Search & Visibility Data Services

Different layer entirely. SERP data covers search engines — Google, Bing, AI Overviews — which is what happens before a shopper reaches a retailer. This service covers what happens inside the retailer's own search box afterwards.

Both matter and they answer different questions. Teams often buy both, joined on brand and product keys, to see the full path from search engine to retailer shelf.

Because sponsored placements occupy a substantial share of first-page slots on major marketplaces, which means blended rank hides how visibility is being achieved. A brand appearing high because it bought the slot looks identical to one earning it, and the cost structures are completely different.

We flag every slot as sponsored or not, count sponsored slots per page, and compute share of shelf both including and excluding paid placement. That makes paid dependency visible rather than flattering.

About 97.8% aggregate, with the residual concentrated in a small number of retailers whose labelling is weaker or inconsistent. We report detection confidence per retailer rather than claiming uniform accuracy.

Retailers also change labelling without notice, which breaks detection until fixed — and this is a maintenance argument rather than a build argument, because an in-house detector degrades silently and produces sponsored shares that drift wrong with no error surfacing.

The absolute position of your first product on the page, as opposed to its organic rank. Organic rank one can be slot four or slot eight depending on how many sponsored placements sit above it.

Since findability depends on absolute position rather than an ordinal within a filtered set, first own slot is usually the more decision-relevant number. We deliver both, along with how many slots sit above your first product.

Yes, explicitly, with an absent flag. Datasets that simply omit rows where a product does not appear hide the most actionable gaps — the keywords where you are entirely invisible.

Absence is a finding rather than missing data, and treating it as a row rather than a gap is what makes the dataset usable for prioritising content and bidding effort.

Yes, and it is one of the faster-growing uses of this service. Sponsored slots are captured with advertiser brand and position across keywords and days, so purchased placement can be checked against what actually appeared and how consistently.

The value is independence: retail media is otherwise verified only by the retailer selling it. Where placement was inconsistent across days or keywords, the data shows it with dates.

Fewer than most teams expect. On-site search behaviour differs from search engine behaviour — shoppers use shorter, more category-like queries — so a tightly chosen set of terms that actually convert outperforms a large set imported from SEO keyword data.

Keyword count is a direct cost multiplier alongside retailers, devices and frequency, so we scope it with you. Starting with dozens rather than hundreds usually produces a more useful dataset for less.

No. We record that a sponsored placement appeared, its position and the advertiser brand. We do not click it.

Clicking would generate a chargeable event for the advertiser and pollute campaign performance data, which would be both unethical and self-defeating — it would corrupt the very measurement clients are buying.

We quote individually, driven by retailers multiplied by keywords multiplied by devices multiplied by frequency. Those four multipliers compound, which makes keyword set discipline the main cost lever.

A focused keyword set across a few retailers on both devices at daily refresh sits at the lighter end. Large keyword sets across many retailers with sub-daily collection sits higher. One scoping call, a free pilot on your own keywords within 48 hours, then a fixed monthly quote. Request a quote.