Planogram data and virtual shelf testing: the UK guide

The short answer

Planogram data is the retailer’s record of where every product sits on every shelf in every store. Virtual shelf testing puts a digital copy of that shelf in front of real shoppers to measure what a change would do before anyone moves a product. Range review optimisation is the work of using both to walk into a retailer’s range review with evidence the buyer can act on.

For suppliers to UK grocery, two retailers now make their planogram data available directly. Tesco planogram data covers every Tesco store, every plan and every product, refreshed every Friday. Co-op planogram data is shared with suppliers through a platform Co-op built with Vazen, so suppliers see the same view Co-op’s own category team works from.

This guide explains what that data contains, how Tesco and Co-op planogram data differ, how virtual shelf testing works and how accurate it is, and how category teams combine the two to prepare for a range review. It is written for heads of category, category managers, shopper insight leads and the commercial teams who sit across the table from UK grocery buyers.

What you will find in this guide

What planogram data is, what a single record holds, and why it has been hard for suppliers to get.
Tesco planogram data: coverage, cadence, and how suppliers access it.
Co-op planogram data: what changes when the estate is convenience-led.
What category teams do with planogram data: share of space, compliance, competitor tracking and gaps.
What virtual shelf testing is, and how accurate it is.
How to build a planogram on live shelf data, and why most plans are built on the wrong side of the data.
Range review optimisation and a week-by-week range review timeline.
Planogram data and AI, and how to get answers from planogram data without an analyst queue.
How to choose a partner, a glossary and frequently asked questions.

Five things to know before you read on

The buyer has always had the planogram. Suppliers have usually worked from an older copy, a field audit or a spreadsheet. Direct access puts both sides of the range review on the same data.
Tesco and Co-op data are available to suppliers today. Tesco data arrives weekly through a formal data partnership. Co-op data refreshes continuously as Co-op’s planogram and sales feeds update.
Planogram data tells you what is on the shelf. Virtual shelf testing tells you what shoppers would do if the shelf changed. A strong range review case needs both.
Virtual shelf testing is fast. Results arrive in days, where an in-store trial usually takes months.
It is validated. Vazen’s virtual test results have shown up to 96% accuracy in predicting real shopper behaviour, with studies typically reading out at 90 to 98% statistical confidence.

Vazen (formerly VST, Virtual Store Trials) runs the platform behind both Tesco and Co-op planogram data access for suppliers, and has run virtual shelf testing for brands since 2017, the year after the company was founded in London. Where this guide describes how something works in practice, it draws on that work.

What is planogram data?

Planogram data is the structured record behind a planogram: which products a retailer has placed on a fixture, in which position, with how many facings, in which stores, from which date. A planogram is the picture of the shelf. Planogram data is the dataset that picture is drawn from, and it can be searched, compared and joined to sales.

Every large grocer runs on it. Space and range teams build planograms in planning software, the planograms are allocated to stores, and store colleagues reset the shelf to match. Each of those steps writes data. Put together across an estate, that data describes the physical shelf of the retailer more completely than any audit could.

What a planogram record contains

A useful way to think about planogram data is as a set of nested levels, from the store down to the single facing.

Level	What it describes	Example question it answers
Store and store cluster	Which stores run which plans, grouped by format, size or shopper profile	Which of my plans run in large stores only?
Fixture and bay	The run of shelving for a category, split into sections	How many bays does my category get in convenience stores?
Shelf	Each shelf within a bay, with its height and depth	Which shelf is my hero product on?
Site (position)	Each instance of a product on a shelf	Is my product on one shelf or two?
Facings	How many units of a product are visible, wide and high	How much of the shelf does each product get?
Product	The retailer’s product identifier and attributes	Which of my listed products appear on no plan at all?
Dates	When a plan goes live and when it is replaced	What changed in this category after the last range review?

From those raw fields, category teams calculate the measures they actually argue with: share of space, linear space, distribution by store, position on the fixture, adjacency to competitors, and how any of these changed between one planogram version and the next.

Planogram data versus sales data and shopper data

Planogram data describes supply: what the shopper could see and pick up. Sales data describes outcome: what the shopper bought. Shopper data describes behaviour: how the shopper moved, searched, compared and chose.

Each one on its own leaves a gap. Sales data shows a product declining without saying it lost two facings in half the estate. Planogram data shows a product holding space without saying it earns it. Shopper research explains the decision without saying how often the shelf in question exists. The value appears when they are joined, which is why the rest of this guide keeps returning to all three.

Why planogram data has been hard for suppliers to get

For most of grocery’s history, planogram data lived inside the retailer’s space planning systems. Suppliers saw it in fragments:

Plan extracts. A copy of a planogram, emailed or downloaded, accurate on the day it was sent and out of date soon after.
Proprietary files. Planogram software from the major space planning vendors saves plans in its own file formats, which need a licence and a trained user to open.
Field audits. A field team visits a sample of stores and records what it finds. Useful, but expensive, slow and limited to the stores visited.
Buyer conversations. The buyer describes the plan in a meeting. The supplier takes notes.

The result is an information gap that shapes the whole commercial relationship. In a range review, the buyer works from the live planogram for every store. The supplier works from a partial, dated picture and spends the first part of the meeting agreeing what the shelf even looks like.

What changes when suppliers have direct access

Direct access to retailer planogram data closes that gap. The supplier sees the same view, in the same format, on the same cadence as the buyer. Conversations start at parity, and the time saved on reconciling data goes into the question that matters: what should change on the shelf, and why.

It also changes who can use the data. When planogram data is live and searchable, a category manager can answer a space question in minutes instead of commissioning a piece of analysis. Field teams can target the stores where something has actually gone wrong. Insight teams can join shelf to sales without rebuilding the dataset each quarter.

