Business

Know the Business

Greggs is a vertically integrated, value-led food-on-the-go specialist that bakes its own product, runs its own logistics and operates its own shops — a UK retail operation with the cost discipline of a manufacturer. The economic engine is small-ticket, high-frequency footfall (a £3 sausage roll sold across 2,739 shops, every weekday) where the company makes money on volume, scale of own-label production, and prime real-estate density. The market is currently underestimating two things: how punishing 2025–26 will be for ROCE as £287m of supply-chain capex peaks against only +2.4% like-for-like sales, and how fast that ROCE can re-rate once Derby and Kettering come online and capex collapses to £150–170m by 2028.

How This Business Actually Works

Greggs is a baker that owns the shops. Every Sausage Roll, Steak Bake and pizza slice is produced in one of the company's own bakeries (Newcastle, Balliol, Leeds, Birmingham, Treforest, plus the new Derby site coming online in H2 2026), shipped through its own radial distribution network, and sold through 2,739 mostly-leased shops. About 75% are company-managed — generating £1.9bn of the £2.15bn FY2025 revenue — and the rest are franchised in roadside, transport-hub and forecourt locations where partners like Moto, Roadchef, Applegreen and EG Group bring real-estate access Greggs cannot easily replicate. Franchised shops contribute via wholesale of finished product and commission, surfacing as the £254m business-to-business line that grew 9.2% in 2025 versus 6.5% in company-managed shops.

The unit economics are deceptively narrow but reliably profitable. A new shop costs around £430k to fit, and Greggs targets a 25% cash return on that investment, typically achieved within two to three years and often exceeding 30% at maturity. Gross margin sits stubbornly at 61–63% because the company manufactures rather than buys finished product — meaningfully higher than the typical QSR — but distribution and selling costs (rent, rates, shop labour, energy, lease depreciation) absorb 47–48 percentage points of revenue, leaving an underlying operating margin oscillating between 8% and 10% in normal years. People are 39% of the cost base and food/packaging another 33%, so the two macro variables that move the P&L are UK wage inflation (+8.3% in 2025 thanks to National Living Wage and employer NIC) and commodity input costs.

Where it really makes money is operating leverage on incremental volume. The bakeries and distribution centres are largely fixed-cost; once a tray of sausage rolls is being baked, baking another tray costs almost only flour, fat and pork. So the marginal contribution from a hot afternoon's footfall, an evening pizza-box order or an extra shop on a forecourt drops disproportionately to operating profit. Conversely — and this is what hit 2025 — when company-managed like-for-like volumes go negative (footfall down because of the heatwave in June/July, plus a consumer-confidence drag), the same fixed-cost network produces -4% operating profit on +6.8% sales. The other amplifier of the model is App-driven frequency: the loyalty App was scanned in 26.7% of company-managed transactions in 2025, up from 20.1% in 2024, giving Greggs a CRM data layer it never had — and which it is now monetising through Greggs Quests gamification and personalised "Baked for You" offers.

What truly drives incremental profit is the gap between three structural growth levers and two structural cost headwinds. The levers: shop count expansion in under-penetrated catchments (current penetration of viable petrol forecourts is only ~35%, supermarkets ~20%, retail parks ~25%), evening daypart growth (9.4% of company-managed sales in 2025 vs 9.0% in 2024, fastest-growing daypart, driven by hot food and delivery), and grocery channel through Iceland and now Tesco (800 large stores plus 1,900 Tesco Express units from January 2026). The headwinds: National Living Wage compounding at 5–10% per year, and lease depreciation creep from estate growth. The arithmetic only works if shop additions and channel partnerships create incremental volume faster than wage costs eat margin — which is why the 2026–28 ROCE recovery thesis hinges on whether like-for-like volumes return to +3-4% as consumer confidence recovers.

The Playing Field

There is no clean comparable for Greggs in UK listed equities, which is itself part of the moat. The peer table below pulls the closest five — pizza delivery (Domino's UK), travel-food (SSP), pub estates (JD Wetherspoon), pub-and-hotel (Whitbread/Premier Inn) and global contract caterer (Compass Group) — and shows the spread of business models that all touch UK out-of-home eating without doing what Greggs does. The takeaway: Greggs sits between Domino's UK (high return on capital, asset-light franchise model, tiny revenue) and Compass Group (massive scale, low margin, huge return on capital) on the value chain, but is unique in combining vertical manufacturing with a high-street retail estate at scale.

