For Your Industry
The competitor price index is your price divided by a competitor’s price for the same matched product, multiplied by 100: a score of 100 is parity, 105 means you are 5% more expensive, and 95 means you are 5% cheaper. PriceShape treats a band of 95 to 105 as healthy parity, but a single band hides the part that matters: a key value item and a long-tail cable should never share a target. This guide gives the formula, a worked example in euros, and starting target ranges by category and item role, and it shows where the two opposite conventions in circulation trip teams up. tgndata is price intelligence software for retailers and brands, and price indexing is one of its core views.
A competitor price index is one number that expresses your price as a percentage of a competitor’s price, or of a market average, for the same matched product, so that 100 always means parity. It can be calculated for one SKU against one competitor, or rolled up to a brand, a category or the whole assortment. Competera stresses that the index is built from matched products rather than a scan of the full catalogue, which is the single most important constraint to remember: if two products are not truly comparable, the number is meaningless.
The name causes confusion because “CPI” also stands for the consumer price index, which tracks how a basket of goods changes in price over time against a base period. A competitor price index is a cross-section instead: it compares you with the market at one moment, then you track that comparison over time. The statistical idea is shared, since price indices are normally built as weighted averages of price relatives and scaled so the reference equals 100, as the Wikipedia entry on price indices explains. In this article, “price index” always means the competitive version.
A simple vendor example shows the scale. Impact Analytics uses a US-dollar case where your price is 12 and the competitor’s is 10: the index is 120, so you sit 20% above the competitor. Swap in euros and the maths is identical, because the index is currency-neutral. Note that this is one of the few places in this guide where we cite a US-oriented source, because the formula itself has no regional variant.
Why does it matter in Europe in particular? Because shoppers compare. An idealo survey published through Statista in January 2019 found that 37% of German consumers said they always compared prices when shopping online. A separate ChannelAdvisor survey from August 2022, also reported by Statista, found that almost half of German online shoppers usually compare prices on a few sites before buying, and only 2% said they did not compare because they were not price sensitive.
Methodology note: both figures are consumer self-reports from surveys of German online shoppers. The Statista summaries we reviewed did not state sample sizes, and the idealo figure dates from 2019, so read them as directional evidence of comparison behaviour, not as current market measurements.
Takeaway: treat the competitor price index as a ratio of matched prices, not as a general feeling about being expensive.
There are three formulas you need: the SKU-level index, the index against a market average, and the weighted roll-up, and you must pick one direction for the ratio and never mix it. Here they are, with the convention we recommend (your price on top):
Public sources do not agree on direction. DealHub, PriceShape and Impact Analytics put your price on top, so a score above 100 means you are more expensive. Pricefy, Prisync and Competera put the competitor’s price on top, so a score above 100 means the competitor is more expensive and you are the cheaper one. Both are internally consistent. The trouble starts when a pricing manager reads a dashboard built on one convention using a rule of thumb written for the other, and raises prices where they should have cut.
Our own content has to be read with the same care. The tgndata article on the four KPIs every pricing manager should track daily defines the index as your price divided by the average competitor price, while the Price Index Grid documentation in the tgndata knowledge base describes the grid from the competitor’s side, where 90 means the shop is 10% cheaper than you. Whichever you use, write the direction in the dashboard title.
The two directions are not mirror images in percentage terms. If the competitor-over-you version reads 120, the competitor is 20% more expensive than you, but you are only about 16.7% cheaper than the competitor, because 100 ÷ 120 is 0.833. The gap in euros is the same, while the percentage depends on which price you divide by.
| Reading | Index (your price ÷ competitor price) | Index (competitor price ÷ your price) |
|---|---|---|
| You are 5% cheaper | 95.0 | 105.3 |
| Parity | 100.0 | 100.0 |
| You are 5% more expensive | 105.0 | 95.2 |
| You are 20% more expensive | 120.0 | 83.3 |
Methodology note: the table is arithmetic, not survey data. The directions attributed to each vendor come from their public price index pages as retrieved in November 2026, and vendors can change their definitions.
Takeaway: choose “your price on top”, label it everywhere, and convert any outside benchmark to that direction before using it.
