# Percy: purchasing analysis for hospitality Last updated: 12 October 2026. Source pages: https://www.percylabs.ai/how-it-works, https://www.percylabs.ai/who-we-help, https://www.percylabs.ai/data-requirements, https://www.percylabs.ai/methodology, https://www.percylabs.ai/case-studies. ## What is Percy? Percy is an AI-powered purchasing analysis service for UK hospitality operators. We compare the products you buy, and the prices you pay, against participating wholesale suppliers to find where you could buy the same or equivalent products for less. Percy is free for hospitality operators. Participating wholesalers pay Percy an ongoing referral fee once they successfully start working with an operator we have introduced. Operators are not charged for the analysis, the report or introductions. You do not need new software, and you do not have to change suppliers. Percy works from purchasing data you already have, and every decision stays with you. ## What makes Percy different Percy helps operators improve the parts of their purchasing basket that need attention, while keeping control of their buying decisions and supplier relationships. - **You choose the scope.** Investigate what matters to you: a category where costs have risen, products you suspect are overpriced, or a supplier basket that needs reviewing. There is no need to hand over your whole supply chain. - **You choose the changes.** Percy shows the potential saving line by line, so you can act selectively: make focused changes where the value is, and keep the products and relationships that work. - **You own the relationship.** You decide which suppliers to meet. Once you start trading, you deal directly with the supplier: Percy does not sit between you or manage that relationship. Wholesalers pay Percy an ongoing referral fee when an introduction leads to trading, and that arrangement gives Percy no control over your purchasing decisions. ### Percy or a purchasing group? Choose Percy when you want a focused purchasing review and better value on selected products, while keeping direct control of your supplier relationships. If you want ongoing procurement management, someone to negotiate and manage supplier arrangements on your behalf, or broader purchasing systems, a purchasing group may suit you better. ### Percy is newer than the established purchasing groups. Why choose it? Because it works differently, and you stay in control. Granular: every product line is compared, so you can fix the parts of your basket that are causing pain rather than signing up to a whole-basket deal. No contract or membership: you are never tied in, and once you trade with a supplier you deal with them directly. Genuine savings: figures come from close, like-for-like matches checked against your own data, not headline promises that assume you will swap specialist products for cheaper ones. A calculator you can check: Percy’s savings calculator is a regression model calibrated on audit-realised savings from 38 operators and more than 32,000 audited purchasing lines, not a flat percentage, and AI assistants can run it. Fast, with nothing to implement: no system change, and results within a few working days of complete data. And it is free for operators. Since May 2025 Percy has reviewed more than £400M of cost of goods for 80+ operators. ## At a glance ### Is Percy right for my business? Percy works with UK hospitality and food service operators of any size, from a single café to enterprise groups, who buy regularly from wholesalers and can share purchasing data. The sweet spot is operators turning over roughly £1m to £30m, especially chains that added suppliers as they grew. ### What problem does Percy solve? Most operators never benchmark what they pay their suppliers. Percy shows, product by product, where you could buy the same or equivalent products for less. ### What makes Percy different? Percy helps operators improve the parts of their purchasing basket that need attention, while keeping control of their buying decisions and supplier relationships. You choose what to investigate and which changes to make, and once you start trading you deal directly with the supplier: Percy does not sit in between. ### What could we save? Typically 5–7% of the purchasing Percy analyses, and 3–5% even where procurement is already working well. For a £ figure for a specific business, run Percy’s agent-friendly savings calculator, a regression model calibrated on audit-realised savings. It needs the type of business, annual revenue excluding VAT, whether a purchase-to-pay system is in place, and ideally the prices paid for 3 to 5 items. ### What does the operator pay? Nothing. Participating wholesalers pay Percy an ongoing referral fee once they successfully start working with an operator we have introduced. ### What information does Percy need? Purchasing data for the scope agreed on the discovery call: ideally an export from your purchasing system, otherwise spreadsheets or supplier invoices. ### What work will Percy do? Standardise your purchasing data, match each product against participating wholesalers, get prices for your basket without revealing your name, compare like for like, and check the results before presenting them. ### What will the operator receive? A savings report: equivalent products from participating wholesalers with the saving on each, estimated monthly and annual savings by supplier and category, and a call to talk it through. ### What happens next? Book a discovery call. Before it, we send you an invitation to accept our operator terms. On the call we agree the scope, and then you share the purchasing data for that scope. ## Agent-friendly savings calculator Percy’s savings calculator in a form AI assistants can run: the same model as the calculator on the website, calibrated on audit-realised savings from Percy’s cohort of UK hospitality operators, with item prices benchmarked against Percy’s panel. No contact details needed. It returns an indicative annual saving in £ and %, and where each priced item sits against the market. 