Share of voice: the metric that survives meetings

Present forty keyword positions to a meeting and watch attention die by row nine. Present “average position improved from 12.4 to 11.1” and watch someone reasonably ask whether that’s good. Both problems have the same root: raw positions don’t aggregate honestly. The metric that does is share of voice.

What it is

Share of voice estimates the fraction of available visibility across your whole keyword set that lands on your domain versus everyone else’s. Instead of averaging positions — which treats a jump from #45 to #40 the same as #6 to #1, though one is worthless and the other is a party — it weights each ranking by how much attention that position actually gets. A #1 contributes a lot; a #9 contributes a little; a #38 contributes approximately nothing, which is exactly what it’s worth.

The result is one percentage. It goes up when you win visible ground, stays flat when you shuffle around in the invisible depths, and — this is the meeting-proof part — it’s directly comparable against competitors measured the same way.

Why it beats the usual suspects

Versus average position: averages get better when you drop a bad keyword and worse when you add ambitious new ones — punishing exactly the behavior you want. SOV is stable under keyword-set growth: new hard keywords start contributing ~0 and grow as you do.

Versus “keywords in top 10” counts: counts treat #1 and #10 as equal, and they’re not even close. Counts also invite gaming with easy irrelevant keywords. A visibility-weighted share is much harder to flatter.

Versus traffic: organic traffic mixes rankings with seasonality, SERP layout shifts, and demand swings. SOV isolates the competitive question — of the attention available, how much are we winning? — which is the part your SEO work actually controls.

Using it without self-deception

• Fix the keyword set before you trust the trend. SOV is relative to the keywords you track; it should be the set that represents your market, not your ego.
• Track competitors on the same set. Your SOV falling while all competitors’ rise says market shift; only yours falling says look inward. The delta is the diagnosis.
• Segment it. SOV per topic cluster or label tells you where you’re winning; one global number hides a product line quietly losing its shelf.

One line going up and to the right, with competitors’ lines for context — that’s a ranking report an executive can act on. The forty rows can live in the appendix, where they belong.