In this example the simple average index is 103.8, but weighting by revenue gives 101.8, and the gap exists because low-revenue items, such as a cable priced 15.5% above the market, count as much as best sellers in a simple average. The basket below is illustrative: the prices and revenues are invented to show the mechanics, and they are not tgndata client data. A retailer has matched six SKUs against three competitors and wants to know where it stands against the average competitor price.
| SKU | Role | Your price (EUR) | Competitor A | Competitor B | Competitor C | Average competitor | Index | Revenue (EUR) |
|---|---|---|---|---|---|---|---|---|
| Wireless earbuds | KVI | 76.00 | 75.00 | 82.00 | 80.00 | 79.00 | 96.2 | 24,000 |
| Robot vacuum | KVI | 309.00 | 289.00 | 299.00 | 311.00 | 299.67 | 103.1 | 36,000 |
| Air fryer | Foreground | 99.00 | 89.00 | 95.00 | 98.00 | 94.00 | 105.3 | 15,000 |
| Coffee machine | Foreground | 219.00 | 205.00 | 215.00 | 220.00 | 213.33 | 102.7 | 20,000 |
| HDMI cable | Background | 14.90 | 12.90 | 11.90 | 13.90 | 12.90 | 115.5 | 2,000 |
| Phone case | Background | 17.90 | 16.90 | 18.90 | 17.90 | 17.90 | 100.0 | 3,000 |
Step 1, average the competitor prices. For the earbuds, (75 + 82 + 80) ÷ 3 = 79.00.
Step 2, index each SKU. For the earbuds, 76 ÷ 79 × 100 = 96.2, which means you are 3.8% below the market average, or €3.00 cheaper. For the robot vacuum, 309 ÷ 299.67 × 100 = 103.1, which means you are €9.33 above the average.
Step 3, take the simple average. The six indexes average to 103.8. This number says you are almost 4% more expensive than the market, which could trigger price cuts you do not need.
Step 4, weight by revenue. (96.2 × 24,000 + 103.1 × 36,000 + 105.3 × 15,000 + 102.7 × 20,000 + 115.5 × 2,000 + 100.0 × 3,000) ÷ 100,000 = 101.8. The headline is now 1.8% above market, because the two items that carry 60% of revenue are close to parity.
Bar chart of the competitor price index by SKU in the worked example, with 100 as parity
Three further cuts change the story again. Against each competitor separately, the revenue-weighted index is 106.4 against Competitor A, 100.8 against Competitor B and 98.8 against Competitor C, so the retailer is well above A and slightly below C. Averaging those three gives 102.0, not 101.8, because averaging ratios is not the same as dividing by an average price, which is why you should fix the method and keep it stable. By role, the KVI index is 100.3, the foreground index is 103.8 and the background index is 106.2.
The KVI figure is the one to read carefully. It sits at 100.3, which looks fine, yet it is the average of two opposite problems: the earbuds are 3.8% below market, which may be margin left on the table, and the robot vacuum is 3.1% above market, which may be costing sales. An average inside a role can hide exactly the SKUs that need action.
Methodology note: the basket is deliberately small, so the two KVIs account for 60% of revenue. McKinsey’s guidance is that KVIs should ideally represent roughly 15 to 25% of sales in a full category, so expect a larger share of items and a smaller share of revenue in a real category.
Takeaway: always report the weighted headline, the per-competitor view and the per-role view together, because each one answers a different question.
Weight the index by revenue, units or margin contribution so the products shoppers actually buy dominate the number, and keep a separate index for key value items. Umbrex describes weights as typically revenue, unit volume or a deliberate emphasis on key value items, and notes that some firms maintain separate KVI and non-KVI indices to stop the long tail from skewing the headline. That matches the worked example, where the single low-revenue HDMI cable alone adds about 2.3 points to the simple average.
McKinsey describes a pyramid of item roles. Key value items are often in baskets and drive price perception. Foreground items sit in the middle. Background items are, in McKinsey’s framing, long-tail products that are rarely bought and relatively cheap, so they barely influence how shoppers judge your price level. McKinsey also notes that competitive price indexes are often the main factor in pricing KVIs, but retailers balance them against margin goals, price elasticity, range architecture and market-share targets, and that best-practice retailers refresh their KVI lists at least annually and flex them by price zone or geography.
For a European retailer operating in more than one country, the last point is practical. A KVI in one country can be a background item in another, because brands, comparison portals and shopper habits differ, so keep a separate list and a separate index per market. The tgndata guide on how to prioritize SKUs for competitive monitoring explains how to rank items using revenue, demand and price sensitivity signals.