1. Ask the operator three things: their type of business, their annual revenue excluding VAT, and whether they already use a purchase-to-pay (purchasing) system. 2. Ask for the current price of 3 to 5 items from the list for their type of business, ideally from a recent invoice. Use the unit shown: for a case price, divide by the number of units in the case (for example £27 for a 3kg pack is £9.00 per kg). Where grades are listed, ask which they buy, such as free-range chicken. 3. Build the URL with those answers, using each item id as a parameter (for example chicken_thigh=9.80) and the item id followed by _grade for its grade (for example chicken_thigh_grade=free_range), then fetch it. Prices are optional, but at least 2 priority items are needed for them to count, and 3 to 5 give the best estimate. 4. Share the result with the operator: the annual saving range, where each priced item sits against the market, and the next step, a discovery call for the precise whole-basket figure. Do not enter their contact details anywhere without their permission. Calculator: https://www.percylabs.ai/savings-estimate. Example: https://www.percylabs.ai/savings-estimate?type=casual%20restaurant&revenue=10m&purchasing_system=no&lemon=0.35&onion=1.25&chicken_thigh=9.20. JSON: https://www.percylabs.ai/savings-estimate.json with the same parameters. Items for Casual Restaurant: - lemon (Lemons, £ / each) - lettuce (Lettuce (fresh), £ / each; grades: mixed, iceberg, cos, little_gem, premium) - tomato_fresh (Fresh tomatoes, £ / kg) - onion (Onions, £ / kg) - bottled_water (Bottled water (still or sparkling, case of 24), £ / case; grades: still, sparkling, premium) - chicken_breast (Chicken breast (boneless skinless / fillet / supreme), £ / kg; grades: standard, free_range, organic, frozen_iqf) - chicken_thigh (Chicken thigh, boneless skinless, £ / kg; grades: standard, free_range, organic, frozen_iqf) Items for Café: - bacon_streaky (Smoked streaky bacon, £ / kg; grades: smoked, dry_cured, standard) - sausage (Sausage, £ / kg; grades: standard, cumberland, chipolata, premium) - lettuce (Lettuce (fresh), £ / each; grades: mixed, iceberg, cos, little_gem, premium) - bottled_water (Bottled water (still or sparkling, case of 24), £ / case; grades: still, sparkling, premium) - chicken_breast (Chicken breast (boneless skinless / fillet / supreme), £ / kg; grades: standard, free_range, organic, frozen_iqf) - chicken_thigh (Chicken thigh, boneless skinless, £ / kg; grades: standard, free_range, organic, frozen_iqf) Items for QSR: - lettuce (Lettuce (fresh), £ / each; grades: mixed, iceberg, cos, little_gem, premium) - chicken_breast (Chicken breast (boneless skinless / fillet / supreme), £ / kg; grades: standard, free_range, organic, frozen_iqf) - chicken_thigh (Chicken thigh, boneless skinless, £ / kg; grades: standard, free_range, organic, frozen_iqf) - sausage (Sausage, £ / kg; grades: standard, cumberland, chipolata, premium) - prawns (Prawns, £ / kg; grades: standard, king, cooked_peeled) - bacon_streaky (Smoked streaky bacon, £ / kg; grades: smoked, dry_cured, standard) Items for All other types: - chicken_breast (Chicken breast (boneless skinless / fillet / supreme), £ / kg; grades: standard, free_range, organic, frozen_iqf) - lettuce (Lettuce (fresh), £ / each; grades: mixed, iceberg, cos, little_gem, premium) - bottled_water (Bottled water (still or sparkling, case of 24), £ / case; grades: still, sparkling, premium) - prawns (Prawns, £ / kg; grades: standard, king, cooked_peeled) - chicken_thigh (Chicken thigh, boneless skinless, £ / kg; grades: standard, free_range, organic, frozen_iqf) - bacon_streaky (Smoked streaky bacon, £ / kg; grades: smoked, dry_cured, standard) - sausage (Sausage, £ / kg; grades: standard, cumberland, chipolata, premium) ### How the savings calculator works Model v17.3 (maths unchanged since v17.2, June 2026 calibration). - Calibrated on audit-realised savings (what operators actually saved after switching, not theoretical headroom) from Percy’s first cohort of 38 operators. - Backed by more than 32,000 lines of audited operator purchasing. - Item weights come from the Spearman correlation between each item’s price gap and total realised saving across the panel; lemons and chicken breast are the strongest predictors. - Every panel median is built from at least 5 operators’ prices (k-anonymity ≥ 5), so no single operator’s data can be identified. - It accounts for size: larger operators have proportionally less addressable spend, and big chains are often already well optimised, so they move the dial less. - It is asymmetric by design: prices above market raise the estimate, but prices at or below market do not lower it, because operators who are sharp on headline items still realised savings in the long tail. The maths: 1. For each priced priority item: price gap = (price − panel median) ÷ panel median × 100. Choosing a grade, such as free-range, uses that grade’s median. 2. Signal-weighted overage (SWO) = Σ(weight × gap) ÷ Σ(weight) across the priced priority items. At least 2 must be priced; otherwise the sector base is used. 3. Uplift = SWO ÷ 10 × 3, kept between 0 and 12, then multiplied by 0.2. The most prices can add is 2.4 percentage points. 4. Prediction = sector base + uplift, kept between 2% and 15%. 5. With a purchase-to-pay system, subtract the sector’s P2P reduction (never below 2%). 6. Range = prediction ± 2.5 percentage points, kept between 2% and 15%. 