → Share 25.0%.

Keywords where competitors win and you are absent. A keyword qualifies when, in its latest scan, your brand is not in the top 20 and at least one competitor is in the top 5. Rows are ranked by an opportunity score:

opportunity score = search volume × max(CPC, 0.50) × (competitors in top 5 / 5)
  • The CPC floor of 0.50 stops keywords without a paid value from scoring zero.
  • The last factor rewards terms that several competitors already win.
  • Up to three competitors are shown per row. The table shows at most 50 rows.

Example: 5,000 searches, CPC €0.30 (raised to 0.50), two competitors in the top 5 → 5,000 × 0.50 × 0.4 = 1,000.

A heatmap of your 20 highest-volume keywords against the twelve calendar months, from the last twelve months of search-volume history. Each row is scaled to its own peak:

cell = volume that month / highest monthly volume in the row

so 1.0 is that keyword’s busiest month and every other cell is relative to it. Rows are not comparable with each other: the card answers “when does demand for this keyword peak”, not “which keyword is biggest”.

Example: “christmas gifts” has 800 searches in January and 20,000 in December → January renders at 0.04 (shown at the minimum shade), December at 1.0.

One line per brand over time. Per day (per week for ranges longer than 92 days) and per brand:

value = sum over the brand's ranked results that day of search volume × CTR(position)

This is expected clicks, so a rank change on a big keyword moves the line more than one on a small keyword. Days without results show 0.

Each brand’s slice of the expected clicks across every ranked appearance in the date range:

weighted score = sum of search volume × CTR(position)
share = weighted score / sum of weighted scores over all monitored brands × 100

The slices sum to 100%.

Example: brand A ranks #1 on a 10,000-search keyword (2,750) and #4 on a 2,000-search keyword (160), total 2,910. Brand B ranks #2 on the first keyword, 1,520. The pool is 4,430 → A 65.7%, B 34.3%.

Brands ranked by the same weighted score. Columns:

Column Calculation
Share As for SERP Share of Voice
Avg. position sum of positions / appearances, unweighted
Trend Share now minus share in the previous range, in percentage points
Sparkline Daily appearance counts over the last 7 days, regardless of the date range

A bubble chart of up to 200 keywords.

  • Horizontal axis: monthly search volume, on a logarithmic scale.
  • Vertical axis: your best position in the latest scan, reversed so higher is better. Keywords where you are absent from the top 20 are drawn at “absent”.
  • Bubble size: the keyword’s uplift (see commercial value); when there is none, its CPC.
  • Colour: search intent.

Large bubbles far from the top are the quick wins: high demand, real money, and room to climb.

The most recent average monthly searches for this keyword. The card below it plots the last twelve months, marks the peak month, and totals the year.

The average cost per click an advertiser pays for this keyword, as reported by the search engine data. The subtitle shows the low and high top-of-page bid range when available. A high CPC signals competing buyers rather than curious browsers.

The best rank your brand holds in the latest completed scan within the date range. Shown as “—” when your brand does not rank in the top 20.

The same calculation as on the overview, for one keyword. When you rank, the tile shows the realised value with the potential as subtitle. When you do not rank, it shows the potential in amber. Without scan data it shows “—”. Below it, an ads budget estimates what buying the #1 traffic would cost per month:

monthly clicks = search volume × CTR(1)
low estimate = monthly clicks × low top-of-page bid
high estimate = monthly clicks × high top-of-page bid

One line per brand: the best position per scan day. Only days with a completed scan are plotted; a scan that is still running or failed adds no day. A day that was scanned but where the brand did not rank shows a gap, not a zero. Brands that never rank in the date range are left out. Lower is better.

Up to 25 suggestions related to this keyword, each with its search volume, CPC and difficulty, and a flag when you already track it.

The most recent results page, with snapshots from the last 30 days to choose from. The ten organic results are shown in order. Ads are not capped and are grouped above and below the organic block as they appeared on the page. Your own row also shows how many ads sit above it.