We have written more about this shift in how planogram data became a new frontier for retailer and supplier collaboration. The next two sections cover the two UK retailers where suppliers can get it directly today.

Tesco planogram data

Tesco planogram data is available to Tesco suppliers through Coplan for Tesco, which Vazen provides through a formal data partnership with Tesco. It covers every Tesco store, every plan and every product, and it updates every week, each Friday. Coplan was formerly called Planogram Publisher, the name dunnhumby lists it under on the Tesco Media & Insight Platform.

That access began in 2022, when Vazen and dunnhumby launched store-level planogram access for brands through the Tesco Media & Insight Platform, covering more than 3,500 stores. Today it is the most complete view of the Tesco shelf a supplier can get outside Tesco itself.

What Tesco planogram data covers

Share of space across the Tesco estate. How much of each fixture your products hold, by store, cluster and format, and how that compares with your share of sales.
Distribution and compliance against listed products. Which of your listed products appear on which plans, and where a listing has not turned into a position on the shelf.
Competitor positioning at the fixture. Where competitor products sit, how much space they hold, and what they sit next to.
Out-of-stocks and gaps by store cluster. Where the shelf shows a gap your product should be filling.
Planogram changes as they happen. What moved between one plan version and the next, so you see the effect of a range review in the data within a week of it landing.

Formats and scale

Tesco runs very different stores under one name: large superstores, mid-sized stores, Express convenience stores in town centres and on forecourts. The same category can run on many plan variants across those formats, each with its own space and range.

This is where store-level planogram data earns its keep. A national distribution figure can say a product is ranged in most of the estate while hiding the fact that it is missing from every convenience plan. Seeing each plan and the stores it runs in shows exactly where a product is present, where it is not, and how the shelf differs by format.

How suppliers access Tesco planogram data

Suppliers access Tesco planogram data inside the Coplan product. Because of the terms of the Tesco data partnership, Tesco data is not available over an API: it stays inside Coplan, where it can be searched, filtered, visualised and combined with the supplier’s own data.

In practice, category teams use it in three rhythms:

Weekly. Check what changed on Friday’s refresh: new plans, lost facings, compliance issues, competitor moves.
Before a meeting. Pull the current shelf for the category and the stores in question, so the conversation starts from the same picture as the buyer’s.
Around a range review. Compare the plan before and after the review, and measure what the change did to space, distribution and sales.

Why Tesco suppliers use it

The short version is parity. The buyer has always had the planogram. Coplan gives Tesco suppliers the same view, in the same format, on the same cadence, so the meeting starts with what to do about the shelf and skips the argument about what the shelf is.

The second reason is execution. Listings get authorised and planograms get distributed, and the fixture in store is sometimes something else. Continuous monitoring means a supplier no longer needs a national audit to find out where compliance has slipped. The data shows it, and the field team goes to the stores that need them. Our guide to planogram compliance covers how to measure and fix it.

Frequently asked questions about Tesco planogram data

How often is Tesco planogram data updated? Every week, each Friday.

Which Tesco formats are covered? Every Tesco store, every plan, every product.

How is the data sourced? Through Vazen’s formal data partnership with Tesco.

Can we get Tesco planogram data through an API? No. The data partnership does not permit sharing Tesco data over an API. It is available through the Coplan product itself.

Co-op planogram data

Co-op planogram data is available to Co-op’s branded suppliers through Coplan for Co-op, a platform Co-op built in partnership with Vazen to change the way it shares shelf data with its supplier partners. Suppliers using it work from the same data Co-op’s own category team works from, refreshed continuously as Co-op’s planogram and sales feeds update.

The partnership was announced in October 2025, starting with a trial group of suppliers before rolling out across the wider supplier base. Co-op’s Commercial Director, Tom Bradley, described the aim as giving brands direct sight of planogram data to remove friction and create transparency.

What Co-op planogram data covers

Share of space across the Co-op estate, by store and by format.
Distribution and compliance against listed products.
Competitor positioning at the fixture.
Out-of-stocks and gaps by store format.
Planogram changes as they happen.

Why convenience changes the picture

Co-op describes itself as the leading UK convenience retailer, and convenience shelves behave differently from superstore shelves. Space is tight, so every facing is contested. Ranges are narrower, so a single delisting removes a brand from a shopper’s choice entirely. Missions are immediate: tonight’s dinner, a top-up shop, something for the walk home.

Co-op’s estate also spans store formats with very different category dynamics. A product can be well placed in larger community stores and absent from the smallest sites. A single-line distribution report hides that. Store-format planogram data shows it as it actually is, which is the level a Co-op buyer thinks at.

The estate is also widely distributed, which makes it hard to audit by field force at speed. Planogram data tells a supplier where compliance has slipped, where listings are missing and where the gaps are, without sending people to hundreds of stores to find out.

How suppliers access Co-op planogram data

Co-op planogram data is available inside Coplan and, unlike Tesco data, through the same APIs as the rest of the Coplan dataset. That means a supplier’s insight or data team can bring Co-op shelf data into its own analysis environment alongside internal sales and forecasting data.

Tesco and Co-op planogram data compared

	Tesco planogram data	Co-op planogram data
Access route	Coplan for Tesco, through Vazen’s formal data partnership with Tesco	Coplan for Co-op, a platform Co-op built with Vazen
Coverage	Every Tesco store, every plan, every product	The Co-op estate, by store and format
Refresh	Weekly, each Friday	Continuous, as planogram and sales feeds refresh
API access	No, available in Coplan only	Yes, through the Coplan APIs
Estate character	Superstores to convenience, many formats	Convenience-led, many small formats
Typical use	Parity in Tesco meetings, compliance across a very large estate	Format-level reality, compliance across a distributed estate

For suppliers who sell into both, the two datasets sit in the same product and the same data model, so a category manager can compare how a range is executed in Tesco Express and in Co-op side by side.

