Every term the product uses, in one sentence each.

These are the same words the screens show beside a figure and an exported file carries on its first sheet. 206 terms, A to Z.

#

10 points or more below MRP: The share of priced products whose median price sits 10 percentage points or more under the MRP the app displays. MRP is the app’s own list price, not an independent one. Method

A

A mark a city: One mark for each city, as large as the figure it stands for: the area follows the figure, down to a smallest mark that can still be seen. It is set at the middle of a square about 9 km across that holds the middle of the city’s stores. It is a place for the city, never the place of a store. Method
A paid slot only: A plain keyword where the brand held a paid slot in five or more cities and none of the first 10 unpaid results was its own. Method
Across India: Stores where a product of the brand was read, out of the stores we read, by zone, state, kind of city and city, on one app. Method
Advertisers: Brands that held at least one paid slot in the searches read on that app. Method
Aisle: Our group of categories, the same across apps: dairy, bread and eggs holds curd, milk and paneer. A category sits in one aisle. Method
Alert: A message when a rival cuts price, goes out of stock, restocks or moves position on a shelf you watch. Method
Almost there: The brand came up in at least three in ten of a keyword’s searches, in the first 10 on fewer than half of them, at a usual best rank of 11 to 20. Method
Another brand’s paid slot where you are in the first 3: A plain keyword where your products are usually in the first three unpaid results, you held no paid slot in five or more cities, and another brand held one there. The line names the brand that held it in the most cities. Method
At more than one price: The share of priced products seen at more than one price across the stores read. It says an app prices by place; it does not say by how much. Method
At the most common price: The share of stores carrying the product at its single most common price. Method

B

Below MRP: How far below MRP your products are priced: the median discount in one category on one app. Method
Best unpaid rank: The highest unpaid position any of your products held in a search; across many searches, the middle value. A paid slot has no unpaid rank. Method
Booked keyword: Instamart only: the keyword a paid slot was served on, as the result itself states it, and whether that is the word searched or a wider match. Method
Both on the shelf: A store read to the end where you and the rival are both listed. Method
Brands: Brands with at least one product the read reached, counted only where the app or its catalogue states the brand. A brand read off a product’s name is not counted. Method
Brands beside you: The brands that most often share a first 10 with you on plain keywords. Sister brands of your own company are left out. Method
Brands for each shelf place: The brands of the category on the app, over the products a typical store lists in it: how many brands compete for each place on a store’s shelf. Where an app’s shelf per store is not known, the count of brands stands in. Method

C

Category: Our own list of categories, the same across apps. A native category is the app’s own. Method
Cities carrying it: Cities where at least one store read had it on the shelf at the read. A city with no store read is not counted either way. Method
Cities seen in: Cities where your products were in the unpaid first 10 on searches for this name, out of the cities where the name was searched. Method
Cities with a gap: Cities holding at least one store read to the end where a rival is and you are not. Method
Cities with a paid slot: Cities in which we saw at least one paid slot of the brand, out of the cities where the keywords it held one on were read. Method
City holes: On a keyword where the brand holds a twentieth or more of the first 10 and shows up in at least half the cities read, the cities (five or more) where none of its products came up at all. Method

D

Delisted: Absent from every shelf it sat on across two windows covering a morning and an evening 24 hours apart, all read to the end. Never on Zepto, Swiggy Instamart or Flipkart Minutes. Method
Depth of distribution: Stores carrying this product divided by stores where your categories were read, on one app. Method
Displayed MRP: The list price the app itself shows. Every discount here is against it, not against a market price. Method

E

Empty price point: A band of the price ladder (the price per 100 g, 100 ml or piece, in eight equal bands from the cheapest twentieth to the dearest) that holds no product, or under 3% of them, between two bands that do. A gap on this shelf at this read, not a promise that shoppers want that price. Method
Export: An Excel or CSV of any table, with its reference and every store id, so a number can be traced after it leaves. Files are counted: a plan that includes them allows a stated number of rows a month. Method