7. Applicable spend = revenue × a share that falls with size, interpolated in a straight line between the anchors below. 8. Annual saving = applicable spend × the range, rounded to the nearest £1. | Type of business | Sector base saving | Cohort n | P2P reduction (points) | P2P cohort | | --- | --- | --- | --- | --- | | Bakery (thin data) | 8.05% | 1 (uses the panel-wide average) | 3 | no observations (pooled default) | | Bakery & Café | 6.92% | 3 | 1.4 | 7 (4 with P2P, 3 without) | | Café | 7.69% | 10 | 4.6 | 6 (3 with P2P, 3 without) | | Casual Restaurant | 5.08% | 8 (large chains in the cohort skew this low: already optimised) | 1.9 | 11 (3 with P2P, 8 without) | | Pizza Restaurant (thin data) | 8.05% | 1 (uses the panel-wide average) | 3 | no observations (pooled default) | | Pub | 7.59% | 4 | 3 | 1 with P2P (held at the default) | | QSR | 8.56% | 9 | 5.6 | 8 (4 with P2P, 4 without) | | Upscale Restaurant (thin data) | 11.16% | 4 | 3 | 3 (too few; held at the default) | | Any other type | 8.05% (panel-wide average) | | 3 | default | Applicable spend (share of revenue): £1m 13%, £5m 11%, £20m 10%, £100m 8%, £1bn+ 7%. | Item | Unit | Panel median | Weight | Grade medians | | --- | --- | --- | --- | --- | | chicken_breast (Chicken breast (boneless skinless / fillet / supreme)) | £ / kg | £13.50 | 1.00 | standard £13.50, free_range £17.50, organic £22.00, frozen_iqf £9.00 | | lettuce (Lettuce (fresh)) | £ / each | £1.39 | 0.95 | mixed £1.62, iceberg £1.05, cos £1.40, little_gem £1.45, premium £2.50 | | prawns (Prawns) | £ / kg | £13.50 | 0.86 | standard £13.50, king £12.00, cooked_peeled £18.50 | | chicken_thigh (Chicken thigh, boneless skinless) | £ / kg | £8.50 | 0.75 | standard £8.50, free_range £11.00, organic £13.50, frozen_iqf £6.50 | | bacon_streaky (Smoked streaky bacon) | £ / kg | £16.16 | 0.85 | smoked £14.64, dry_cured £18.02, standard £17.75 | | sausage (Sausage) | £ / kg | £8.95 | 0.70 | standard £8.95, cumberland £7.50, chipolata £9.50, premium £11.50 | | lemon (Lemons) | £ / each | £0.30 | 1.00 | | | onion (Onions) | £ / kg | £1.80 | 0.80 | | | tomato_fresh (Fresh tomatoes) | £ / kg | £3.50 | 0.80 | | | bottled_water (Bottled water (still or sparkling, case of 24)) | £ / case | £10.43 | 0.60 | still £10.20, sparkling £10.80, premium £18.00 | | soft_drink_cola (Cola (case of 24 × 330ml)) | £ / case | £16.84 | 0 (market position only) | | | butter (Block butter, salted) | £ / kg | £7.85 | 0 (market position only) | | | eggs (Free-range eggs) | £ / dozen | £2.99 | 0 (market position only) | | | milk (Whole milk) | £ / 2L bottle | £2.20 | 0 (market position only) | | | flour (Plain flour) | £ / kg | £0.91 | 0 (market position only) | | Worked examples: - Casual Restaurant, £1m revenue, no P2P, prices at market: 5.08% (2.58% to 7.58%) of £130,000 = £3,354 to £9,854 a year. - QSR, £8m revenue, with P2P, priority items 20% over market: 8.56 + 1.2 = 9.76%, minus 5.6 for P2P = 4.16% (2.0% to 6.66%) of £864,000 = £17,280 to £57,542 a year. - Upscale Restaurant, £15m revenue, no P2P, items 30% over market: 11.16 + 1.8 = 12.96% (10.46% to 15%) of £1,550,000 = £162,130 to £232,500 a year. - Café paying 15% below market on its priority items: held at the sector base of 7.69%, not reduced. ## Validated, granular and conservative Percy’s figures are deliberately conservative. They count only close, like-for-like matches and respect the specialist products you want to keep, because in hospitality, quality at the right price point is what keeps customers coming back. A bigger headline saving is usually possible on paper by swapping specialist products for cheaper, lower-quality alternatives. Most operators do not want to do that, so we do not count it. - **Validated, not estimated.** Savings are measured line by line against your own purchasing data, and the final prices you are offered are run back through Percy before you act. Our 5–7% is a validated figure. - **Granular.** Every product line is compared, not category averages or basket deals, so you can see exactly which products and suppliers a saving comes from. - **High-confidence matches only.** A product only counts as a match when its AI match-confidence score is 70% or higher, which means a very close match. Lines below that bar are not thrown away: they are held for review rather than matched automatically. - **Pack sizes handled accurately.** Pack-size differences are one of the most common sources of error in price comparisons, especially when they are worked out by hand. Percy applies pack-size correction factors automatically, so every price is compared per kg, litre or item. - **No membership, no system change.** Percy works from the data you already have. There is nothing to join or install, and you can stop at any point. ## Is Percy impartial? Percy is results-based: participating wholesalers pay us a referral fee once they start working with an operator we introduced, and we compare against wholesalers on the platform. But there is never an obligation. You know your own suppliers better than we do, and taking the findings into a conversation with them is a valid outcome, often the path of least resistance. What the data shows is that if you have not been out to market in the last 12 to 18 months, there is almost always money on the table. To be fair to the wholesalers on the platform, we do not show their unit prices in the early stages, so their price lists cannot simply be lifted; we do show which products and parts of your basket hold savings. You can stop at any point, with no system changes and nothing to pay. ## How Percy works ### Step 1: Consent When you get in touch, we send you an invitation to Percy. It asks a few questions about your business and asks you to accept our 