South African Forecourt Convenience Retail Market

How to read this report

Three principles govern the numbers below.

1. Anchoring. Market size is anchored on the most recent authoritative trade estimate — the Trade Intelligence / Nedbank Forecourt Retail Report 2025/26 (released March 2026), which values forecourt convenience retail at R40bn in 2024, +4% y/y, forecast to ~R48bn by 2028. Every other figure is cross-checked against it.
2. Explicit modelling. Where no published figure exists (site-level P&Ls, category revenue splits, non-store forecourt income), we build the number bottom-up and label it Modelled with the assumptions shown.
3. Discrepancies surfaced, not hidden. South African forecourt data is genuinely contradictory — site counts range from 3,800 to 6,000+ depending on source and definition, and market-size estimates were re-based between 2024 and 2026. Section 12.4 reconciles each conflict and states which figure we use and why.

A companion workbook, SA_Forecourt_Data_Appendix.xlsx, contains all tables in machine-readable form.

1. Executive summary

1.1 The headline

South Africa’s forecourt convenience channel is a R40bn–R42bn retail market growing at 4–5% a year inside a fuel business that is shrinking at 4–6% a year. It is the only structurally growing revenue line most fuel retailers have, and it now supplies roughly a third of forecourt gross profit on barely a sixth of forecourt turnover.

Sources: Trade Intelligence/Nedbank Forecourt Retail Report 2025/26; Kalibrate SA retail fuel surveys 2021–2024; CEF Group/DMRE fuel price schedules; Stats SA P6343.2; own modelling. See Section 12.

1.2 Key findings

• Fuel is in structural, not cyclical, decline. Consumption fell 6.3% in 2024; Stats SA records fuel sales income down 5.6% in real terms in 2025. Trade Intelligence attributes this to hybrid working, constrained consumers and improved vehicle efficiency — not to a downturn that will reverse. Some metro sites have lost 15–20% of volume since 2019.
• Yet the network keeps growing. Forecourt numbers rose ~10–12% over five years (~600 new sites since 2019) while volumes fell. The result is cannibalisation: more sites chasing fewer litres. Average throughput in Gauteng fell from 304 kl/month (2022) to 282 kl/month (2024).
• Convenience has decoupled from fuel. Retail sales at forecourts are effectively uncorrelated with fuel volumes. Revenue per litre sold is rising consistently — Trade Intelligence pegs convenience store revenue at R1.36 per litre of fuel (our volume-based recalculation suggests R2.10–R2.40; see §12.4).
• The shopper is no longer the driver. 46% of forecourt store customers are not buying fuel. The average South African visits a forecourt ~10 times a month, few of them to refuel. 74% of visitors use the store, 68% buy takeaway food.
• Supermarket brands have won the perception battle. Roughly half of surveyed consumers have a strong preference for a supermarket-branded forecourt store over a fuel-branded one, and essentially all the channel’s growth is coming from retail partners — very few fuel-branded stores are growing consistently.
• Profitability is excellent where it is done well. On-site gross profit above 30% is achievable (FreshStop claims >30%, versus mid-20s for a typical independent shop); a well-run branded store conversion has been reported to roughly double turnover and lift fuel volumes ~10%.
• The channel is now bigger than corner shops and contributes 15% of South Africa’s primary FMCG convenience channel (~R265bn), against R207bn for the informal/spaza trade.