F

Few paid slots: A search of the brand’s categories where paid slots were seen and fill under a tenth of the first ten positions. Method
First 10: The first ten unpaid results of a search, in the order the app returned them. Paid results are counted apart. Method
First page: The first ten positions of the ranked slice Instamart served us at the last read. Method
First read: The first read: nothing to compare against, so every product is "first seen" and no absence is claimed. Method
First result paid: The share of searches read whose very first result was a paid one. Method
First seen: The first time we read this product. It is not called a launch, because the previous read could not show it was missing. Method
First seen on our list: The read at which we first saw a product. Not a launch date: the October read reached deeper into every shelf than the reads before it, so most products first seen then were on the shelf already. Method
First unpaid result: Searches in which one of your products was the first unpaid result on the page. Method

H

Hatched: The hatch means the number cannot be known, not that it is zero. Method
Head or long tail: Where a keyword sits in our list: a head keyword is a broad word many shoppers type (milk, chips), a body keyword is narrower, a long-tail keyword is a specific phrase. A judgement made when the list was written, never a count of searches. Method
Held by the top 3 brands: Listings of the three largest brands over all listings of the category on one app. Listings that came with no brand count in the total and for nobody, so it is a minimum. Withheld where under 90% of the category’s listings carry a brand. Method
Held by the top brands: How much of a first 10 the largest brands take, averaged over the keywords of that kind. 100% means one brand holds every slot; 10% means ten brands share them evenly. Method
History we have: How far back your record goes, counted back from today over the reads we have. Nothing is filled in between reads. Method
Holds the most: The brand whose products took the largest share of the unpaid first-10 slots, over every search read for the keyword (or the kind of keyword) on that app, with that share. Method
How crowded: How a keyword compares with the other keywords on the same app: open (the least crowded third), contested, or crowded (the most). Read from the leading brand’s share of the first 10, the first three brands’ share and, where ads were read, how many brands held a paid slot in five or more cities. Method
How crowded the shelf is: How hard the shelf is to get onto, against the other categories on the same app, in thirds: open, contested, crowded. Read from the share of listings held by the three largest brands, the brands for each place on a typical store’s shelf, the app’s own labels, how deep the discounts run and, where the app prints ratings, how much of them the three most-rated brands hold. Listings, not sales. Method
How this is read: We open each app as a guest at a store’s location, type the search, and record the page that comes back: the results in order, and which ones the app itself marks as paid. Method

I

In stock at read: Of the times we looked in this read, the share that found the item buyable. It is the rate at that moment, not a fill rate and not a monthly figure. Method
In the first 3: The share of searches where at least one of the brand’s products sat in the first three unpaid results. Method

K

Kept for later: A rival you ticked beyond what your plan watches. It stays on your account, out of the screens, and is watched when you move to a plan that watches more or swap it in. Method
Keyword finder: Every keyword we read on an app, with how crowded its first 10 is, the signs of demand the app shows, your usual rank and the first five brands with theirs. Our own analysis of the searches read. Method
Keyword gap: A relevant keyword where your products held none of the first-10 slots in any search we read. Method
Keyword landscape: One dot a keyword. Left to right is the share of the first 10 the brand holds; bottom to top is the share of the first 10 that is paid. Two counts side by side, never a forecast. Method
Keywords in the first 10: Plain keywords of this kind where the brand held at least one of the unpaid first-10 slots in the searches read, out of the plain keywords of the kind. Method
Keywords that name it: Plain keywords read on the app whose words name the category (“atta” and “multigrain flour” for atta and flours), with how crowded each one’s first 10 is and its signs of demand. A brand name is never one. Method
Keywords with a paid slot: Keywords on whose searches the brand held at least one paid slot, in any city read on that app. "Tea" and "green tea" are two keywords. Method
Kind of city: Three groups. Tier 1 metro: Delhi, Noida, Greater Noida, Gurgaon, Ghaziabad, Faridabad, Mumbai, Thane, Navi Mumbai, Bengaluru, Hyderabad, Chennai, Kolkata and Pune. Tier 1 non-metro: Ahmedabad, Jaipur, Lucknow, Chandigarh, Kochi and Indore. Tier 2: every other city an app serves. Method
Kind of keyword: The group a keyword is filed under (dairy, snacks, personal care and so on). Every keyword is in one kind, the same on every app. Method
Kind of search: Our keyword list files every keyword under one kind (snacks, beverages, personal care and so on). A brand’s paid slots are added up by the kind of the keyword they appeared on. Method