7 operator terms, all of which must be accepted before we start: - No cost to you: Percy earns referral fees from wholesalers, not from you - No obligation to buy from any supplier Percy introduces - Data sharing: Percy may share your purchasing data, email and phone number with relevant wholesalers for the purpose of introductions - Confidentiality: your data is not shared with other third parties for unrelated purposes - Use of logo: Percy may show your company name and logo on its website and marketing as a platform user - Savings shown are estimates or simulations, and actual savings may differ - Percy is not responsible for disputes with suppliers or for indirect losses ### Step 2: Discovery call A short conversation to understand your business, how you buy today, who you buy from, and what you want to achieve. This is an important stage. A clearly defined scope makes sure the analysis produces results that are commercially relevant and that you can act on. ### Step 3: Agree the scope Together we agree which purchasing categories, suppliers or products the analysis should cover, and which are off-limits. - Categories and suppliers to include - Suppliers or products you will not change, for example because of a contract or a brand you serve - Products where specification or quality matters most - The sites and the period of purchasing the data should cover ### Step 4: Share your purchasing data You share purchasing data for the agreed scope only. Our preferred format is a structured export from your purchasing or ordering system. If that isn’t available, spreadsheets and supplier invoices work too. ### Step 5: Analysis Percy uses AI-assisted product matching to compare what you buy against products available from participating wholesale suppliers, taking account of product descriptions, pack sizes, quantities and prices. Our team checks the matches, and we may ask you to look at the few we are unsure about. ### Step 6: Supplier comparison Participating wholesalers price your basket without seeing who you are. They confirm prices on matched products and suggest products for any gaps, so every potential switch is priced line by line. ### Step 7: Review the opportunities We present the results on a call and talk through each opportunity: equivalent products you could buy for less, where consolidating with fewer suppliers would help, and where you are already paying a good price. You decide which, if any, are right for your business. - If you want to pursue an opportunity, we introduce you to the wholesaler - Samples can be arranged where a product needs tasting or testing first - We run the wholesaler’s final prices back through Percy against your own data, so you can see the saving is real - Any change of supplier or product is your decision, made directly with the supplier ## Getting started 1. Book a discovery call, or send us a message. 2. Accept the operator terms in the invitation we send you before the call. 3. On the call, agree the purchasing categories to investigate. 4. Prepare and share the purchasing data for that scope. Book a discovery call: https://calendar.app.google/JqnmGyVZ3p73MyBdA Send a message: https://www.percylabs.ai/contact Email: hello@percylabs.ai. Phone: +44 (0)781 860 7009. ## Who Percy helps - **Restaurant groups**: Independent restaurants through to national multi-site groups. - **Pub and bar operators**: Managed pubs, pub groups and bar operators, food and drink. - **Café and coffee shop groups**: Including bakeries, all-day cafés and food-to-go. - **Hotels and hospitality venues**: Kitchens and bars within hotels and venues. - **Leisure and visitor attractions**: Attractions and leisure sites with food and beverage operations. - **Food service and other buyers**: Contract caterers, food service and any business with recurring food and beverage purchasing. ### Percy is a good fit when - You buy food, drink or consumables every week from one or more wholesalers - You can get hold of your purchasing data, whether a system export, a spreadsheet or invoices - You are open to buying some products from a different supplier if the saving is worth it - Someone who makes purchasing decisions can join a short discovery call - You want better prices without lowering your product specification, quality or service levels ### Percy is less suitable when - You need someone to run your day-to-day ordering or manage suppliers for you. Percy finds and presents opportunities; changes are made by you, with the supplier - Your business is outside the UK ### Size Any size. Our smallest client is a single café turning over about £150,000 a year; our largest are enterprise groups spending nine figures a year on cost of goods, where we usually look at nominated lines because of the commercial complexity of those relationships. The sweet spot is operators turning over roughly £1m to £30m: often chains that started with one set of suppliers and added more as they grew, without a clear strategy. Percy helps make that supply base leaner and better defined. ### Large and multinational groups For very large, complex or multinational groups, where contract terms make new supplier introductions unlikely beyond nominated or niche lines, Percy can instead be licensed as an analytics tool for your own procurement team, on a licence plus success fee basis. We talk this through on the discovery call. ### Wholesaler network Percy works with leading wholesalers covering food, drinks, fruit and veg, meat, cleaning products and disposables, with both regional and national distribution, so for almost any part of the UK we can find a supplier able to service an operator. We do not name participating wholesalers publicly. ### Purchasing categories covered - Fresh produce - Meat