Tracking Competitor Rankings and What It Can't Tell You

Positions are public, and everything behind them isn’t. For any keyword you track, the whole result page is visible, so you can watch a competitor’s position with exactly the same fidelity as your own. What you cannot see is their traffic, their conversion, their intent, or why anything moved.

That asymmetry defines what competitor tracking is good for: detecting change in the competitive field, not explaining it.

What you genuinely observe

Who occupies the page. The full set of domains ranking for your terms, over time. This is the most underused competitive dataset in SEO and it costs nothing extra, because your tracker already parsed the page to find you.

Arrival and departure. A domain that wasn’t in the top ten last month and is now fourth is a fact, and usually a more useful prompt than your own movement.

Concentration. Whether a keyword set is dominated by a few domains or spread across many. Concentrated sets are harder to enter and more stable once entered.

Which of their URLs ranks. Sometimes the most informative single field. It tells you how they’ve structured their targeting — one hub page for a cluster versus a page per term — and whether they changed that structure.

Feature ownership. Who holds the featured snippet, who’s in the local pack, who appears in the shopping blocks. These are separate competitive races from the organic ordering.

What you’re guessing at

Everything causal. Some specific things people routinely assert and cannot know:

Their traffic. Third-party traffic estimates are models built from keyword data and assumed click curves. They’re useful for rough comparison of very different sizes and are not measurements. Never present one as their traffic.

Why they moved. You can see that a competitor’s page changed, if you were watching, and you still can’t see whether their gain came from the change, from links, from a broad reassessment, or from something you can’t observe at all.

Whether it’s working for them. Ranking is not revenue. A competitor gaining positions on terms that don’t convert has bought themselves nothing, and you can’t tell the difference from outside.

What they’ll do next. Extrapolating a trend line of someone else’s positions is not forecasting.

Picking a competitor set

The useful set is smaller than the obvious one.

Pick by result-page overlap, not by business rivalry. Your commercial competitors and your search competitors are different lists. Track the domains that actually appear on your keywords, which frequently includes marketplaces, publishers, aggregators and forums you’d never call rivals.

Cap it at a handful. Three to five domains, tracked properly, tells you more than twenty tracked nominally — the same attention argument as choosing which keywords to track.

Review it quarterly. The set changes, and a competitor set assembled two years ago is describing an old result page.

Include one domain that beats you everywhere and one that’s roughly level. The first shows the ceiling; the second is where movement is actually meaningful.

Reading it without fooling yourself

Two traps specific to competitive data.

Attributing your own drop to them. If you fell three places and a competitor rose three, that’s one event described twice, not evidence of cause. Someone arriving above you can also be a symptom of the same reassessment that moved you, rather than its cause. Establish your own blast radius first — how to diagnose a ranking drop.

Reading noise as strategy. Competitors’ positions wander for the same sampling reasons yours do — location, device, layout, personalisation removed or not. A competitor “gaining” this week is usually nothing. Apply the same baseline discipline you’d apply to your own series — you can’t read a position without a baseline.

The comparison that’s actually worth reporting

Share of the tracked keyword set, over time, for you and three named competitors. One chart, four lines, same tool, same settings, frozen keyword set.

It won’t tell you why anything moved. It will tell you whether the field is shifting, which is the question competitor tracking can legitimately answer, and it’s a far better use of the space than a table of their individual positions nobody reads. The construction and caveats are in visibility scores and share of voice.

What to actually do

1. Build the competitor set from your own result pages, not from the sales team’s list.
2. Keep it to three to five domains and review quarterly.
3. Record which of their URLs ranks, not just their position.
4. Never present a third-party traffic estimate as their traffic.
5. Report share of the tracked set over time, and say explicitly that it describes change, not cause.