Frequently asked questions about Co-op planogram data

How often is Co-op planogram data updated? Continuously, with updates landing as Co-op’s planogram and sales feeds refresh.

Can we get Co-op planogram data through an API? Yes. Coplan for Co-op data is available through the same APIs as the rest of the Coplan dataset.

How is the Co-op partnership structured? Co-op partnered with Vazen to build the platform that powers Coplan for Co-op. The commercial detail sits inside that arrangement.

What category teams do with planogram data

Category teams use planogram data for five jobs: measuring share of space against share of sales, checking that listings reach the shelf, tracking competitors, finding gaps, and reading the effect of every plan change. Each one turns a question that used to need an audit or an analyst into something a category manager can answer before lunch.

Share of space is the proportion of a fixture a brand or product holds. On its own it is a vanity number. Set against share of sales, it becomes the most persuasive argument in a range review.

A product that sells more than its space suggests is under-spaced, and the shelf is costing the retailer sales. A product holding more space than it earns is a candidate for giving some back. Both are useful. A supplier who walks in with the first case, backed by store-level data, is asking the buyer to grow the category. A supplier who volunteers the second is building the trust that makes the first case land.

Planogram data makes this calculation exact. Space can be measured as facings, linear space (width multiplied by facings wide), or the visible face of the product on the shelf, and it can be cut by store cluster or format, so the conversation moves from national averages to the stores where the imbalance is biggest.

2. Distribution and compliance

A listing is a commercial agreement. A position on the planogram is what makes it real. Planogram data shows which listed products appear on which plans, and which stores run those plans, so a supplier can see where a listing has quietly failed to reach the shelf.

Compliance is the second half: whether the store shelf matches the plan. Continuous planogram data narrows the search. Instead of auditing a random sample of stores, the field team goes to the stores where the data points to a problem.

3. Competitor positioning

Planogram data shows the whole fixture, so it shows every competitor too: how much space they hold, where they sit, what they are next to, and when that changes. A competitor gaining eye-level space in a group of stores is visible within a refresh cycle. So is a new entrant being tested in a handful of plans before a wider rollout.

4. Gaps and out-of-stocks

Where the data shows a gap on the shelf, or a position for a product that is persistently unavailable, the supplier has a precise, store-specific issue to raise with the retailer or fix with the supply chain. Gaps are lost sales for both sides, which makes them one of the easiest conversations to open.

5. Reading the effect of plan changes

Every range review, reset and relay changes the planogram. Planogram data with dates on it lets a category team compare the shelf before and after, then join that to sales to see what the change actually did. This is the evidence base for the next review: what worked last time, in which stores, and by how much.

Beyond the category team

The same data serves other teams in a supplier business:

Field sales plan store visits around the stores where the data shows problems.
Insight teams join shelf to sales and shopper data for analysis that used to start with building the dataset.
Retail media teams check that the product being promoted is actually on the shelf in the stores where the campaign runs.
Supply chain teams see how many facings and shelf positions they are supplying to, store by store.

We wrote about the most common failures this helps catch in the top five shelf execution mistakes and how to avoid them.

What planogram data cannot tell you

Planogram data describes the shelf as it is. It cannot tell you how shoppers would respond to a shelf that does not exist yet: a new layout, a new pack, a new product, a different block. For that, category teams need to put the proposed shelf in front of shoppers. That is what virtual shelf testing does.

What is virtual shelf testing?

Virtual shelf testing is shopper research that puts a digital replica of a real supermarket shelf in front of real shoppers, changes one thing, and measures what they do. The replica is built from the retailer’s planogram, so shoppers face the fixture they would find in store. Results arrive in days, and they show which option wins and why.

It answers the question planogram data cannot: what would happen if the shelf changed? A new product, a new pack, a new block, a different number of facings, a move to eye level, a new range architecture. Any change a shopper would notice at the shelf can be tested.

How a virtual shelf test works

Define the decision. Every good test starts with a decision someone has to make, usually framed as options: keep the current shelf, or move to layout A or layout B.
Build the shelf. The current fixture is rebuilt digitally from planogram data, product by product, with real pack images at the right size and position. This current shelf is the control.
Build the variants. Each option under test becomes its own version of the shelf, identical to the control except for the change being tested.
Recruit shoppers. Panels of real category shoppers are recruited to match the retailer’s shopper profile, and each shopper sees one version of the shelf.
Shop. Shoppers are given a realistic mission and shop the virtual fixture: they search, browse, pick up, compare and choose. Every interaction is recorded.
Ask why. After shopping, shoppers answer questions about what they noticed, what they looked for and why they chose what they did.
Read out. Results compare each variant with the control on the measures that matter to the decision, with the statistical confidence of each result stated.

What virtual shelf testing measures

Measure	What it tells you
Purchase and switching	Whether a change grows the category or moves spend between products, and who gains and loses
Spend and units per shopper	Whether shoppers spend more, buy more, or trade up or down
Penetration	The share of shoppers who buy from the category, brand or product at all
Findability	Whether shoppers can locate a product, and how long it takes
Navigation	How shoppers move across the fixture, and where they get lost
Stand-out	Which packs catch the eye, and which disappear
Reasons	Why shoppers chose, switched or walked away, in their own words

The most useful of these for a range review is switching. Retailers care about the category total. A change that simply moves spend from one brand to another is neutral for the buyer. A change that brings new shoppers in, or gets existing shoppers to spend more, is growth, and virtual shelf testing can tell the two apart.