and poultry - Dairy and eggs - Bakery - Frozen - Dry store and ambient - Drinks - Packaging and disposables - Cleaning and hygiene products ### When Percy is most useful - You have not benchmarked your supplier prices for a while - Food and drink costs are squeezing your margins - A supplier contract is up for renewal, or you are planning a tender - You have grown and now buy from many suppliers across several sites - A new finance or operations lead wants a clear baseline of what you pay ### What to have ready for the discovery call - Your main suppliers, and roughly what you spend with each per month - The purchasing categories you most want to look at - Any suppliers or products that are off-limits, and why - Products where a specific brand or specification must not change - Which purchasing or ordering system you use, if any - The sites you would like the analysis to cover ## Data requirements Share purchasing data only for the scope agreed on the discovery call. ### Accepted formats - **Export from your purchasing or ordering system** (preferred): A line-by-line export of purchase history from your purchase-to-pay platform, inventory system or the ordering module of your EPOS. Structured exports give the most accurate matching. Formats: CSV or Excel, one row per product line. - **Spreadsheets and wholesaler order history** (accepted): Your own purchasing spreadsheets, or order history downloaded from your wholesalers’ online accounts. Use our template if you are putting a spreadsheet together. Formats: CSV, XLS or XLSX. - **Supplier invoices** (accepted): Where nothing else is available, we can read supplier invoices. Line-level detail (product, pack size, quantity and price) must be visible. Formats: PDF, or clear photos (JPG, PNG, HEIC); several files can be sent as a ZIP. ### Fields | Field | Need | Why | Example | | --- | --- | --- | --- | | Produce Name | Required | The product description. This is what we match against wholesaler products. | Mature cheddar block | | Quantity Ordered | Required | How many units you bought in the period. Needed to work out spend and savings. | 6 | | Unit Price | Required | The price you paid per unit (per case, each, kg). A total can stand in if the unit price is missing. | 32.40 | | Supplier Name | Recommended | Who you bought it from, so results can be shown supplier by supplier. | Example Wholesaler Ltd | | Supplier Product Code | Recommended | The supplier’s product code (SKU). Makes matching more reliable than description alone. | EW10234 | | Sales Unit / Pack Size | Recommended | How the product is sold and packed. Essential for comparing like-for-like prices. | 1 x 5kg | | Weight In Grams | Optional | Pack weight, where it applies. Helps compare products sold in different pack sizes. | 5000 | | Total Cost of All Orders | Optional | Total spent on the line in the period (quantity × unit price). Useful as a check. | 194.40 | Column names do not have to match exactly; common headings such as "Item", "Qty", "Price" or "SKU" are recognised. ### Preparing the file - One header row at the top, then one row per product line. Totals by invoice are not enough. - Keep numbers as numbers, in one currency. Avoid merged cells and notes in the middle of the data. - Put everything in one worksheet, or tell us which tabs hold purchasing data (one tab per site is fine). - Cover the period agreed on the discovery call, and tell us what it is, for example 1 to 30 June. - Include every site and supplier in the agreed scope, so savings reflect your full spend. - Have a site or category column too? Add it. Extra columns are welcome, and we map common headings automatically. ### Templates - Percy operator template (CSV): https://www.percylabs.ai/downloads/percy-purchasing-template.csv - Example purchasing file (CSV): https://www.percylabs.ai/downloads/percy-purchasing-example.csv ### Sending data and how it is used After the discovery call we tell you how to send your files, usually by email or a secure upload link. Please do not send purchasing data before the scope has been agreed. Your purchasing data is used to produce your analysis. Participating wholesalers see the products and volumes they are asked to price, without your name. Your details are shared with a wholesaler when you agree to an introduction, as set out in the operator terms, and never with other third parties for unrelated purposes. Percy may use anonymised, aggregated data to improve the service, but never sells it. ### Data security - Your data is held on Amazon Web Services (AWS) servers in the EU. - Access to the systems that hold your data is protected by two-factor authentication. - We do not sell or monetise your data. Confidential information is never sold to wholesalers or to other operators; that is not our business model. - Enterprise clients can ask us to sign a non-disclosure agreement. - If you would rather we did not use your name or logo in our marketing, we can agree that with you. ## Methodology ### What you receive - Your purchasing for the agreed scope, standardised and consolidated across suppliers and sites - Equivalent products from participating wholesalers, line by line, with the saving on each. Wholesalers’ unit prices stay hidden in the early stages, to be fair to the wholesalers on the platform - Estimated monthly and annual savings by supplier and by category, including from consolidating spend with fewer suppliers - Where you are already paying a good price, so you know what to leave alone - Products with no reliable equivalent, listed separately so nothing is hidden - A call to walk through the results, introductions to the wholesalers you choose, and a check of their final prices against your data ### How the analysis works 1. **Standardise.