The peer set reveals where the moat is real and where it is imaginary. The real edge is value at premium frequency: Greggs ranks number one for value in YouGov's BrandIndex across the entire UK quick-service-restaurant, coffee-shop and delivery sector — not number one in food-to-go, number one across all of QSR. That price perception is built on three structural advantages no peer can replicate at this footprint. First, vertical manufacturing — Domino's franchises bake locally, SSP buys finished product, Compass cooks on-site at corporate locations, but only Greggs runs five bakeries plus a national logistics fleet feeding 2,700 shops. Second, real-estate density — 2,739 UK shops gives Greggs a footprint roughly 4x Pret's UK estate and 3x Costa Coffee's owned shops, which means brand presence on virtually every high street, retail park and motorway service area. Third, a 33,000-person operating organisation tuned to a £3 average ticket; matching this requires either inheriting a similar baker (Cooplands, Pound Bakery, Bakers Plus — all far smaller) or rebuilding from scratch, which nobody is doing.

What the peer set also reveals is what Greggs is not. It is not Compass, whose 108% return-on-capital reflects an asset-light contract-catering model where the customer (a hospital, an airline, a corporate cafeteria) provides the kitchen; Greggs owns or leases every brick. It is not Domino's UK, whose 83% return-on-capital comes from a master-franchisor model where franchisees fund all the stores; Greggs runs ~75% of its estate on its own balance sheet. It looks more like Whitbread/Premier Inn — a fully-owned UK retail estate — except Whitbread's customer is a £100 hotel night and Greggs' is a £3 lunch, which makes Greggs much more transaction-intensive and far less weather-resilient than its margin profile suggests. The "good in this industry" benchmark is Domino's UK at the asset-light end and Compass at the global-scale end; Greggs sits in the awkward middle, defending its position with brand, vertical integration and footprint density rather than capital efficiency.

Is This Business Cyclical?

Yes — more than the 60-year track record of bakery growth would suggest. Greggs is exposed at three points: footfall (consumer disposable income and weather), wage inflation (UK labour-market policy), and capital-cycle timing (its own choice). The 2020 COVID year drove revenue down 30% to £811m and operating profit to a £7m loss; that demonstrated that a 2,200-shop high-street retailer with no delivery infrastructure dropped through 100% of operating leverage when shops physically closed. Since then management has built delivery (now 6.8% of mix) and grocery (Iceland + Tesco) as cyclical buffers, but the 2025 result is the more interesting cycle test: in a year of 6.8% sales growth, operating profit fell 4% because like-for-like volumes were only +2.4% (further softening to ~+1.6% in early 2026), wage inflation ran 8.3%, and the supply-chain investment programme front-loaded fixed costs. ROCE went from 20.3% to 16.0% in twelve months without a recession, on rising sales.

Three episodes carry teaching value. FY2013 saw operating margin compress from 7.2% to 5.2% as the company executed a then-painful estate transformation (closing in-mall bakeries, investing in shop refits) — a self-inflicted capital-cycle dent that took two years to reverse. FY2019 was the post-IFRS 16 jump: EBITDA margin appears to leap to 19.3% but that is largely the lease-accounting reclass; the genuine signal is that 2018–19 was the high-water mark for unit economics before Covid intervened. FY2020 was the binary shock: £811m revenue, operating loss, no dividend, but a quick recovery to record £153m profit in 2024 once the estate normalised. The current cycle, 2024-26, is the third lesson — capital-cycle plus weak consumer plus wage shock. The signature is that capex peaks at £287m in 2025 against £52m of FCF, drawing the £125m net cash position down to £46m, and is set to reduce to £200m in 2026 and £150–170m from 2027.

Where the cycle hits hardest is operating margin, not revenue. Demand for an affordable lunch is sticky — even in 2025 the food-to-go market overall lost 3.1% of visits while Greggs gained share — but the company has limited ability to flex its fixed cost base downwards when volumes soften. Wage and lease costs are largely contractual; food and packaging are partially hedged (about four months forward cover, 100% of 2026 electricity fixed) but the underlying commodity exposure to wheat, fats, dairy and pork is real. Working capital is structurally favourable: Greggs collects from customers (cash and card at the till) faster than it pays suppliers (standard terms), giving it a chronic net-current-liability position (£152m at year-end 2025, up from £67m). That working-capital float is a permanent funding source — but it also means a sales decline mechanically compresses cash availability faster than P&L because the float reverses.

The Metrics That Actually Matter

Forget P/E and forget gross margin — they are noise in this business. The five metrics below explain whether Greggs is creating value, and the 2025 results show why three of them are flashing yellow.

Underlying ROCE is the single most important number because it sits at the intersection of profit margin and capital intensity, and management has explicitly anchored its strategy on a ~20% target. Underlying ROCE was 16.0% in 2025 versus 20.3% in 2024, and management has guided that it will fall further in 2026 as the Derby and Kettering distribution centres come into the asset base before generating their associated profits. The mechanical recovery to ~20% requires three things to happen by 2028: capex falls from £287m to £150-170m, like-for-like sales return to +3-5%, and the new logistics capacity supports continued shop additions toward the 3,500-shop medium-term ambition.