1.3 Biggest opportunities

1. Foodservice (“foodvenience”) — barista coffee, hot ready-to-eat, fresh prepared meals. 57% of shoppers already buy hot RTE food; margins run 55–65% versus ~25% on packaged grocery. This is the single highest-return investment on a forecourt.
2. The 1,900–2,200 stores that are not yet branded. Only 849 of ~3,000+ forecourt stores carry a supermarket brand. Converting the rest is the largest addressable growth pool in the channel.
3. Township and late-night dayparts. Forecourt stores are becoming destinations specifically for younger, township and after-dark shoppers (convenience stops at midnight run at twice the rate of fuel stops). This is white space that supermarkets and spazas cannot serve at 24/7 with security and parking.
4. Loyalty arbitrage. Bank/retailer reward tie-ins move extraordinary volume: Astron Energy’s Standard Bank UCount partnership (up to R10/l cashback) drove an +83% increase in customer traffic versus competitors.
5. Services densification — parcel lockers (Pargo, Pudo), licence-disc renewal, ATM/banking, car wash, battery swap, EV charging dwell time.
6. Vape/e-cigarette category shift — significant migration from combustible tobacco, concentrated in upper-LSM stores, ahead of regulation.

1.4 Biggest risks

1. Margin compression from saturation. More sites, fewer litres, and price-sensitive shoppers. This is the number one threat to operator economics.
2. A live fuel-price shock. In 2026 the Middle East conflict and Strait of Hormuz disruption pushed inland 95 ULP from R21.41/l (Dec-25) to R23.36/l (Apr-26) and then R26.92/l (Sep-26); diesel hit a record R26.11/l. Treasury temporarily cut the fuel levy by R3.00/l. SARB expects CPI above 5% into 2027. High pump prices suppress volumes and disposable income for store baskets simultaneously.
3. Crime, shrinkage and cash handling. Forecourts are cash-heavy, 24-hour, high-value targets. Shrinkage and security cost are structural, not episodic.
4. Channel competition. Checkers Sixty60 grew 34.5% to R25.5bn in FY26; quick commerce, spaza densification and supermarket small formats all attack the same “top-up” mission.
5. Regulatory overhang. The Tobacco Products and Electronic Delivery Systems Control Bill (plain packaging, vape restriction, penalties to 10 years) directly threatens the highest-margin fixed-rent category. Fuel price deregulation debate continues.
6. Unequal competition. Operators cite unregulated informal and foreign-owned retail competition and illicit fuel trading as active margin eroders.
7. EV transition — slow, but real. NEVs were 2.8% of 2025 new-vehicle sales (16,716 units); BEVs actually fell 13.4%. Trade Intelligence expects EV refuelling to gain traction only from ~2040. This is a 2035+ problem, not a 2028 one — but site selection decisions made today have 20-year horizons.

1.5 Strategic implications

For fuel brands (Shell, BP, Engen/Vivo, Astron, TotalEnergies, Sasol, Puma)

• Treat the shop as the primary asset and fuel as the traffic generator, not the reverse. Set non-fuel gross-profit targets as a board-level KPI, not a residual.
• Choose a retail partner or build a credible own-brand — but do not do neither. Fuel-branded stores are losing the perception war. Engen (Woolworths Foodstop + Quickshop & Co + Café 365), Astron (FreshStop + Seattle), BP (Pick n Pay Express) and TotalEnergies/Puma/Sasol (OK Express, Circle K) have effectively partitioned the market.
• Shell’s exit is the decade’s structural event. ~591 sites, ~10% of the network, are in divestment (Adnoc and Gunvor reportedly shortlisted; Puma withdrew). Whoever buys it acquires SPAR Express, Vida e Caffè counters and a Wild Bean/Café estate — and will either consolidate or reshape convenience competition.
• Monetise dwell time. bp SA’s programme (up to 40 new sites by 2030, licence-disc renewal, tyre service, expanded QSR, bPOWERd battery rental) is the template.