What you can test

Range changes: adding a product, delisting a product, or changing the range mix, including what happens to the shoppers of a product that disappears.
Layout and blocking: blocking by brand, by type, by need state or by price tier.
Space: adding or removing facings, or moving a product to a different shelf.
Pack and design: a new pack against the current one, seen in context on a full shelf.
New product launches: where a new product should sit and what it will take sales from.
Signage and navigation: shelf-edge labels, dividers and category signage.
Whole-fixture resets: a complete new category layout against the current one.

Why testing on a replica of the real shelf matters

A pack test on a blank background tells you whether a pack is liked. It says little about whether the pack is found and chosen on a crowded shelf of forty competitors. Building the test from real planogram data means the shopper faces the right neighbours, the right number of facings and the right position, so the result applies to the shelf the retailer actually runs.

This is also why planogram data and virtual shelf testing belong together. The better the planogram data, the more faithful the replica, and the more confident everyone can be that the test result will hold in store.

For a deeper introduction, read what shelf testing is and why it is replacing in-store trials. Testing is Vazen’s virtual shopper testing product, and the service the company was founded on.

How accurate is virtual shelf testing?

Well-run virtual shelf testing predicts in-store behaviour accurately. Validated against in-store trials, real launches and actual sales data, Vazen’s virtual test results have shown up to 96% accuracy in predicting real shopper behaviour, and a perfect record to date in identifying which variant wins. Individual studies typically read out at 90 to 98% statistical confidence.

Those two numbers answer different questions, and both matter.

Statistical confidence answers: is the difference between the variant and the control real, or could it be noise? Panels are sized to the question so the answer is clear.
Predictive accuracy answers: does behaviour at a virtual shelf match behaviour at a real one? This is tested by comparing virtual results with what later happened in store.

Why a controlled test can be more reliable than a live one

It sounds backwards, but a virtual test often gives a cleaner answer than a real store. In-store results carry noise a trial can rarely control: weather, local promotions, competitor activity, store-to-store differences, and imperfect execution of the trial itself. Isolating the effect of the one change you made is genuinely hard, and many physical trials read out ambiguous.

A virtual test is a controlled experiment. Every shopper sees the same fixture except for the variable under test, and the current shelf runs as a true control. Behaviour is captured at the level of the individual decision, so the result explains itself: who switched, from which product, and why.

What makes a virtual shelf test trustworthy

If you are commissioning or judging a virtual shelf test, check for these:

A real control. The current shelf, built from current planogram data, tested alongside the variants.
A faithful fixture. Real pack images, true proportions, the right competitors and the right number of facings.
The right shoppers. Recruited to match the category and the retailer’s shopper base, sized to reach a clear answer.
A realistic mission. Shoppers asked to shop as they would, rather than asked what they think of a pack.
Stated confidence. Every result reported with its statistical confidence, so a weak result is labelled as one.
A validation record. Evidence that the provider’s results have been checked against real in-store outcomes.

Virtual shelf testing compared with in-store trials

	Virtual shelf test	In-store trial
Time to result	Days	Months, often around six
Cost	Research budget; a failed variant costs only its share of the study	Operational budget: production, logistics, resets and lost sales from a failed variant
What can be tested	Anything that changes how shoppers interact with the shelf	What a retailer will permit on live trade
Control	A clean experiment against a true control	A noisy read from live trading
Diagnostics	Individual-level behaviour explains the result	Aggregate sales describe the result
Realism	Validated prediction of in-store behaviour	In-store behaviour itself

The least discussed difference is often the decisive one: permission. A physical trial asks a retailer to risk live sales on your idea, so retailers rationally allow only conservative changes. A virtual shelf needs no permission. The ideas with the most upside can be tested, and the evidence is what earns the conversation with the buyer.

The two methods also work in sequence. Virtual testing explores the options cheaply and identifies the winner. If the retailer wants physical proof, an in-store trial runs late, on one variant that has already earned confidence. Our full comparison of virtual shelf testing and in-store trials covers cost, speed and evidence in detail.

How to build a planogram on live shelf data

The strongest planograms are built on the current shelf, tested with shoppers before they are proposed, and developed with the retailer in the same workspace. Most are built the other way round: on an old extract, in desktop software, sent back and forth as files, and judged months later when the sales come in.

Planogram data describes the shelf that exists. Virtual shelf testing measures a shelf that does not exist yet. Planogram software is where that proposed shelf gets built, and it is the step where the most evidence tends to get lost.

Where planogram building goes wrong

Category and space planners describe the same frustrations again and again.

Building against a frozen extract. The plan starts from a copy of the shelf taken weeks or months ago. By the time it is proposed, the shelf has moved and the plan is already out of date.
Specialist desktop tools. The industry-standard space planning applications need installing, licensing and a trained user. The category manager who owns the argument often cannot open the file that holds it.
Files bounced between teams. Supplier drafts a plan, emails it to the buyer, the buyer’s space team rebuilds it in their own system, comments come back in a spreadsheet. Every round trip loses detail and adds a week.
One plan, many stores. A master plan has to be adapted by hand for every store cluster and format, which is slow, error-prone and the first thing cut when the deadline tightens.
No evidence before it ships. The plan goes to store on judgement. Whether it worked is known months later, by which point nobody remembers exactly why each decision was made.

Each of these separates the plan from the data. A range review case built on the most current planogram data in the market still loses force if the proposed shelf is drawn on something else.

What modern planogram software changes

Studio is Vazen’s planogram software, built on the same data foundation as Coplan and Testing. It is designed around the five problems above.