L

Leader: The brand seen in the most stores of the category on one app at the read; brands level on that count are named together. Method
Least seen: A product’s stores, out of the stores where any product of the brand was seen on that app. Pages stop at a limit, so a product low on a long shelf is not seen there: least seen means not listed or low on the shelf, and is never a count of stores that do not sell it. Method
Listings with a brand: The share of a category’s listings on one app that came with a brand the app or its catalogue states. Below 90%, brand figures are withheld for that category on that app. Method
Location precision: How precisely the app located the store: exact, approximate, area (about 1.8 km) or unknown. On a map a store is drawn as the 1 km square it is in. Every Instamart store is area-level and has no pincode. Method
Looked for, not found: Listed in that store at the previous read. At this read its shelf was read to the end there, and the product was not on it. One read is not enough to call it delisted: a product can come back, or be on another shelf we did not reach. Never said on Zepto, Swiggy Instamart or Flipkart Minutes. Method
Lowest on: The one app whose median price for the product is clearly below every other app that lists it. Two apps within 1% of each other name nobody. Method

M

Map of sizes: Every item drawn as a rectangle whose area is its product count. Depth of ink is one second figure, named under the picture; a hatched rectangle is one where that figure was not read. Method
Map of states: Every state is one square of the same size, set roughly where the state is. A small state with many stores is then as easy to read as a large one with few. Method
Match confidence: How a pack was matched across apps: exact is a barcode; strong, probable, name-only and commodity are brand, pack and name, never shown as exact; confirmed is a product a person checked on each app. Method
Matched on another app: Of the products on the app where the brand has the most, how many we matched to a listing on each other app. The rest are "no match found": a different pack, a different name, or not listed, and we cannot tell which. Method
Matched pack: A group of listings, one on each app, that we are confident are the same pack. Method
Median price: The middle of every price read on this product across stores in this read. Method
Missing where your brand is: Stores that list your brand, where the shelf this product sits on was read to the end, and that do not list this product. Method
Morning or evening window: A reading window: 05:00 to 13:59 IST is morning, 14:00 to 04:59 is evening. Method
Most listings: The brand with the most listings in the category on one app: the widest range on the shelf, not the brand on the most shelves. Named only where its lead over the next brand is larger than all the listings that came with no brand. Method
Movements: Every change we caught between two reads of the same product in the same store: a price cut, a rise, a stockout, a restock, an MRP change, a position move. Method
MRP changed: The displayed list price changed between two reads. Rare, and the steadiest sign of a price change we have. Method
My keywords: Keywords you chose to keep in front of you, from the keywords we search. A tracked keyword is followed on every app that read it. Method

N

Neither brand on the shelf: A store read to the end where neither you nor the rival is listed. Method
New product: A product on a shelf read to the end at this read, and not on the same shelf, read to the end, at the previous read. Method
No match found: We found no listing on that app that we are confident is this product and this pack. That is not proof it is not sold there. Method
Not enough reads yet: We withhold a rate below 20 reads over 5 stores and print what we have instead of a number we cannot stand behind. Method
Not proven either way: We read this store but could not reach the end of its list, so we say nothing about what it lacks. Method
Not seen in the latest read: The product was not seen in the latest read. Said only where that read covered at least 80% of the app’s stores, and never for Zepto, Swiggy Instamart or Flipkart Minutes. It is not the same as delisted: a read takes a limited number of products from each shelf. Method

