Written by: JJ Tan, Founder, Jelly

Key Takeaways For UK Multi-Site Operators

  • Centralised purchasing puts supplier negotiation, pricing, and stock visibility at head office while sites handle day-to-day ordering.
  • The model delivers bulk discounts, lower waste, consistent quality, and less admin through standard supplier lists and automated workflows.
  • UK operators face rising food inflation and margin pressure, so centralised control now plays a key role in protecting profitability.
  • Implementation depends on clear governance. Head office owns supplier agreements and rules. Sites manage forecasting, ordering, and receipt recording.
  • Jelly delivers invoice-led purchasing and stock control with first-week onboarding, flat-rate pricing, and native EPOS integrations for UK multi-site groups.

See Jelly In Action For Your Sites

How Centralised Purchasing Works Across Multiple Restaurant Sites

The governing principle is simple: centralise control, standardise what matters, decentralise day-to-day ordering. That principle only works when the rules are written down first. Centralisation starts as a governance decision before it becomes a software project, so the rules exist on paper before any platform is configured.

The operating model follows four stages:

  1. Head office negotiates suppliers and sets approved products and prices. One agreement covers the whole estate. Supplier relationships, pricing benchmarks, and the approved product list sit centrally. The group gains consolidated leverage instead of fragmented site-by-site buying power.
  2. Head office sets ordering rules and monitors group spend. Approval thresholds, spend limits by category, and ordering windows are defined at group level. Head office tracks aggregate spend across all sites and flags deviations, without approving every individual order.
  3. Sites forecast, raise orders, approve, receive, and record discrepancies. Site managers and chefs work inside the framework set by head office. They raise orders against the approved supplier list, accept deliveries, and record discrepancies at receipt. Data stays accurate without head office stepping into routine operations.
  4. The system compares actual versus theoretical usage and reports food cost by site. Once invoices flow and POS sales data connects, the platform reconciles what was bought, received, and consumed against what was sold. The result is a food-cost report per site and for the group as a whole.

For a deeper look at how stock control fits into this model, see What Is Stock Control In Restaurant Inventory Management?

Head Office Versus Site Responsibilities For Ordering

Clear responsibility splits make centralised purchasing work in practice. Define this split before configuring any software so every site understands how the model runs day to day.

  • Head office owns: supplier negotiation, approved product lists, price agreements, ordering rules, group spend monitoring, and food-cost reporting across the estate.
  • Sites own: demand forecasting, raising orders against the approved list, approving deliveries, recording discrepancies at receipt, and day-to-day stock counts.

This split keeps restaurant managers moving without waiting for head office to approve every order while spend still sits inside agreed guardrails. Purchase order permissions should be scoped per cost centre rather than globally, with approval thresholds set by branch and order value. A single global permission lets unauthorised orders reach suppliers without detection.

For a comparison of stock control and inventory management responsibilities across a multi-site estate, see Stock Control vs Inventory Management: Two Jobs, One System.

Talk Through Your Head Office / Site Split

The Data Flow From Purchase Order To Food Cost

Once head office and site responsibilities are clear, the next step is to map how data moves between them. The chain that connects a purchase order to a food-cost report runs through five stages: purchase order → delivery → invoice matching → recipe costing → food-cost reporting. Each stage depends on the accuracy of the previous one.

Jelly puts invoice data at the centre of this chain. Invoices are captured by email or photo. Every line item is then digitised automatically, including quantity, SKU, price, and tax. Price alerts flag increases and decreases by supplier the moment a new invoice arrives. Operators gain the evidence needed to challenge a supplier or claim a credit note before the cost hits the P&L. Amber restaurant saves £3,000–£4,000 per month using this approach, achieving approximately 68× ROI.

The integration points that complete the data flow are:

  • EPOS: Jelly integrates natively with Square, Lightspeed, EPOS Now, and Toast via real-time API. That connection delivers item-level sales data the moment a transaction completes. POS setup across all four systems takes under five minutes. POS-to-dish linking only surfaces items sold since the integration was connected, which keeps mapping clean and free of legacy menu clutter. These are complementary tools Jelly works alongside.
  • Accounting: Jelly integrates directly with Xero, with Sage integration coming soon. Digitised invoices push into the accounting platform in one click and reduce bookkeeping time by 90%.

The EPOS records what was sold. The purchasing and stock layer records what was bought, received, and consumed, then reconciles the two into food cost. Sushi Revolution reduced its monthly stocktake from 2–3 hours to 5–20 minutes using Jelly, while improving gross profit by 2–3%.

For a detailed look at stocktake methodology that complements this data flow, see Advanced Restaurant Stocktake Techniques That Cut Waste.

Central Kitchen And Commissary Design For Multi-Site Groups

Central production changes how purchasing, stock, and reporting work across the group. Most multi-site restaurant groups start feeling the pressure to centralise production around three to five sites, once informal training no longer holds quality consistent.

Two financial design decisions define how a central kitchen appears inside the group accounts:

  • Cost-centre model: The central kitchen sits under a parent entity. Stock moving to a branch is recorded as an internal transfer at cost, with no inter-site invoice. This works cleanly when all sites are commonly owned and no external billing is required.
  • Separate-location model: The central kitchen operates as a standalone site. It ships on a tax invoice at a defined transfer price and carries its own profit-and-loss. This structure is required when the kitchen bills external clients or when each site’s P&L must stand independently.

In both models, inter-site stock movements must be recorded as transfers rather than purchases. If internal movement between a central kitchen and branches is not recorded at all, group cost-of-goods variance becomes structurally invisible.