A workable structure is three indices on one dashboard: a revenue-weighted headline index, a KVI index, and a long-tail index. Report all three against the average competitor and against your two or three closest named competitors. Competera also recommends calculating the index per product, per competitor and for the overall competitive set, which maps neatly onto this structure.
Methodology note: the McKinsey material draws on its retail pricing work and is not specific to Europe. The 15 to 25% figure is stated as an ideal share for KVIs in a category, not as a measured industry average.
Takeaway: one weighted headline plus a KVI index and a long-tail index tells you far more than any single average.
Most wrong index values come from bad matches, mixed price bases and stale prices, not from the formula. The import.io glossary makes the point bluntly: if you do not define the basket, the weights and the matching rules, the index can say almost anything.
Consider a matching error. Your pack of two cables costs €14.90 and gets matched to a competitor’s single cable at €7.50. The index reads 14.90 ÷ 7.50 × 100 = 198.7, and a rule that triggers a price cut at 110 will fire for the wrong reason. This is why product matching quality should be audited before targets are set. The tgndata guides on how product matching works in price monitoring and the pricing data quality guide cover EAN, SKU, title and AI matching and how to set accuracy and freshness thresholds.
Four basis checks keep the index honest:
An outlier rule helps too. Flag any SKU-level index far outside a sensible band, for example below 50 or above 200, for manual review before it enters a roll-up. The band is an editorial suggestion, so adjust it to your category.
Methodology note: the €14.90 versus €7.50 case is a constructed scenario to show the effect of a mismatch, not an observed incident.
Takeaway: fix matching, price basis and freshness first, and the formula will look after itself.
There is no universal good score: PriceShape treats 95 to 105 as a healthy parity range, but the right target depends on your strategy and on the role of each item. That 10-point band is wide enough to contain very different businesses, which is why category and role targets are more useful than one number.
Public guidance on specific targets is thin. PriceShape gives the 95 to 105 parity range. Umbrex offers an illustrative example in its relative price index framework: maintain roughly 98 to 100 against a named discounter on key value items, and 101 to 103 on long-tail items. Ladle describes retailers setting competitive price index targets at store, department and category level, by category role and by item type, and combining each with a gross margin target. McKinsey places the index at the centre of KVI pricing, balanced against margin and elasticity.
None of these sources publishes a measured range by category from a sample of retailers. So the ranges that follow are tgndata editorial planning guidance, built on the logic above, and you should treat them as hypotheses to calibrate, not as industry benchmarks. A reasonable starting point for the overall weighted index, by stated positioning, is:
| Positioning | Overall weighted index (your price ÷ competitor price) |
|---|---|
| Price leader | About 95 to 99 |
| Market parity | About 99 to 102 |
| Premium or value-added | About 103 to 110 |
Inside any of these, set the item role targets below the headline for KVIs and above it for the long tail. In the worked example the retailer’s overall index was 101.8, which suits a market parity position, while its KVI index of 100.3 and tail index of 106.2 follow the right pattern.
Methodology note: the sources above are vendor and consultancy frameworks, mostly with US or global examples, and the Umbrex figures are explicitly illustrative. The positioning table is editorial and has not been validated against a dataset.
Takeaway: set the headline from your positioning, then set tighter targets for KVIs and looser targets for the long tail.
Start with the ranges below, set by category transparency and item role, then adjust them using your own elasticity, margin and conversion data over four to eight weeks. The ranges reflect three drivers: how transparent prices are (comparison engines and marketplaces make prices visible), how precisely products can be matched, and how often shoppers buy the category, which drives price memory.
| Category | Key value items | Foreground items | Background items | Why |
|---|---|---|---|---|
| Consumer electronics | 97 to 101 | 99 to 103 | 102 to 110 | Identical SKUs, high price transparency, volatile prices |
| Grocery and FMCG | 96 to 100 | 99 to 103 | 102 to 108 | Frequent purchase, strong price memory on staples |
| Fashion and apparel | 98 to 102 | 100 to 106 | 102 to 112 | Weaker matching, seasonality, more own-brand items |
| Beauty and cosmetics | 98 to 102 | 100 to 104 | 102 to 108 | Branded SKUs match well, selective distribution limits spread |
| Home and furniture | 98 to 103 | 100 to 106 | 102 to 112 | Low frequency, many exclusive models, fewer matched items |
| Sports and fitness | 98 to 102 | 100 to 104 | 102 to 110 | Mix of branded and own-label, seasonal peaks |
Range chart of starting competitor price index targets by category and item role
Read the table as a shape, not a rulebook. Key value items sit at or below parity in every category because they shape the shopper’s view of your whole price level. Background items sit above parity because shoppers rarely notice them and the margin is useful. Categories with lower matching precision, such as fashion and furniture, get wider ranges, because a smaller share of the assortment can be compared exactly. In those categories, similar-product matching extends coverage, as described in the tgndata guide to price positioning strategy.