O

Of: Every number is printed as N of M. M is what we actually read, never what exists. Method
On promo: Share of your listings with an MRP that are 10 points or more below it. Method
On your name: Another brand held a paid slot on a search for your brand name, in five or more of the cities read. It says where the slot appeared, not what that brand chose. Method
One place above: The brand ranked just above you on share of search for this kind of keyword, with its share: the next one to pass. Method
One point a read: A product has a point for a read only where that read saw it in at least 20 stores: its median price over everything we read, the MRP most listings showed, and the share of reads in stock. A read with no point is a break in the line, and not a zero. Method
One product, every app: The listings of one product on different apps, matched as the same product and the same pack, shown as one row under one title. Method
Open 24 hours: The hours the app stated for the store span the whole day. A store with shorter stated hours closes at night; a store with no hours read is in neither group. Method
Openings: Searches of the brand’s own categories where another brand held a paid slot in five or more cities and the brand did not, and searches where paid slots fill under a tenth of the first ten positions. A search with no paid slot at all is never one: an app may not sell slots on it, or the law may keep ads off it. Where a slot appeared, never what it cost or would. Method
Other brands’ keywords: Keywords that are another brand’s name (not its own or a sister brand’s) on whose searches the brand held a paid slot in five or more cities. Several keywords can be one brand’s name. Where a slot appeared, not what the brand chose. Method
Out of stock at the last read: Listings whose most recent read said out of stock; inside one read of the apps they may be days apart. Method
Out of stock by kind of store: For each store, the share of the products seen there that were out of stock at its last read; then the middle store of each kind. Counts set side by side, our own reading, with no cause claimed. Method
Out of stock in search: Products the app itself listed under a search as unavailable at that store. Method
Outside your plan: Stores this plan counts but does not list. Every plan shows the full count. A paid plan lists the stores. Method
Own labels, share of listings: Listings of the app’s own labels over all its listings in categories, from our list of which labels belong to which app. A label missing from that list counts as a national brand, so the share is a minimum. Method

P

Pack sizes: The pack sizes the category is sold in on the app, most products first, with how many brands sell each size and the widest reach one product of that size has. Method
Page as read: The first results of one search, in order, as returned to a guest at the time shown. A page changes between searches; shares over many searches are the steady figure. Method
Paid banner: A banner or sponsored strip above or between the results, counted when the app ties it to a brand. Method
Paid position: Where on the page a paid slot sat, counting every result from the top; across many searches, the middle value. Method
Paid pressure: How much of a keyword’s first ten positions were paid, as a band of our own: no paid slot, lightly paid (under a tenth), paid (under a quarter), heavily paid. The number beside it is the brands named on its paid slots. Method
Paid share of the first 10: Of the first ten positions on the page, the fraction the app marked as paid, across the searches read. Method
Paid slot: A result the app marked as paid for. We count where it appeared; we never see what it cost. Method
Paid slots held here: Cities in which the brand on screen held a paid slot on this keyword, out of the cities read; under it, the other brands you watch that held one in five or more cities. Method
Paid where already in the first 3: Plain keywords where a brand held a paid slot in five or more cities while its own products were already in the first three unpaid results in at least half the searches read. Method
Popularity marks: What an app itself prints on a product about how much it is bought: the number of ratings it shows, and on BigBasket a "sold in a month" badge. Ranked inside a category, one app at a time. Method
Position: Where a listing sat in the ranked slice Instamart served us. Not a search rank. Method
Position moved: An Instamart listing that moved 5 or more places inside the top 30 between two reads of the same shelf. Method
Presence only: Instamart returns a rotating ranked slice, a Flipkart Minutes list is never read to its end, and no reading of Zepto shows a whole shelf, so on those three we can prove you are there ("at least N") and never that you are not. Method
Price across apps: The median price of the same product on each app, over the stores read there in the read. Medians are set side by side; nothing is compared store by store. Method
Price change across reads: From the first read with a point to the latest read with a point, so two products can span different reads. Within 1% is level. Method
Price cut: The displayed price fell by at least one rupee between two reads. Method
Price per unit: The median price divided by everything in the pack: per 100 g, per 100 ml, or per piece. A two-pack is priced on both units together. A pack whose size no app states has no unit price. Method
Price position: Your median product price divided by the category median on one app, times 100. A position, not a discount, never compared across apps. Method
Price rise: The displayed price rose by at least one rupee between two reads. Method
Price vs rival: Your category median divided by the rival’s, times 100, on one app. Category medians, not matched packs. Method
Priced apart: The same product is priced 5% or more apart between the app where it is lowest and the app where it is highest. Method
Prices seen: Distinct displayed prices for one product across the stores read. Method
Print station: The app’s print service answered for this store. Read on Blinkit only; on every other app this is blank, which means not read. Method
Product research: Every category we read on an app, set by how crowded its shelf is beside its signs of demand, then one category in depth and a rival’s range beside yours. For a brand choosing what to launch next, and for anyone studying a market before entering it. Method
Product search: Finds products whose title, brand name and pack carry every word typed, in any order, across every app we read. Titles that say the words side by side come first, then the product seen in the most stores. Method
Product-store listings: One product on the shelf of one store; a brand with 10 products in 100 stores has up to 1,000 listings. Not a physical facing. Method
Products: Distinct products of yours this read reached. What the read reached, not the size of your catalogue. Method
Products a typical store lists: The middle value, over the stores carrying the category, of the products listed on its shelves there. It says how crowded the shelf is where you compete. Method
Products listed, by store: How many products a store’s shelves of the category listed at the shelf read. Where the shelf was not read to its end the count is a minimum, and many stores then sit level at the page cap. Method
Products seen: The different products the shelf read saw at a store. A minimum of what the pages listed, and not the size of the store’s shelf. Method
Products that carry the rank: For each of the brand’s products: the unpaid first-10 slots it held across every search read, the times it was the first result, and the keywords it held them on. Method
Products the read reached: Products of the app that the shelf read saw in at least one store. Pages stop at a limit, so it is what we reached and never the size of the app’s range. Method
Products with a count: Products of the brand for which the app printed a rating count at the read. A product with no count shown is not in this figure. Method
Promised minutes: The minutes the app promised at the store’s own location at the reading: its shortest trip. Not a typical delivery time. Method