The two-tier recipe problem is the most common source of branch GP lag in a central kitchen model. A raw ingredient price change at the central kitchen must propagate to every branch assembly recipe using that component. Without that link, branch GP reports lag actual production costs by weeks. Without linked two-tier recipe costing, groups with multiple branches using a common semi-finished component face the same update propagating through every branch-level recipe record manually every time a central kitchen input price changes.

UK groups with a central production kitchen must also post, price, and evidence inter-company recharges between group companies for HMRC. Inter-company recharges are one of five recurring monthly jobs a multi-site restaurant group accountant performs that a single-site accountant does not. Companies under common control can register as a single VAT group under section 43 of the VAT Act 1994, filing one return and ignoring VAT on supplies between members, which simplifies the central kitchen invoicing workflow considerably.

Where Each UK Platform Fits In Your Toolkit

The UK market for centralised purchasing and stock control software has matured. Each platform below serves a specific operator profile, so match the choice to your group size, complexity, and internal resources.

  • Fourth: Enterprise workforce and procurement platform suited to larger groups with dedicated operations teams and complex labour management requirements.
  • Supy: Inventory and procurement platform with strong stock control capabilities, suited to operators needing deep central kitchen and multi-entity support.
  • Apicbase: Recipe and menu management platform suited to food-led operations where menu engineering and nutritional compliance drive decisions.
  • MarketMan: All-in-one inventory platform with a broad feature set, typically associated with a longer onboarding process.
  • Growyze: Stock and compliance platform suited to multi-site operators where compliance reporting sits alongside stock management.
  • Freshmarkets: Procurement marketplace suited to operators whose primary need is supplier-led buying rather than internal stock control.
  • Jelly: Purpose-built for growing UK restaurant, pub, and boutique hotel groups running a handful of sites where speed of deployment matters. Jelly is invoice-led, so value appears as soon as invoices flow and price alerts and spending insights become available. Pricing is flat-rate at £129/month per location with no per-user charges. It integrates natively with Square, Lightspeed, EPOS Now, Toast, and Xero, and first-week onboarding is standard. Larger enterprise groups with complex central kitchen or multi-entity requirements may prefer a platform with deeper ERP integration.

For a broader comparison of stock control software options for UK multi-site venues, see Best Stock Control Software For UK Multi-Site Venues.

Find Out If Jelly Fits Your Group

Implementation Steps And Time-To-Value

A multi-site rollout starts with three inputs: a current supplier list, EPOS admin access for each site, and accounting platform access. With those in place, Jelly typically onboards in the first week.

Value appears as soon as invoices flow. Once suppliers send invoices to a dedicated email address, or the kitchen photographs invoices into Jelly, price alerts and spending insights appear within 24 hours. Jelly users cut food costs by 3% on average in the first three months.

Sequencing matters for a multi-site rollout. Configure the head-office cost centre or central kitchen location before the branches that order from it. The centre defines the price lists and ordering rules the branches inherit, so setting branches up first means redoing the configuration once the central framework is in place. Configuring the central kitchen before the branches is the recommended sequence precisely because the kitchen defines the price lists and production rules the branches inherit.

POS connections follow the same five-minute setup flow across all four supported systems. Open Jelly, click Integrations, sign in to the POS, grant permissions, and select which categories to sync. The only common friction point appears when the user lacks admin access to their POS account, so Jelly flags this requirement upfront. Once connected, the POS automates 2–5 hours of weekly work and delivers real-time margins and sales mix data.

Plan Your Multi-Site Rollout With Jelly

Frequently Asked Questions

What Is Centralised Purchasing And Stock Control For Restaurants?

Centralised purchasing and stock control means head office consolidates supplier negotiation, approved product lists, pricing, and stock visibility while individual sites raise orders and record deliveries against that framework. The model gives a multi-site group a single source of truth across the estate, with governance held centrally and day-to-day ordering delegated to sites.

How Much Effort Does It Take To Set Up Across Multiple Sites?

The main inputs are a current supplier list, EPOS admin access for each site, and accounting platform access. With those ready, Jelly usually completes onboarding in the first week. Follow the earlier guidance and configure the head-office or central kitchen cost centre before the branches, since the centre defines the price lists and rules the branches inherit.

How Quickly Do Results Appear?

Price alerts and spending insights appear as soon as invoices flow, usually within 24 hours of the first invoice being photographed or emailed into Jelly. As mentioned earlier, Jelly users typically cut food costs by around 3% in the first three months. One operator improved gross profit from 65% to 72% within 12 weeks on approximately £500,000 in revenue.

Who Owns Centralised Purchasing Internally?

Head office owns supplier negotiation, approved product lists, price agreements, ordering rules, and food-cost reporting across the estate. Sites own demand forecasting, raising orders, approving deliveries, recording discrepancies at receipt, and day-to-day stock counts. This split is a governance decision and should be documented before any platform is configured.

Conclusion

Centralised purchasing and stock control for restaurants gives a multi-site group the foundation to protect margin, maintain consistency, and make decisions on current data rather than last month’s accountant report. The evaluation lens that matters most is whether the governance model is clear and whether the software delivers control in the first week. Centralise control, not every order.

For UK groups running 2–5 sites and ready to implement this quarter, Jelly provides invoice-led purchasing and stock control with first-week onboarding, flat-rate pricing, and native integrations with the EPOS and accounting tools already in use.

Start Centralising Purchasing With Jelly