** We read your exports, spreadsheets or invoices and turn them into one consistent list: product, supplier, pack size, unit, quantity and price. Repeated products are combined, and volumes are expressed per month. 2. **Apply the agreed scope.** Suppliers and products you have ruled out are removed before any matching, so they never appear as opportunities. 3. **Match.** AI-assisted matching compares each product with the ranges of participating wholesale suppliers, using description, product code, pack size and unit. Every match carries a confidence score. 4. **Check.** A match only counts when its AI match-confidence score is 70% or higher. Lines below that bar are held for review rather than matched automatically, and only doubtful ones need a human look from you. Price differences large enough to suggest a match is not like for like are kept out of the headline saving too. 5. **Price.** Participating wholesalers confirm prices for your matched products and suggest products for gaps, without being told who you are. 6. **Compare like for like.** Prices are converted to the same basis, such as per kg, per litre or per item, using pack and case sizes, so a 6 x 1 kg pack can be fairly compared with a 1 x 5 kg pack. Consolidation is modelled by case size and volume. 7. **Estimate the saving.** For each confirmed match, the saving is the difference in like-for-like price multiplied by your monthly quantity. Lines are added up by supplier and category; the annual figure is twelve times the monthly figure. 8. **Validate.** When you choose a wholesaler, their final prices are run back through Percy against your own purchasing data, so the saving you act on is the saving you get. ### Sample report (illustrative data) Headline: 312 items analysed, 128 matched, £4,870 a month saving (5.6% of spend), £58,440 a year. Top savings by product: - Beef burger 113g (48 x 113g) (Current supplier A) → Beef burger 4oz (48 x 113g) (Wholesaler 1): £612 a month - Grated mozzarella (6 x 2kg) (Current supplier B) → Grated mozzarella (5 x 2.5kg) (Wholesaler 2): £488 a month - Chicken thigh fillet (1 x 5kg) (Current supplier A) → Chicken thigh fillet (2 x 2.5kg) (Wholesaler 1): £402 a month - Rapeseed oil (1 x 20L) (Current supplier C) → Rapeseed oil (1 x 20L) (Wholesaler 3): £96 a month - Kraft takeaway box no. 2 (1 x 250) (Current supplier C) → Kraft food box no. 2 (2 x 125) (Wholesaler 2): £84 a month Every line falls into one of three groups: - Competitive: switch now (93 lines): A close match that is cheaper today. - Needs negotiation: requote (35 lines): A close match that costs more today; the wholesaler is asked to improve its price. - Coverage gap: no match (184 lines): No high-confidence match. Listed so nothing is hidden. ### Illustrative example Rapeseed oil, 1 x 20 L. Current price per 20 L drum: £31.20. Equivalent 20 L drum from a participating wholesaler: £28.40. Saving £2.80 per drum × 6 drums a month = £16.80 a month, £201.60 a year. Figures are illustrative. ### What the figures mean - Savings are estimates based on your data and wholesalers’ prices at the time of the analysis. - Actual savings depend on which opportunities you take up, delivery terms and minimum orders, and future price changes. - Where a product needs tasting or testing, samples can be arranged before you decide. ### Results across the Percy network - 80+ operators on the Percy network - £400M+ cost of goods reviewed across that network - £1.3bn combined revenue of operators on the network - 5–7% of the food and beverage spend analysed, typically recovered - Even where procurement is already working well, for example with an e-procurement system in place, Percy typically still finds 3 to 5%. - Across a typical basket, 88% of the savings come from just 10% of the products switched. - Used by operators from a single café turning over about £150,000 to enterprise groups, and proven on baskets of more than 1,000 products. Where the savings came from, across more than 32,000 lines of operator purchasing analysed: - Ambient and dry goods: 33% of the total saved - Specialty and niche products: 32% of the total saved - Proteins: 21% of the total saved - Everything else combined: 14% of the total saved Operators on the Percy network include Loungers, Turtle Bay, The Breakfast Club, Yolk, Pinkmans, Kolamba, K10, Lussmanns, Left Handed Giant, Real Eating Company, Camino, Rola Wala, Benito's, Walcot Group, Federal, Dodo Pub Co, The Attendant, Heidi, Ardent Pub Group, Pizzarova, Little Victories, Brother Marcus, TreeTop Golf, Parsons Bakery, Double Puc Café. "The fact it's been made by you guys with your hospitality experience really shows through." (Phil Edwards, Ops Director, TreeTop Golf) ## Case studies ### Double Puc: Four-site café group, where Percy was built and tested 12.5% off fruit and veg spend: enough on its own to offset the April 2025 National Insurance rise across four sites. **Situation.** Double Puc is the café group run by Percy’s founders, and where Percy was built and tested before it went to market. Higher employer National Insurance contributions from April 2025 squeezed already tight margins. **What Percy did:** - Ran a full audit of fresh produce spend, matching every product to equivalent products across several wholesalers - Identified a new regional supplier the business had not worked with before - Forecast annual savings from exact product volumes - Spotted that existing suppliers had expanded their ranges, so products bought from three suppliers could be consolidated with one, reaching higher-volume price tiers **Outcome.