R

Range band: Our own split of an app’s fully read stores into thirds by distinct products seen: compact, standard, wide. It ranks what the pages listed, not floor area. Method
Rank among brands: A brand’s place when the brands of one kind of keyword are ordered by share of search. Only brands that held at least three first-10 slots are ranked, and the count of them is printed beside the rank. Method
Rank and paid slots, together: Our own reading of two facts side by side: where a brand’s products rank unpaid on a keyword, and whether the brand held a paid slot there. Method
Rank by stores: Where the brand stands among every brand of the category by the number of stores carrying it at the read, on one app. First is the widest. Method
Ratings held by the top 3: Of every rating the app prints in the category, the share on the three most-rated brands’ products. Ratings accumulate while a product is listed, so it leans to older brands; it is shown only from 20 rated products. Method
Ratings shown: How many ratings the app prints for the product. It grows for as long as a product is listed, so it favours older products: it ranks, it is not a rate of sale. Method
Reaches the store: A wider range (long-tail stock, pharmacy stock) is offered at the store’s own location. It may ship from another store, so the store is reached by it, it does not necessarily hold it. Method
Read: One reading of an app’s stores, start to finish. Every number is from one read. Method
Read reference: The read of the apps these figures came from. Method
Read to the end: We read the whole category list in that store, so a product that is not on it is genuinely absent there at that read. Method
Read to the end, of read: Stores read to the end divided by stores where your categories were read: the share of the shelf on which an absence can be claimed. Method
Read window: The first and the last moment of the read behind these figures. Method
Reads: One look at one product in one store. Every rate says how many it stands on. Method
Reference: The exact set of figures a screen or a file came from. Quote it if you write to us about a number. Method
Relevant keywords: The plain keywords where you hold part of the first 10, or whose first 10 is filled by a category you sell in on that app. This is our own analysis of where a brand belongs, worked out from what the app returned. Method
Restock: A listing that was out of stock at one read and in stock at the next. Method
Restock lag: Hours from a listing going out of stock to coming back, for the stockouts that came back inside the read. Method
Result slots that were paid: Of every result on the pages read, the whole page and not only its first ten, the share the app marked as paid, across the searches read on that app. Method
Room to enter: Our own filter, the same rule for every category: strong signs of demand on a shelf that is open or contested, or moderate signs on an open one. It compares a category with the others on the same app; it is not a forecast of sales. Method
Rows outside your plan: Rows outside your plan are not shown. The counts still include them. Method