The problem	What Studio does about it
Building on a frozen extract	Plans pull current shelf and sales data straight from Coplan, so every plan starts from today’s shelf
Specialist desktop tools	Browser-based and cloud-native, with nothing to install
Files bounced between teams	Retailer and supplier teams work in one shared workspace, with access controls and audit logs that make joint planning safe across organisations
One plan, many stores	One master plan tailored across the estate, with variants created at store-cluster level
No evidence before it ships	Variants go to shoppers through Testing before they leave your desk

Studio also imports and exports industry-standard planogram formats and offers API access for planogram data in and out, so it works alongside the space planning systems a retailer already runs.

What it looks like in practice

Two results show the pattern.

Princes: +12% category growth on the ambient tomato fixture. Princes rebuilt the fixture from type-led to brand-led across three- and four-bay plans. Studio designed the variants, and the team iterated with the buyer across several sessions until the fixture was right. Read the Princes result.

A confectionery supplier: +15% value growth in chocolate pouches. The pouch fixture had grown organically without ever being properly rethought. Studio restructured it with better brand blocking and clearer format groupings. Read the confectionery result.

In both cases the growth came from the category, which is the result a buyer is looking for in a range review.

The loop that makes it work

The value is in the connection. Coplan shows the shelf as it is. Studio builds the shelf you want to propose on that data. Testing puts the proposal in front of shoppers and shows whether it grows the category. The winning plan goes to the buyer in a form their team can use, and after the reset, Coplan shows whether it landed in store.

Coplan shows the shelf as it is today.
Studio builds the proposed shelf on that data.
Testing puts the proposal in front of shoppers.
The retailer agrees the plan.
The store resets, and Coplan shows whether it landed.

Each turn of the loop leaves evidence behind, which is what makes the next range review easier than the last.

Range review optimisation with planogram data and virtual shelf testing

Range review optimisation is the process of using evidence to shape what a retailer decides in a range review: which products stay, which go, which come in, and how much space each one gets. The most effective suppliers optimise for the category, prove it with shopper evidence, and show it on the retailer’s own shelf data.

A range review is a retailer’s periodic reassessment of the assortment in a category. The buyer’s questions are consistent: is this category growing, is every product earning its space, what is missing, and what would happen to shoppers if something changed? Planogram data and virtual shelf testing between them answer all four.

What the buyer is actually weighing

Buyers are accountable for the category, which shapes every decision in the review.

Category growth over brand growth. A proposal that grows your brand at a competitor’s expense is neutral to the buyer. A proposal that grows the category is interesting.
Space productivity. Every facing has to earn its place. Under-spaced winners and over-spaced passengers are both on the table.
Shopper impact of delistings. The buyer wants to know where shoppers go when a product disappears: to another product in the range, or out of the category and the store.
Evidence they can use. Buyers take proposals to their own leadership. Evidence that is clear, on their own data, and specific to their stores is evidence they can reuse.
Execution risk. A great plan that cannot be executed across the estate is a bad plan.

The range review evidence stack

Think of the case you bring to a range review as four layers, each built on the one below.

Layer	The question it answers	Where the evidence comes from
1. The shelf today	What does the category look like, store by store?	Tesco or Co-op planogram data in Coplan
2. Performance on the shelf	Which products earn their space, and which do not?	Planogram data joined to sales
3. What would happen if it changed	Which proposal grows the category, and why?	Virtual shelf testing against the current shelf
4. The plan to agree	What exactly should the new shelf look like, by store cluster?	A planogram built on live data in Studio

Most suppliers arrive with layer two and a view. The suppliers who win arrive with all four, on the retailer’s own data.

Five ways planogram data optimises a range review

Start from the same picture as the buyer. Bring the current plan for every relevant store and format. The meeting moves straight to what should change.
Find the space imbalance. Compare share of space with share of sales by cluster. Lead with where the category is losing sales to the shelf.
Show what the last review did. Use dated planogram data to compare the shelf before and after the previous change, joined to sales. Evidence of what worked last time is the strongest argument for what to do next.
Expose the execution gap. Show where listed products never reached the shelf. Fixing distribution on the current range is often worth more than any new listing.
Size the opportunity by store. Store-level data lets you estimate what a change is worth in the stores where it would apply, instead of a national average that flatters or undersells it.

Five ways virtual shelf testing optimises a range review

Test the delisting before it happens. Remove a product from the virtual shelf and see where its shoppers go. Transferable demand is the most contested number in any review, and a test measures it directly.
Prove incrementality. Show that a new product or layout brings new spend into the category, and measure how much.
Settle the layout argument. Brand block or type block, price tiers or need states: test the options against the current shelf and let shoppers decide.
Test what the retailer would never trial. Bold ideas rarely get live trial permission. Virtual evidence is what earns the conversation.
Explain the result. Shopper reasons turn “sales went up” into “shoppers found the premium range faster because it was blocked together”, which is the insight a buyer can repeat to their own team.

Where Studio and Ask Vazen fit

Two more steps usually slow a range review team down: building the proposed plan, and answering the questions that come up between meetings.

Studio removes the first. The proposal is built on the current Coplan shelf, varied by cluster, tested in Testing and shared with the buyer’s team in a workspace both sides can open. There is no rebuilding a plan from an old extract a week before the deadline.

Ask Vazen removes the second. “How has my share of space in Tesco shifted this quarter by store cluster?” or “Where are my biggest distribution gaps in Co-op, and which competitors are filling them?” get answered in plain English in seconds, with every answer linked back to the underlying data. The section on getting answers without an analyst queue covers this in more detail.

Range review and retailer sell-in

A range review is one moment in a longer relationship. The same evidence works for any conversation where you are asking a retailer to change something: a brand day, a joint business plan, a new product launch, a merchandising proposal. We cover the conversation itself, including what to say when the retailer says no, in How to win the retailer sell-in.