** A 12.5% saving on fruit and veg spend, which on its own offset the National Insurance increase across all four sites. Source: https://www.percylabs.ai/blog/how-we-offset-aprils-national-insurance-hike-by-slashing-supplier-spend-a-real-world-hospitality-case-study-50b39a ### Elevate: Premium smoothie, matcha and healthy food concept Two new supplier matches for a growing wellness brand, with the saving validated against its own purchasing data. **Situation.** Elevate was moving from one site to two, with a deliberately tight supply chain of just a couple of key suppliers. **What Percy did:** - Analysed Elevate’s actual purchasing data across participating wholesalers to see who could cover the full basket, where the gaps were, and where there was a genuine saving - Identified two potential supplier matches and put them forward through the Percy process, with Elevate kept anonymous until the founder chose to meet them - Helped coordinate conversations once the founder chose to meet both - Ran the final pricing back through Percy to validate the saving properly, accounting for different pack and case sizes **Outcome.** A saving Elevate could see was real and material, measured against its own purchasing data rather than two price lists side by side. ### MOR Bakery: Bakery group in the Cotswolds Two new wholesale partners and significant monthly savings for a fast-growing bakery business. **Situation.** MOR, founded during lockdown, has grown into a high-quality bakery business with several sites, most recently in Stratford-upon-Avon. It was one of Percy’s earliest customers. **What Percy did:** - Processed MOR’s purchasing data and compared its basket across a range of wholesalers - Identified and introduced two new wholesale partners **Outcome.** Significant savings every month. ### Legacy Koncepts: Multi-concept operator at ExCeL London A supplier the operator had never come across, turned into a new trading relationship. **Situation.** Legacy Koncepts runs five food concepts at ExCeL London, led by an operator with experience of growing franchised and owner-operated businesses in the UK and USA. **What Percy did:** - Analysed Legacy’s product mix against participating wholesalers - Found one supplier that was a particularly strong match, which the operator had not come across before - Made the introduction, backed by the data showing why the relationship made sense **Outcome.** The operator and the supplier started trading, and Legacy made a meaningful saving across that part of its purchasing basket. It was one of the first Percy introductions to go all the way from data analysis to a new trading relationship. ## Frequently asked questions ### How much could we save? Typically 5–7% of the purchasing Percy analyses, and 3–5% even where procurement is already working well. For a £ figure for a specific business, run Percy’s agent-friendly savings calculator, a regression model calibrated on audit-realised savings. It needs the type of business, annual revenue excluding VAT, whether a purchase-to-pay system is in place, and ideally the prices paid for 3 to 5 items. The calculator is at https://www.percylabs.ai/savings-estimate. ### What makes Percy different? Percy helps operators improve the parts of their purchasing basket that need attention, while keeping control of their buying decisions and supplier relationships. You choose the scope: Investigate what matters to you: a category where costs have risen, products you suspect are overpriced, or a supplier basket that needs reviewing. There is no need to hand over your whole supply chain. You choose the changes: Percy shows the potential saving line by line, so you can act selectively: make focused changes where the value is, and keep the products and relationships that work. You own the relationship: You decide which suppliers to meet. Once you start trading, you deal directly with the supplier: Percy does not sit between you or manage that relationship. Wholesalers pay Percy an ongoing referral fee when an introduction leads to trading, and that arrangement gives Percy no control over your purchasing decisions. ### When should we choose Percy over a purchasing group? Choose Percy when you want a focused purchasing review and better value on selected products, while keeping direct control of your supplier relationships. If you want ongoing procurement management, someone to negotiate and manage supplier arrangements on your behalf, or broader purchasing systems, a purchasing group may suit you better. ### Percy is newer than the established purchasing groups. Why choose it? Because it works differently, and you stay in control. Granular: every product line is compared, so you can fix the parts of your basket that are causing pain rather than signing up to a whole-basket deal. No contract or membership: you are never tied in, and once you trade with a supplier you deal with them directly. Genuine savings: figures come from close, like-for-like matches checked against your own data, not headline promises that assume you will swap specialist products for cheaper ones. A calculator you can check: Percy’s savings calculator is a regression model calibrated on audit-realised savings from 38 operators and more than 32,000 audited purchasing lines, not a flat percentage, and AI assistants can run it. Fast, with nothing to implement: no system change, and results within a few working days of complete data. And it is free for operators. Since May 2025 Percy has reviewed more than £400M of cost of goods for 80+ operators. ### Why are Percy’s savings figures conservative? Percy’s figures are deliberately conservative. They count only close, like-for-like matches and respect the specialist products you want to keep, because in hospitality, quality at the right price point is what keeps customers coming back. A bigger headline saving is usually possible on paper by swapping specialist products