S

Same pack, different app: The same pack matched across apps; the median store price on each, compared. Only matched packs count. Method
Same read: A change seen between two looks at one store inside the same read. Real, but not a change from one read to the next. Method
Same-product basket: The products seen in at least 20 stores in both of two reads, and the middle value of later price over earlier price. A basket of what the reads reached: no weights, no sales, not a consumer price index. Method
Searched for, not available: The share of searches of a keyword where the app itself showed products that shoppers searched for and the store could not supply. Shown on the apps that print such a block; blank on the others, never a zero. Method
Searches read: One search is one keyword typed at one store of one city, read as a guest. Every figure here counts searches, cities or result slots, and says which. Method
Searches with a slot: Of the searches read for the keyword, the share where this brand held at least one paid slot. Paid slots rotate, so this is a share over many reads, never one page. Method
Second store: A second store of the same app also serves the location, usually with a wider range and a longer promise. Method
Seen in: Stores where at least one product of the brand was on a shelf we read. A count of presence: the stores it was not seen in are not claimed as absences here. Method
Seen there: Which of the two brands was seen in the place at all in this read: both, only you, only the rival, or neither. Method
Share of ratings shown: A brand’s ratings shown, over all ratings shown for the category’s products on that app. A share of attention to date, not of sales. Method
Share of search: Of all the unpaid first-10 slots across the plain keywords of one kind (dairy, snacks, personal care) on one app, the fraction a brand’s products held. Method
Share of shelf: Your listings in one category on one app divided by all listings in that category on that app, over the stores read. Never summed across apps, and not sales. Method
Share of stores: Stores where we saw you divided by stores where your categories were read, on one app. Method
Share of the first 10: Of all the unpaid first-10 slots across the searches read for a keyword, the fraction your products held. One app at a time, never added across apps. Method
Shelves asked for: Category lists we asked a store for in this read, out of the lists planned for it. Under nine in ten means the app was still being read when the figures were made. Method
Signs of demand: What the apps show of how much a category is wanted, against the others on the same app, in thirds: strong, moderate, light. Read from the share of stores that carry it, how many of its products a typical store lists (the app’s own range decision), and, where the app prints ratings, the ratings shown per rated product. No app publishes sales, so none is shown. Method
Sponsored share of reads: Instamart only: reads that carried an ad marker divided by all reads. Not share of voice, not spend. Method
Spread, at most: When reads span more than one window, two prices may be two stores or two days, so the spread is the most that the difference between stores could be. Method
Status at the reading: What the app answered at the store’s own location at the minute we read it. Paused means it was not taking orders then, not that the store has closed. Method
Still being read: These figures were made while the apps were still being read, so they cover part of this read. They are replaced when it ends. Method
Still out at last read: A stockout with no restock before our last read of that store. It may have come back after we left. Method
Stockout: A listing that was in stock at one read and out of stock at the next. Method
Store role: The lines a store was seen shipping on in any location’s answer: express, long-tail, extended long-tail or pharmacy on Blinkit; main or second store on Zepto and Instamart. Method
Stores carrying it: Stores where a shelf of the category listed at least one product at the shelf read. A minimum. Method
Stores carrying the category: Stores where at least one product of the category was read, out of the stores we read, by zone, state, kind of city and city. Method
Stores in its base: For one product: the stores that list your brand and where the shelf this product sits on was read to the end. Every count for the product is over these. Method
Stores moved together: Stores where one product moved the same way in one window, over stores carrying it. Only a count from one window says the stores moved together. Method
Stores on our list: The stores we set out to read this time, whether or not an answer came back. Method
Stores on record: Store ids we have for the app: those on our monthly list of stores, plus any the shelf read met that the list does not have. Method
Stores read: Stores where we got a shelf back this read. Counted for a brand, they are the stores where a shelf of its own categories came back, the base of its shares; counted for an app (Market), every store where any shelf came back, so the app’s figure is the larger. Method
Stores where you were seen: Stores where at least one of your products was on the shelf at a read. Method
Stores with shelves read: Stores where the shelf read got an answer. Every count of a brand by kind of store is out of these, never out of every store on record. Method
Street map: Each store drawn as the 1 km square it is in, over the streets around it. It is loaded only when you ask for it. No latitude or longitude is published. Method