Company-managed like-for-like sales is the cleanest read on whether the brand is winning or losing share. The +2.4% printed in 2025 was below internal expectations and weakened to +1.6% in the first nine weeks of 2026, well under the ~+5% the model needs to leverage operating costs. The market context matters: the broader UK food-to-go market lost 3.1% of visits in 2025 per Circana, so Greggs gained share (8.1% to 8.6% of food-to-go visits) — but absolute volume growth is what drives operating leverage, and 2.4% is not enough to absorb 5.6% LFL cost inflation.

Cash return on new shops (25% target, 30%+ at maturity) is the unit-economic anchor. As long as new openings deliver the target, capital allocated to estate growth compounds well above cost of capital and the strategy works. The risk is that in a softer consumer environment the maturation curve slows from two-three years toward four-five years, lowering the IRR on the £430k-per-shop investment.

Net shop openings (121 in 2025, ~120 planned for 2026) is the volume side of the growth equation. More than half of new openings are now in non-traditional locations — petrol forecourts, retail parks, transport hubs, supermarkets, university and hospital campuses — which the company calls "underrepresented catchments." Penetration data from the 2025 prelim shows just 35% of viable petrol forecourts and ~20% of viable supermarkets currently host a Greggs, against near-saturation in central London Underground and university/hospital sites. This is the runway for the medium-term 3,500-shop target; the question is whether the 25% cash ROI hurdle holds in less-trafficked locations.

Food-to-go visit share (8.6% in 2025, +50bps in a down market) is the credibility check on the brand. Greggs is now ranked number one at breakfast, number two at lunch, number three at snacking and number four at dinner and delivery in UK food-to-go. The dinner ranking is structural — three years ago dinner was barely a Greggs occasion, and pizza, hot savouries and Mac & Cheese have built it into a 9.4% daypart share — but the absolute weight of dinner in the overall food-to-go market is small relative to lunch, where competition (M&S, Pret, Subway, supermarkets, contract caterers) is intense.

The metrics to de-emphasise are gross margin (rangebound at 61–63% because of the manufacturing model — fluctuations there are mix and commodity noise) and EPS (heavily affected by IFRS 16 lease interest, exceptional items like the £14.1m 2024 disposal gain and the £4.5m 2025 VAT correction). Underlying diluted EPS was 122.8p in 2025, but the more relevant figure is operating cash inflow per share at 267.1p — up 4.6% — which is what funds dividends and reinvestment.

What I'd Tell a Young Analyst

This is a great brand executing a hard business model in a phase where the model is paying for its own growth, and the share price is reflecting the present rather than the through-cycle. Three things to watch.

First, the ROCE-recovery clock starts when capex stops. Management has guided capex from £287m (2025) to £200m (2026) to £150–170m (2027 onwards). When that landing is verified — likely at the H1 2026 result and again at the FY2026 prelim — the FCF picture transforms: from £52m in 2025 to potentially £200m+ by 2028 on the same revenue base. If you believe the guide, 2025 is the cycle low for FCF margin (2.4%) and 2027–28 normalises to 8–10%, which is what funds restored special dividends and buybacks (point 5 of the stated capital allocation policy).

Second, like-for-like sales is the swing factor — and the early-2026 print of +1.6% says the consumer headwind has not lifted. The 20% ROCE target needs underlying volumes to grow 3-4% per year; at +2% LFL the operating leverage simply does not arrive. Watch quarterly trading updates for any sign of volume recovery, and pay particular attention to the daypart mix (evening growth and delivery basket size are the structurally encouraging signals; lunch share is the at-risk number because that is where competition is most intense).

Third, do not confuse the brand moat with the capital moat. The brand moat is real and durable: 8.6% UK food-to-go visit share, number-one value perception, a vertically integrated supply chain that no peer can copy in under five years. But the capital moat is weak — Greggs owns its bakeries and depots and leases its shops, which means any volume softness drops through to operating profit hard, and the only way to grow is to put more capital out (every new shop is £430k of fitting plus a 25-year lease commitment). The thesis-changing event is not a brand failure (very unlikely) but a structural shift in UK out-of-home eating habits — GLP-1 weight-loss drugs reducing snack frequency, persistent real-wage compression keeping consumers at home, or a competitor (likely a supermarket meal-deal player rather than another bakery) breaking the value-perception lead. None of these is imminent, but each is a tail risk that the 2025–26 trading pattern has made marginally less academic.

The trade you can run against this name is patience: Greggs is a high-quality UK consumer compounder going through a self-inflicted capital cycle in an unfriendly consumer environment. If you believe in the medium-term 3,500-shop story and the ROCE recovery to ~20%, the right entry is during the period when the market is most worried about ROCE — which is now. If you do not believe management can recover ROCE because volume growth has structurally slowed, then 16% ROCE on a fully-priced consumer brand is mediocre and the dividend yield (~4.5%) is not enough to compensate. The next four prints — H1 2026, Q3 2026 trading update, FY2026 prelim and H1 2027 — will tell you which world you are in.