A range review timeline for category teams

The best range review preparation starts about twelve weeks before the review meeting. That leaves time to read the current shelf, test the options with shoppers, build the plan and rehearse the case. Here is how that time is best spent.

When	Focus	What to do	Tools
12 weeks out	Read the shelf	Pull the current plan for every store cluster and format. Map share of space against share of sales. List distribution gaps and competitor moves since the last review	Coplan
10 weeks out	Frame the decision	Agree the two or three changes worth proposing. Write each as a decision the buyer would have to make	Coplan, Ask Vazen
9 weeks out	Build the options	Build each proposed shelf on current planogram data, varied by cluster where needed	Studio
8 to 6 weeks out	Test with shoppers	Run the options against the current shelf. Read switching, spend, findability and reasons	Testing
5 weeks out	Choose and refine	Take the winning option. Refine it with what shoppers said. Re-test only if the change is material	Studio, Testing
4 weeks out	Build the case	Write the story in the buyer’s terms: category growth, space productivity, shopper impact, execution plan	Ask Vazen
2 weeks out	Pre-empt the pushback	List the three objections the buyer is most likely to raise and the evidence that answers each	Coplan, Testing
Review week	Present	Lead with the category. Show the shelf today, the evidence, then the plan. Leave the plan in a form the buyer’s team can use	Studio
After the reset	Measure	Check the new plan reached the shelf. Compare before and after on space, distribution and sales	Coplan

The timing flexes. Some retailers run category reviews on a fixed annual calendar, others review more often, and a launch or brand day can bring the cycle forward. The order holds either way: read, frame, build, test, refine, present, measure.

Three questions to answer before the meeting

If the preparation has worked, you should be able to answer these in a sentence each, with a number and a source.

What is the shelf costing the category today? Usually a space imbalance or an execution gap, shown on the retailer’s own planogram data.
What would shoppers do if the shelf changed? The tested result, with its statistical confidence.
What exactly should the new shelf look like? The plan, by cluster, ready for the retailer’s space team.

Common mistakes in range review preparation

Starting too late. Six weeks is enough to present data. It is rarely enough to test, refine and build.
Leading with the brand. Buyers listen for the category. A brand-first story invites a brand-first rebuttal.
Using national averages. The decision is made cluster by cluster. Averages hide the stores where the case is strongest.
Arguing about the shelf. If the first ten minutes are spent agreeing what the planogram looks like, the case has already lost time. Bring the same data the buyer has.
Stopping at the meeting. A plan that is agreed but not executed is worth nothing. Measure the reset, and bring that evidence to the next review.

Planogram data and AI

AI is only as useful to a category team as the shelf data underneath it. A general-purpose AI model knows nothing about what is on the Tesco shelf in a particular store cluster this week. Connected to clean, current, governed planogram data, the same technology can answer shelf questions in seconds and draft the analysis a category manager used to wait days for.

Why planogram data suits AI

Planogram data is unusually well structured. Every record has a store, a fixture, a position, a product and a date. Those fields join cleanly to sales and to shopper research. That structure is what lets an AI system answer precise questions reliably, as long as three conditions are met.

The data is normalised. Tesco and Co-op record their shelves differently. Before analysis, the data has to be brought into one consistent model, so a question means the same thing in both retailers.
The data is current. An answer about last quarter’s shelf is a history lesson. Weekly or continuous refreshes keep answers tied to the shelf as it is.
The data is governed. Retailer data comes with terms. Access controls decide who can see what, and every answer should be traceable to the records that produced it.

Vazen’s platform does this work underneath all four products: it ingests retailer data, brand data and open-source signals, normalises and governs them, and makes them available to people and to AI. We wrote about the wider opportunity in the AI opportunity hiding in your planogram data and why the real question for AI in category management is your data.

What AI does well with shelf data today

Answering questions in plain language. Share of space shifts, distribution gaps, competitor moves, compliance issues, by store cluster and period.
Joining datasets in one question. Shelf, sales and shopper evidence combined without an analyst building the join.
Spotting change. Flagging the plan changes, gaps and competitor moves worth a category manager’s attention after each refresh.
Drafting meeting preparation. Turning the numbers into a first draft of the story for a retailer meeting, which the category manager then sharpens.

What AI does not replace

AI does not replace shopper evidence. It can describe the shelf that exists and find patterns in what shoppers bought. It cannot tell you what shoppers would do on a shelf that has never existed. That still needs a test with real shoppers.

It does not replace the category manager either. The judgement about which case to make, and how to make it to a particular buyer, stays human. What changes is how much of the week goes on finding numbers and how much on using them.

How to get answers from planogram data without an analyst queue

The fastest way to get answers from planogram data is to ask for them in plain English and get a traceable answer in seconds. For most category teams the reality is a queue: the category manager has the question, the analyst has the data, and between them sits a backlog of requests and a turnaround measured in days.

The analyst queue, and what it costs

The same pattern shows up in almost every supplier business with good data and a busy insight team.

Every question becomes a ticket. “What happened to our share of space in Express after the last reset?” is a five-minute question and a two-day wait.
Meeting prep takes a week. A retailer meeting needs a dozen small numbers. Each one goes into the queue, and the deck is finished the night before.
Analysts spend their time on routine pulls. The people best placed to do deep analysis spend their week answering the same shelf questions for different category managers.
Follow-up questions die. The buyer asks something in the meeting. By the time the answer comes back, the moment has passed.
General-purpose AI does not help. A public chatbot runs on whatever it can find. It has no access to the retailer’s shelf data, and nothing it says can be traced to a source a buyer would accept.

The biggest cost is the questions that never get asked because the queue is too long to bother.