For convenience retailers (FreshStop, OK Express, SPAR Express, Pick n Pay Express, Woolworths Foodstop, Circle K, independents)

• FreshStop is the scale player (330–350+ stores, 126 Seattle Coffee outlets, ~R1bn invested, 40m²–450m² formats, 346 of 350 sites franchised, 10-year Astron agreement, 25 new sites targeted for 2026). Competing head-on with it on the Astron network is not viable; compete on catchment-specific formats elsewhere.
• Woolworths Foodstop at Engen has hit 101 sites and the group has committed to doubling that footprint in the medium term. “Woolies After Dark” via Uber Eats runs from ~80 Foodstops to midnight and has more than doubled revenue — proof that a forecourt store is also a dark-store node.
• OK Express (Shoprite) at 128 forecourt stores across TotalEnergies, Puma and Sasol is the fastest-scaling traditional-retailer entry, backed by Shoprite’s supply chain and a franchise-friendly model. The OK Urban flagship (Wellington, WC) signals an upmarket forecourt play. Note Shoprite has explicitly ruled out putting Checkers on forecourts (cost structure and premium positioning don’t fit) — but has bought Vida e Caffè (~400 outlets incl. Shell counters), which changes the coffee map.
• Pick n Pay Express at 188 bp sites is the most mature franchise estate; PnP Go with Vivo Energy (Botswana) shows the model exporting.
• Independents (KAL Group/PEG Retail Operations with 300+ stores across 75 sites; Relish at Astron Rosebank; Pantry by Marble at Sasol) demonstrate that differentiated independents can outperform branded stores in affluent catchments.

For FMCG suppliers and distributors

• This is a R40bn channel growing at 4–5% with a 3.5-minute average dwell time. That single fact should drive your entire forecourt playbook: pack size, price point, impulse placement and cold-chain availability matter more than range breadth.
• Cold drinks are the entry ticket (75% purchase incidence), airtime second (58%), hot RTE third (57%). Energy drinks are the growth engine within beverages.
• Sell daypart solutions, not categories: breakfast combo, midday meal deal, afternoon energy hit, late-night snack. Combo offers (coffee + pie), daypart-based energy-drink discounts and cross-merchandising are the tactics operators report using to build basket.
• Build direct-to-forecourt route-to-market capability. Franchise-heavy structures (FreshStop 346/350; PnP Express; SPAR Express; OK Express) mean you are selling to hundreds of independent P&Ls, not one head office.
• Price perception is the barrier to non-shoppers, not availability. 39% of forecourt shoppers call pricing “fair” and 43% “great/good” — but for non-shoppers, price remains the key deterrent. Fund value promotions to convert them.
• Priority accounts for 2026–2028: Food Lover’s Market/FreshStop, Shoprite/OK Express (and Vida post-acquisition), Woolworths Foodstop, Pick n Pay Express, Astron Energy retail, Engen retail, KAL/PEG.

2. Market definition and scope

2.1 What “forecourt convenience retail” means in South Africa

In the South African context a forecourt is the whole trading site operating under a fuel retail (site) licence: the driveway and dispensers, the convenience store, any QSR or coffee operation, and the ancillary services. The industry’s own term for the non-fuel part is “Alternative Profit Opportunities” (APOs).

For this report, forecourt convenience retail is defined as all retail and foodservice revenue generated within the boundary of a licensed fuel retail site, excluding the fuel itself. It comprises three layers:

Where a single figure is needed, this report uses R57bn total non-fuel forecourt revenue for 2024 and R40bn for the convenience store component, consistent with the published anchor.

2.2 Included

• Convenience stores physically located on a licensed fuel retail site, whether fuel-branded (Engen Quickshop & Co, Shell Select, bp Express, Astron/Relish, Sasol, TotalEnergies), supermarket-branded (FreshStop, Woolworths Foodstop, Pick n Pay Express, SPAR Express, OK Express/OK Urban, Circle K, Pick n Pay Go), or independent/unbranded.
• QSR and coffee operations co-located on the forecourt, including drive-throughs and kiosks.
• Ancillary forecourt services listed in Layer C.
• All site types: urban/metro, highway/inter-city, rural/town, township, and dual-sided highway sites.
• All ownership structures: dealer-operated, oil-company-owned company-operated (COCO), franchisee-operated, and retail-partner-franchised.