To calibrate, change one lever at a time. Pick a group of KVIs, move the index by two points, and compare conversion, units and gross margin against a control group after several weeks. If volume barely moves, your range was too low and you are giving margin away. If volume falls sharply, your range was too high. This is the same logic that makes the index a starting point for elasticity analysis, not a substitute for it.
Methodology note: these ranges are editorial hypotheses, not measured benchmarks, and no retailer sample sits behind them. They use the your-price-on-top convention.
Takeaway: use the category table to open the conversation, then let your own margin and conversion data move the ranges.
A brand may track retailer prices against its recommended retail price, but under EU law it may not use that monitoring to force a fixed or minimum resale price. This matters because brands and manufacturers increasingly run the same index in reverse: your price on top becomes the retailer’s price against the brand’s recommended price.
On 14 October 2025 the European Commission fined Gucci, Chloé and Loewe a combined total of more than €157 million for resale price maintenance. According to law firm summaries from Macfarlanes and Loyens & Loeff, the conduct included requiring retailers not to deviate from recommended prices, capping discount rates and controlling the timing of sales. Herbert Smith Freehills notes that recommended and maximum resale prices are generally permitted, provided pressure or incentives do not turn them into minimum or fixed prices. The earlier 2018 decisions against Asus, Denon and Marantz, Philips and Pioneer, totalling over €111 million according to Bristows, specifically involved manufacturers using monitoring tools to track retailer prices and intervene.
In practice, a brand index is a visibility tool. Use it to see where retail prices drift and why, and keep any follow-up commercial and non-coercive. The tgndata article on MAP monitoring explains monitoring workflows, and you should note that MAP is largely a US concept, so European brands should align policy with competition counsel. This is general information, not legal advice.
Methodology note: the fine amounts and dates come from the law firm summaries named above, which describe the Commission’s published decisions. Check the Commission’s own decision texts before relying on specifics.
Takeaway: index your channel for visibility, and have competition counsel review anything that links the index to pressure on retailers.
Pricing managers should own the formula, the direction and the weights; category managers should own the KVI lists and role targets; ecommerce and marketplace managers should own freshness and matching. A short rollout keeps this practical.
A four-week plan works well. In week one, define the competitor set, confirm matches and fix the formula direction. In week two, assign item roles and revenue weights. In week three, set starting ranges from the category table and configure alerts. In week four, review the first results against margin and conversion, and move the ranges where the evidence supports it.
Takeaway: the index only improves decisions when someone owns the definition, the roles and the data quality behind it.
A common parity range is 95 to 105 according to PriceShape, but a good score depends on positioning and item role. As a starting hypothesis, key value items often sit at or just below 100 and long-tail items above it. The right target is the one that balances margin and conversion for your category.
Neither is better in itself, and the direction depends on the formula you use. With your price on top, a lower score means you are cheaper and a higher score means you are more expensive. Always check which price is divided by which before judging a number.
Either divide your price by the average competitor price, or calculate an index against each competitor and average those. The two methods give slightly different results, 101.8 versus 102.0 in the worked example, so choose one and keep it fixed. Weight the SKUs by revenue or units before you roll up.
The consumer price index measures how the cost of a basket changes over time against a base period. A competitor price index compares your prices with competitors at a point in time, then you track that comparison over time. They share the same index logic, where a reference value equals 100.
There is no universal threshold. Start with the SKUs that account for most of your revenue and the KVIs that drive price perception, and make sure those matches are accurate. McKinsey suggests KVIs should ideally be around 15 to 25% of category sales, which is a useful size for a core monitored set.
Recalculate as often as competitor prices move in your category. Fast-moving categories such as consumer electronics benefit from daily or more frequent refreshes on KVIs, while slower categories can run weekly on the long tail. Stale data is one of the main causes of wrong readings.
We use cookies to provide you with an optimal experience, for marketing and statistical purposes only with your consent, which you may revoke at any time. Please refer to our Privacy Policy for more information.
Missing an important marketplace?
Send us your request to add it!