What conversational analytics changes

Ask Vazen is Vazen’s conversational analytics product. It turns plain-English questions into answers across shelf, sales and shopper data on the Vazen platform, including Tesco and Co-op planogram data in Coplan, planograms in Studio and results from Testing.

The pain	What Ask Vazen does
Every question becomes a ticket	Ask in plain English, with no SQL and no formulas, and get the answer from live retail data in seconds
Meeting prep takes a week	Questions that would queue for an analyst get answered directly, so meeting prep collapses from a week to an afternoon
Analysts stuck on routine pulls	Every query the analyst does not have to write is time back for the analysis that needs a human
Data scattered across tools	Shelf, sales and shopper data in one question
Answers nobody can check	Every answer links back to the underlying records, so the insight team can audit exactly what produced it
Governance worries	Granular access controls, audit logs and data segregation by organisation. Every question and answer is logged

For recurring questions there are answer templates, and for teams that want to build Ask Vazen into their own tools there is API access.

The kind of questions category managers ask

“How has my share of space in Tesco shifted this quarter, by store cluster?”
“Where are my biggest distribution gaps in Co-op, and which competitors are filling them?”
“Which of my products are underperforming relative to the shelf they are getting?”
“What changed in my category on Friday’s Tesco refresh?”

Each of these used to be a request to an analyst. Each is now a question the category manager can ask directly, the evening before a meeting or in the corridor after one.

Why traceability matters more than speed

A fast answer that cannot be checked is a liability in a range review. If a buyer challenges a number, the supplier needs to show where it came from. Ask Vazen runs on structured retail data with lineage back to the source, so every answer can be traced to the planogram and sales records behind it. That is what makes it safe to put an AI-generated number in front of a retailer.

How it connects to the rest of the process

Ask Vazen draws on what the other products produce. Coplan supplies the shelf, Studio the plans, and Testing the shopper evidence. The more of the process runs on the platform, the more questions Ask Vazen can answer, and the more of the range review preparation described above moves from the queue to the category manager’s own desk.

How to choose a planogram data or shelf testing partner

Choose a planogram data partner on the source of the data, its coverage and its freshness. Choose a virtual shelf testing partner on how faithfully it rebuilds the real shelf and how well its results have been validated in store. The best partners do both on one data foundation.

Questions to ask a planogram data provider

Where does the data come from? A formal partnership with the retailer is the only route to complete, authorised planogram data. Ask to see it.
What does it cover? Every store, every plan and every product, or a sample? Which formats?
How fresh is it? Weekly, continuous, or a quarterly extract?
Can you see change over time? Dated plan versions are what let you measure a range review after the event.
Does it join to sales? Space without sales is half an argument.
Is it the same data the buyer uses? Parity is the point. If the data differs from the retailer’s own, the meeting opens with a reconciliation.
How do you get it out? In the product, over an API, or both, and within what terms?

Questions to ask a virtual shelf testing provider

Is the shelf built from real planogram data? Or approximated from photographs and guesswork?
Is there a true control? The current shelf, tested alongside the variants.
Who are the shoppers? How are they recruited, and how is the panel sized?
What does it measure? Switching and category incrementality, or only preference?
How is confidence reported? Every result should come with its statistical confidence.
How has it been validated? Ask for evidence that virtual results matched in-store outcomes.
How fast are results? Days is the standard to expect.

Why one data foundation matters

When planogram data, planogram software, shelf testing and analytics sit on separate systems, every handover is a reconciliation. The shelf in the test differs slightly from the shelf in the data. The plan in the proposal is drawn on a different extract. The analyst reconciles all three before anyone can ask a question.

On one foundation, the loop closes. The shelf that is tested is the shelf in the data, the plan is built on it, and the answers draw on all of it. That is how Vazen is built: Coplan, Studio, Testing and Ask Vazen each stand on their own, and the value compounds when they are used together.

Glossary

The terms category teams meet most often when working with planogram data, virtual shelf testing and range reviews.

Bay. A section of a fixture, usually a run of shelves, used to display and sell products.

Category. A group of products a shopper sees as related, managed by the retailer as one unit, such as laundry, pet food or ambient tomatoes.

Compliance (planogram compliance). How closely the shelf in store matches the planogram it is meant to follow.

Control. In a virtual shelf test, the current shelf, tested alongside the proposed variants so each variant can be compared with what exists today.

Delisting. Removing a product from a retailer’s range.

Facing. One visible unit of a product on the shelf. A product can have several facings wide and several high.

Findability. A shopper research measure of whether shoppers can locate a given product on the shelf, and how quickly.

Fixture. A group of bays displaying products from the same category or subcategory. Sometimes used interchangeably with planogram.

Incrementality. The extra spend a change brings into a category, as distinct from spend moved from one product to another.

Linear space. Shelf space measured as product width multiplied by facings wide.

Listing. A retailer’s agreement to stock a product.

Merchandise group. A set of products or planograms that sit in the same category or subcategory.

Penetration. The share of shoppers who buy from a category, brand or product.

Planogram. A schematic of a fixture showing where every product sits, used to plan placement and maximise sales.

Planogram data. The structured record behind planograms: products, positions, facings, stores and dates, across a retailer’s estate.

Planogram software. Software used to build, edit and manage planograms. Studio is Vazen’s planogram software.

Range review. A retailer’s periodic reassessment of the products it stocks in a category and how much space each gets.

Range review optimisation. Using planogram data, sales data and shopper evidence to shape the outcome of a range review.

Rate of sale. How fast a product sells over a given period.

Share of space. A brand’s or product’s share of the shelf in a category, usually compared with its share of sales.

Shelf-edge label. The label on the front edge of a shelf carrying price and product information.