2.3 Excluded

2.4 Definitional cautions

• “Forecourt” vs “service station” vs “site”. Used interchangeably in South African trade reporting. We treat them as synonymous.
• The 4,600 vs 5,000 vs 6,000 site-count problem. SAPIA and the Fuel Retailers Association cite ~4,600; Kalibrate has surveyed 5,000+; Wood Mackenzie cites “over 5,000”; an NEF paper cited 6,000+. The differences reflect (a) whether unlicensed/informal and dual-branded sites are counted, (b) whether commercial-only depots are included, and (c) survey vintage. We use ~4,750 (2024) rising to ~4,900 (2025) and ~5,050 (2026). See §12.4.
• Stats SA “convenience store sales at filling stations” is narrower than our Layer A — it covers quickshop and café sales at filling stations including “other” sales and trading income, from a sample of 897 enterprises drawn from a 9,325-enterprise motor-trade population. It excludes separately-franchised QSR and most services, and under-covers franchise stores classified outside SIC motor trade. It reads ~R28–30bn/annum versus our R40bn. See §12.4 for the reconciliation.

3. Market size and growth

3.1 Forecourt convenience store sales — history and forecast

CAGRs: 2024–2028 = +4.7% (published-consistent). 2024–2030 = +4.8%. Trade Intelligence’s own stated forecast for 2026–2028 is 4.5%–5.4%.

Sensitivity. If the R33bn (2023) figure is used unchanged rather than the restated series, the implied 2023→2024 jump is +21% and the 2024–28 CAGR is unchanged at 4.7% (the 2028 target is published). If the channel grows only at CPI (~3.4% 2026), 2030 lands at R48.4bn; if it grows at 6.5% (branded-store conversion accelerating), 2030 lands at R58.4bn. Planning range for 2030: R48bn – R58bn.

3.2 Total non-fuel forecourt revenue (Layers A + B + C)

Modelled, Layer A published-anchored. Layers B and C built from site counts × format archetypes (§8.4). One trade source cites “retail + fast food together generate R42bn”, which we read as a narrower cut (branded retail stores + franchised QSR only, excluding unbranded shops and services).

2024–2030 non-fuel CAGR: +6.3%. This is the number that matters strategically: non-fuel is growing about 2.5× faster than nominal fuel revenue and about 8× faster than fuel volumes.

3.3 Fuel sales for context

Volumes. South Africa consumed ~27bn litres of liquid fuels annually at the 2017–2019 plateau, falling to 23.19bn in 2020 and 25.18bn in 2021. Petrol consumption was ~11.1bn litres in 2020 and ~11.8bn in 2023 (Kalibrate). Total 2024 consumption fell 6.3%.

Modelled retail forecourt volumes (excluding bulk/commercial/aviation/marine direct supply):

Assumptions: petrol volumes decline ~2.8% p.a. (efficiency + hybridisation + hybrid work); diesel volumes decline ~1.0% p.a. (road freight and generator demand partially offset by efficiency); 2024–25 declines calibrated to the published −6.3% consumption and −4.2%/−5.6% fuel-sales figures.

Revenue at the pump (volume × average inland/coastal weighted pump price, excl. VAT where Stats SA convention applies):

3.4 Convenience as a share of forecourt and of the FMCG convenience channel

Share of total forecourt revenue

Gross-profit split Modelled: fuel GP at 7.5–8.5% of pump revenue (regulated retail margin of ~R3.06–3.15/l less retail rental of ~R0.85/l, plus diesel’s unregulated retail margin); non-fuel GP at 30% for in-store, ~55% for own-operated foodservice, ~30% net for franchised QSR, ~50% for services.

Share of South Africa’s primary FMCG convenience channel

Sources: Trade Intelligence (2023: R240bn total primary convenience market, R188bn informal, R40bn forecourt estimate in the 2023 Forecourt & Convenience Report; 2024: R207bn informal trade, forecourt at 15% of channel).

