Digital shelf analytics: five metrics, read by pin code

The short answer

Digital shelf analytics measures how a product shows up on the apps and sites that sell it: is it listed, is it in stock, is it priced as intended against its MRP, where does it rank in search, and is the listing described correctly. Those five readings, repeated and compared over time, are what the term means in practice.

On quick commerce each of the five changes by pin code, because each order is served from a nearby dark store. So the useful reading is per pin code, per app and per day, not one national number. To collect the readings yourself, follow the digital shelf audit guide, which this page supports.

What the digital shelf is

The word shelf is borrowed from a physical store on purpose. A brand that would never accept a bare shelf or a wrong price tag in a supermarket often has no such view of Blinkit, Zepto or Amazon. There, the shelf is different for every app, every pin code and every hour.

The digital shelf is what one shopper sees for your product at one moment: whether it appears for the words they typed, whether it can be added to a basket, the price beside the MRP, and the title, images and pack size on the listing. Analytics is the discipline of reading that view in a fixed, repeatable way, so that a change can be told apart from noise.

The five metrics, in order

The five are read together because each one only makes sense beside the others. A good rank is worth little on a product that cannot be bought, and a sharp discount says little if the listing shows the wrong pack. Read them in this order, because each depends on the one before.

Listing comes first: is the product on the app at all, at that pin code? A product missing from the catalogue is a different problem from one that is listed and out of stock, and it needs a different owner.

Availability is whether the product could be added to a basket at that pin code and that moment, as a shopper would see it. It is read per dark store, not averaged across a city, and a product listed but unavailable is not the same as one never listed. The guide to listed but unavailable covers the difference.

Price is the selling price read together with the MRP, so a discount can be read against the MRP it was calculated from. Discount depth explains how to calculate it and why the same pack can show a different depth at another pin code or on another day.

Search rank is where the listing sits for the terms a shopper would type, read from the unpaid results only, with sponsored slots recorded apart. Organic rank on Blinkit and Zepto sets out how to record it.

Content is whether the title, images, pack size and description are present and match the same product on other apps. It changes far more slowly than price and stock, so it needs reading less often.

Why the pin code matters

On quick commerce, each dark store serves a small area, so the shelf differs between nearby pin codes. The same product can be in stock in one pin code and not in the next, and a national in-stock rate can look healthy while specific neighbourhoods are empty. Pin-code shelf truth explains why a reading without its pin code is hard to act on.

That is why every reading in digital shelf analytics carries four things beside the value: the app, the pin code, the product and the time it was read. Two readings can only be compared when those match, or when the difference between them is the thing you set out to measure.

Measures built on top of the five

Several measures that teams report are built from the same readings. They are useful when their definition is written down and kept the same from one round to the next.

• Digital share of shelf: the portion of the visible results for a search term that your products fill, at one pin code and time.
• Share of search: the portion of category searches that name your brand, a search-box measure distinct from share of shelf.
• First own slot: your first position on the whole screen, sponsored tiles included, which is not the same as organic rank one.
• Rank drop or delisting: whether a product that fell out of view is still in the results lower down, or has gone.

How it differs from a shelf audit

The two terms overlap. A digital shelf audit, as the audit guide describes it, is the method a team runs itself: a fixed list of products, apps and pin codes, read by hand and repeated on a regular day. Digital shelf analytics is the set of measures and the comparison over time, whoever collects the readings and however often.

In a physical store, planogram compliance checks that products sit where they were agreed. The closest digital counterpart is a mix of content and search: is the listing complete and correct, and does it appear for the terms that matter? The difference is that the digital version has to be read per app and per pin code, because no two shoppers see quite the same shelf.

Mistakes that make a reading wrong

Most errors come from the reading rather than the analysis. Four are common enough to check every round.

• The wrong location: an app keeps the last delivery address it was given, so a reading taken without resetting it describes a different pin code.
• A different pack: the same product often has several pack sizes and multipacks, so record the pack you meant and treat a missing pack as missing.
• Counting ads as rank: a sponsored slot can sit above every organic result, so count organic position apart from the slots marked as ads.
• Reading at different times: stock can change between morning and evening, so two rounds taken at different hours can disagree with no real change.

What digital shelf analytics cannot tell you

It records what a listing showed, not why. It cannot say whether a rival funded a discount, why an app ranked a listing where it did, or whether a stockout cost a sale. A reading can inform those questions but not answer them alone.

It is also not a sales or margin figure. A product that is in stock, correctly priced and well ranked can still sell poorly for reasons the shelf does not show. Keep the evidence about what a shopper was shown apart from claims about outcomes.

Where the readings can come from

Reading by hand is real evidence and the right place to start. It stops scaling at a few dozen products and pin codes, and it is easy to drift from one round to the next. The alternative is to start from records someone else has already read and dated.

Rewalto sells dated readings of prices, stock and search positions by app and pin code. Check the fields, coverage and freshness for your own list in commerce data before you plan around them, and see digital shelf analytics for how teams use them. If the readings feed an ad budget, read Blinkit and Zepto ads: five checks before you spend.

Common questions