Site (position). One instance of a product on a shelf. A product on two shelves has two sites.

Store cluster. A group of stores that share an attribute, such as format, size or shopper profile, and often run the same planogram.

Switching. In shopper research, the movement of purchases from one product to another when the shelf changes.

Transferable demand. The share of a delisted product’s sales that moves to other products in the range instead of leaving the category.

Virtual shelf testing. Shopper research that puts a digital replica of a real shelf, built from planogram data, in front of real shoppers to measure the effect of a change before it is made in store.

Frequently asked questions

What is planogram data? (FAQ)

Planogram data is the structured record behind a retailer’s planograms: which products sit on which fixture, in which position, with how many facings, in which stores, from which date. It describes the physical shelf across a retailer’s whole estate, and it can be searched, compared over time and joined to sales.

How can suppliers get Tesco planogram data?

Tesco suppliers can get Tesco planogram data through Coplan for Tesco, which Vazen provides through a formal data partnership with Tesco. It covers every Tesco store, every plan and every product, and updates weekly, each Friday. The data is available inside Coplan and is not shared over an API. Coplan was formerly called Planogram Publisher.

How can suppliers get Co-op planogram data?

Co-op’s branded suppliers can get Co-op planogram data through Coplan for Co-op, a platform Co-op built in partnership with Vazen. It updates continuously as Co-op’s planogram and sales feeds refresh, and it is available inside Coplan and through the Coplan APIs.

Is Tesco planogram data the same data the Tesco buyer uses?

Coplan gives Tesco suppliers the same view, in the same format, on the same cadence as the buyer. Conversations start at parity.

What is the difference between a planogram and planogram data?

A planogram is the diagram of one fixture. Planogram data is the full set of records behind every planogram in an estate, including which stores run each plan and when it changed. One is a picture of a shelf. The other lets you analyse every shelf.

What is virtual shelf testing? (FAQ)

Virtual shelf testing is shopper research that puts a digital replica of a real shelf, built from planogram data, in front of real shoppers. One thing changes between versions, such as a layout, a pack, a product or the number of facings, and the test measures what shoppers do. Results arrive in days.

How accurate is virtual shelf testing? (FAQ)

Validated against in-store trials, real launches and actual sales, Vazen’s virtual test results have shown up to 96% accuracy in predicting real shopper behaviour, with a perfect record to date in identifying which variant wins. Individual studies typically read out at 90 to 98% statistical confidence.

How long does a virtual shelf test take?

Results arrive in days. An in-store trial usually takes months, often around six, once negotiation, production, store resets, trading time and analysis are counted.

Is virtual shelf testing cheaper than an in-store trial?

Yes, typically. A virtual test is a research cost, and a failed variant costs only its share of the study. An in-store trial carries production, logistics and store labour costs, plus the sales lost if the variant underperforms in live stores.

What is range review optimisation?

Range review optimisation is the use of planogram data, sales data and shopper evidence to shape the outcome of a retailer’s range review. It covers which products stay, which go, which come in and how much space each one gets, and it works best when the case is built around category growth.

How far ahead should a category team prepare for a range review?

About twelve weeks. That allows time to read the current shelf, frame the decision, build and test the options with shoppers, refine the winner and build the case. Six weeks is usually enough to present data but rarely enough to test.

How do you prove a delisting will not hurt the category?

Remove the product from a virtual shelf and measure where its shoppers go. If most of its sales move to other products in the range, demand is transferable and the category is protected. If shoppers leave the category, the delisting costs the retailer sales, and the test shows how much.

What is modern planogram software?

Modern planogram software builds and manages planograms in the browser, on live shelf and sales data, with retailer and supplier teams working in the same workspace. Studio is Vazen’s planogram software. It pulls current shelf data from Coplan, tailors one master plan across store clusters, sends variants to shoppers through Testing, and imports and exports industry-standard planogram formats.

Can retailers and suppliers build planograms together?

Yes. Studio supports shared workspaces with the access controls and audit logs needed to plan safely across organisations, so a supplier and a retailer’s space team can work on the same plan.

Can AI answer questions about planogram data?

Yes, when it is connected to clean, current and governed planogram data. Ask Vazen answers plain-English questions across shelf, sales and shopper data in seconds, with every answer linked back to the underlying records.

How is Ask Vazen different from ChatGPT?

A general chatbot runs on whatever it can find and has no access to retailer shelf data. Ask Vazen runs on structured retail data on the Vazen platform, with lineage back to the source, access controls and a full audit log.

Does Vazen cover retailers outside Tesco and Co-op?

Vazen’s connected retailer planogram data covers Tesco and Co-op. Virtual shelf testing and Studio can be used for any retailer’s shelf.

Who is Vazen?

Vazen is a London-based retail data AI platform for consumer goods brands and retailers, known as VST (Virtual Store Trials) until July 2026. Its four products are Coplan for shelf, sales and planogram data, Studio for planogram building, Testing for virtual shopper testing, and Ask Vazen for conversational analytics.

Where to start

If you supply Tesco or Co-op, start with the shelf as it is. Pull the current planogram data for your category, set share of space against share of sales, and list where listings have not reached the shelf. That picture alone changes the next conversation with your buyer.

Then add the evidence the buyer cannot get anywhere else: what shoppers would do if the shelf changed. Build the proposal on live data, test it against the current shelf, and take the winner into the range review with the plan ready for the retailer’s space team.

Vazen runs every step of that on one data foundation. Coplan gives Tesco and Co-op suppliers the same shelf data as the buyer. Studio builds planograms on it. Testing puts them in front of real shoppers. Ask Vazen answers the questions in between, in seconds.

Bring us the change you are weighing up, and we will show you what the evidence looks like on your category, before a single store commits to it.