Forecourt is now larger than the corner-shop market and is the second-largest formal convenience sub-channel. Its 15% share of the channel is achieved from ~3,000 stores versus ~150,000 spaza outlets — i.e. forecourt averages ~R13m of sales per store per year against ~R1.4m for a spaza, a 9× productivity gap that explains why FMCG suppliers prioritise it.

3.5 Petrol-heavy vs diesel-heavy sites

Kalibrate’s provincial benchmarking confirms this: Cape Town has the highest shop revenue per site of the seven markets surveyed (a legacy of oil majors piloting global shop concepts there), Durban/KZN second, Gauteng third — despite Gauteng having the highest fuel throughput. Shop performance and fuel performance are only loosely coupled.

3.6 Urban/metro vs highway vs rural vs township

Modelled site shares and revenue bands, calibrated to Kalibrate provincial surveys, Trade Intelligence format data (40 m²–450 m²) and live site-sale listings (§8.5).

4. Channel structure and key players

4.1 The forecourt ecosystem

4.2 Branding and ownership models

4.3 Fuel brands — footprint and convenience strategy

4.4 Convenience brands — footprint and positioning

Scroll horizontally — the row label stays fixed.

4.5 The QSR and coffee layer

4.6 Service and enablement partners

5. Shopper behaviour and segmentation

5.1 The core behavioural shift

The defining change of the last five years is that the forecourt shopper and the fuel buyer are no longer the same person.

Read-across: the forecourt has moved from captive impulse retail to chosen destination retail. That changes the operating model — you now have to earn the visit rather than intercept it.

5.2 Category purchase incidence (what shoppers actually buy)

Source: Trade Intelligence Forecourt Retail Report 2025/26 (ranks 1–3 published; 4–7 estimated from report commentary).

5.3 Price perception

Implication: 82% of existing shoppers are satisfied or neutral on price. The growth pool is the non-shopper, and the barrier there is perception, not necessarily reality. Operators are already responding — shifting away from “convenience surcharges” toward value promotions and regular deals: coffee-and-pie combos, daypart-based energy-drink discounts, creative cross-merchandising.

5.4 Shopper archetypes

Scroll horizontally — the row label stays fixed.

Modelled archetype shares, calibrated to the published metrics above (46% non-fuel, 74% store usage, 68% takeaway, daypart data, LSM commentary). Not from a single published segmentation study.

5.5 Segmentation by geography

5.6 Segmentation by income / LSM and age

5.7 Behavioural shifts from macro forces

Hybrid / work-from-home. A primary driver of the structural fuel decline. Fewer commuting trips means fewer intercepted shoppers on weekday peak dayparts — but it also redistributes demand: suburban and township sites gain, CBD and office-node sites lose. Operators are responding by shifting assortment from “commuter breakfast” toward “neighbourhood top-up grocery” and by extending daypart relevance into mid-morning and mid-afternoon.

Economic pressure. Consumers remain highly price-sensitive; Nedbank’s economist is explicit that “margins will stay tight” and growth must come from diversification and operational efficiency. Observed responses: trading down within categories, smaller packs, more promotions, more combo deals, and a shift of scheduled shopping into value formats — including forecourt stores, which now compete on price rather than only on convenience.

Load-shedding and power reliability. Load-shedding was suspended from March 2024, returned briefly in early 2025 (Stage 6 on 23 February 2025), and Eskom has gone over a year without load-shedding as of mid-2026 (unbroken run since mid-May 2025), supported by Kusile, better maintenance and a large build-out of private solar. Actual load-shedding energy fell ~76% in 2024 versus 2023. But the operational lesson is permanent: sites invested in generators, inverters and solar to protect refrigeration, POS and dispensers, and those investments remain necessary because “there are still power outages that are not necessarily linked to load shedding” (bp SA, on its battery-rental concept). Rising electricity tariffs keep the pressure on. Backup power is now a standing cost line, not a contingency.