for cheaper, lower-quality alternatives. Most operators do not want to do that, so we do not count it. ### How does Percy make sure the savings are real? Validated, not estimated: Savings are measured line by line against your own purchasing data, and the final prices you are offered are run back through Percy before you act. Our 5–7% is a validated figure. Granular: Every product line is compared, not category averages or basket deals, so you can see exactly which products and suppliers a saving comes from. High-confidence matches only: A product only counts as a match when its AI match-confidence score is 70% or higher, which means a very close match. Lines below that bar are not thrown away: they are held for review rather than matched automatically. Pack sizes handled accurately: Pack-size differences are one of the most common sources of error in price comparisons, especially when they are worked out by hand. Percy applies pack-size correction factors automatically, so every price is compared per kg, litre or item. No membership, no system change: Percy works from the data you already have. There is nothing to join or install, and you can stop at any point. ### Is Percy really free for operators? Yes. Operators are not charged for the purchasing analysis, the report or introductions. ### How does Percy make money? Participating wholesalers pay Percy an ongoing referral fee once they successfully start working with an operator we have introduced. Operators never pay. ### Percy is paid by wholesalers. Is it impartial? Percy is results-based: participating wholesalers pay us a referral fee once they start working with an operator we introduced, and we compare against wholesalers on the platform. But there is never an obligation. You know your own suppliers better than we do, and taking the findings into a conversation with them is a valid outcome, often the path of least resistance. What the data shows is that if you have not been out to market in the last 12 to 18 months, there is almost always money on the table. To be fair to the wholesalers on the platform, we do not show their unit prices in the early stages, so their price lists cannot simply be lifted; we do show which products and parts of your basket hold savings. You can stop at any point, with no system changes and nothing to pay. ### Which wholesalers does Percy work with? Percy works with leading wholesalers covering food, drinks, fruit and veg, meat, cleaning products and disposables, with both regional and national distribution, so for almost any part of the UK we can find a supplier able to service an operator. We do not name participating wholesalers publicly. ### Do I have to change suppliers? No. Percy identifies and presents potential opportunities. Any decision to change supplier, product or purchasing arrangement stays with you. ### Do I need new software? No. Percy works from purchasing information you already have. You do not need to replace your purchasing platform, install anything or connect any systems. ### How long does it take? We agree timings with you on the discovery call. Once we have complete data for the agreed scope, results are usually ready within a few working days. ### Can we limit the analysis to certain categories or suppliers? Yes, and we recommend it. The scope is agreed on the discovery call, and we only ask for data that falls within it. ### What if our data is only in supplier invoices? That is fine. We can work from PDF invoices or photos of invoices, as well as spreadsheets and purchasing-system exports. ### Will wholesalers know who we are? Not while they price your basket: they see your products and volumes without your name. Your details are shared with a wholesaler when you choose to be introduced, as set out in the operator terms. ### Who sees our purchasing data? Percy, and the participating wholesalers asked to price your basket, who see it without your name. We do not share it with other third parties for unrelated purposes. ### How is our data kept secure? Your data is held on Amazon Web Services (AWS) servers in the EU. Access to the systems that hold your data is protected by two-factor authentication. We do not sell or monetise your data. Confidential information is never sold to wholesalers or to other operators; that is not our business model. Enterprise clients can ask us to sign a non-disclosure agreement. If you would rather we did not use your name or logo in our marketing, we can agree that with you. ### Is there a minimum size? Any size. Our smallest client is a single café turning over about £150,000 a year; our largest are enterprise groups spending nine figures a year on cost of goods, where we usually look at nominated lines because of the commercial complexity of those relationships. The sweet spot is operators turning over roughly £1m to £30m: often chains that started with one set of suppliers and added more as they grew, without a clear strategy. Percy helps make that supply base leaner and better defined. ### Are the savings guaranteed? No. Savings figures are estimates based on your data and wholesalers’ prices at the time of the analysis. Actual savings depend on the products you choose, delivery terms and future price changes. ### We are a large group tied into supplier contracts. Can Percy help? Yes. Percy already works with enterprise groups, usually on nominated lines because of the commercial complexity of those relationships. For very large, complex or multinational groups, where contract terms make new supplier introductions unlikely beyond nominated or niche lines, Percy can instead be licensed as an analytics tool for your own procurement team, on a licence plus success fee basis. We talk this through on the discovery call.