T

The app’s own label: A brand the app itself owns. Marked so a house label is not mistaken for a rival advertiser. Method
The app’s sold badge: BigBasket prints a badge such as "150+ sold (1 mo)" on some products. We show it in the app’s own words. A badge printed in this read ranks a product by the minimum it states; one last printed before this read is shown with that day and ranks nothing. Method
The category’s range: On one app, the unit prices of the products filed under the same category: the bar runs from the price a tenth of them are under to the price a tenth are over, with the middle marked. Never set against another app’s range, and only shown per 100 g or 100 ml. Method
The same products on two apps: Two apps are compared only on products matched as the same product and pack on both, by median price over the stores read on each. We count how many are more than 1% lower on one, on the other, or level. It is a count, not a saving, and the matched products are a small part of either range. Method
The shelf, by category: Every product we read at this store, grouped by the category of the shelf it was read on, with the price and stock shown at that read. Pages stop at a limit, so a category’s count is a minimum. Method
The usual price: The price the most stores showed for this product on this app, at each store’s last read. On the map a store below it is green and a store above it is red, as a price cut and a price rise are. It is not a list price. Method
Their categories, not yours: Categories where a brand has products on the app and the brand set beside it has none, at the latest read. One app at a time. Method
Thin shelf: A city where the app carries the category in a share of its stores at least 15 points below its share across India, over at least 10 stores read. A store that does not carry it may still deliver it from another store. Method
Thinner here: Our own reading: a kind of store where the brand was seen in a clearly smaller share of stores than in its opposite. Both groups hold at least 30 stores and the shares are at least 10 points apart. Method

W

Watching: A rival your plan follows on every screen: its stores, prices, search rank and paid slots, beside yours. Each plan watches a set number at a time, and you can swap which once in 30 days. Method
What to do, in search: Our own reading: keywords where the searches we read suggest something to do, ranked by a fixed rule. Every line is counts from this screen; nothing is modelled. Method
Where a rival is and you are not: A store whose category shelf we read to the end, where a rival is on it and you are not. Method
Where you sit: Ten cells for the first ten unpaid results. A cell is as dark as the share of searches where your best result sat in that slot. Method
Who ranks on it: The five brands with the largest share of the unpaid first 10 on this keyword over the searches read, each with its usual best rank and its share. Method
Wide listing: A product listed in at least half of its base. A hole in one of these is a store to raise; a product in fewer is a narrow listing, missing from most of its base by design. Method
Worst city: The city with the most stores where a rival is and you are not, with the stores it is counted over. Method
Worth a try: Our own filter, the same rule for every brand: plain keywords in your main categories (where at least a tenth of the keywords you hold a first-10 slot on are), open or contested, with moderate or strong signs of demand, where you are not usually in the first 3. Most promising first. Method

Y

You against your rivals: Your paid slots beside those of each rival you watch, on one app: slots seen, keywords, and the cities with a slot out of the cities read. Counts of what was on the page, never spend. Method
You are and they are not: A store read to the end where you are on the shelf and the named rival is not. Method
Your categories: The categories holding at least 3% of your listings on that app. They set which shelves count for you. Method
Your list: Products you chose to keep in front of you. A product on your list is followed on every app it is matched on. Method
Your own name: Cities read in which your brand held a paid slot on a search for its own name. Method

Z

Zone: One of the six groups of states the Zonal Councils and the North Eastern Council use: North, Central, East, West, South and North-East. The grouping is theirs, not ours. Method

Keep reading

Where these terms are used.

See the shelf, every day.

Every store read every day. Pick a category and see who leads it, what sells out, who ranks in its searches and who pays for its ads. Free.

Start free Open Shelf Talk to us

No card and no seller login needed.