E-hailing and delivery apps. A structural growth driver for forecourts in three ways: (i) ~70,000 e-hailing vehicles refuel and eat at forecourts with unusual frequency; (ii) forecourt stores become fulfilment nodes — Woolies After Dark delivers from ~80 Foodstops to midnight and accounts for close to 20% of sales at participating stores; (iii) 58% of surveyed consumers use e-hailing more frequently than a year earlier (67% in Johannesburg) — Discovery Bank/Visa SpendTrend 2026, drawing on 2.6bn card transactions. Regulation formalised the sector in September 2025 (National Land Transport Act amendment, gazetted 12 Sep 2025), which should professionalise and grow it further.

Fuel-price shocks. The 2026 spike is a double-edged sword: it destroys discretionary income (bad for baskets) but increases forecourt dwell and trip salience, and it drives a switch to closer, more frequent, smaller shops — which favours forecourt convenience over destination supermarkets.

6. Product and service mix

6.1 Revenue mix within forecourt convenience (Layer A, R40bn, 2024)

Modelled — no published category-level revenue split exists for the SA forecourt channel. Built from published purchase-incidence data (cold beverages 75%, airtime 58%, hot RTE 57%), format and size distribution (40–450 m²), operator commentary on the growth of QSR/deli/coffee and the vape shift, and comparable c-store category economics in South Africa and NACS benchmarks.

6.2 Non-store forecourt revenue (Layers B and C)

6.3 Fastest-growing categories

6.4 Highest-margin categories

1. Own-operated barista coffee — 60–70% GP; ~R15–R25k capex per unit for a kiosk; the best return on incremental forecourt square metreage anywhere in the channel.
2. Own-operated hot food / prepared meals — 45–58% GP; higher capex and waste risk, but the FreshStop generation stores prove the model at scale.
3. Vapes and next-generation nicotine — 40–55% GP versus 12–18% on combustibles. Regulatory risk is the offset.
4. Confectionery and impulse snacks — 32–40% GP, low shrink if well controlled, high impulse conversion in a 3.5-minute dwell.
5. Car care and automotive accessories — 35–45% GP, low perishability.
6. Services (ATM commissions, parcel, advertising, licence renewal) — very high margin on minimal incremental cost, and they build visit frequency.
7. General merchandise, gifting, flowers — 35–50% GP.

Lowest margin but strategically essential: airtime/prepaid (3–8%), cigarettes (12–18%), staple grocery (20–26%).

6.5 Under-penetrated opportunities

7. Competitive dynamics and positioning

7.1 Forecourt convenience vs other convenience and grocery channels

Scroll horizontally — the row label stays fixed.

7.2 Positioning map

7.3 Price positioning

• The historical model — “convenience at a premium” — is under active attack. Regular shoppers accept it; non-shoppers are deterred by it. Operators are therefore pivoting to value promotions and regular deals rather than surcharges.
• Supermarket partners import price credibility. Half of consumers strongly prefer a supermarket-branded forecourt store, and the reason cited is familiarity and quality — but the practical effect is also price trust. A Pick n Pay Express or Woolworths Foodstop carries its parent’s price architecture and promotional calendar.
• The “Shoprite next door” paradox. One forecourt owner reported that a Shoprite opening adjacent to his site increased his footfall and turnover — because of increased activity on the site, shoppers wanting shorter queues, and shoppers preferring the forecourt despite higher prices because they felt they got better service. Service and speed can beat price in a convenience mission. This is the most important single qualitative finding in the report for pricing strategy.
• Diesel is the pricing lever nobody uses. Petrol retail prices are fully regulated and identical everywhere. Diesel retail prices are not regulated — only the wholesale maximum is — so diesel can differ by up to ~R1/l between garages in the same suburb. For diesel-heavy sites this is a genuine, legal competitive instrument for both fuel and cross-subsidised store promotions.

7.4 Assortment depth and localisation

Localisation is the emerging battleground. Trade Intelligence reports that forecourt owners “have a comprehensive understanding of what their customers want and are customising on the fly, including going beyond what the retail or fuel brand is able to offer.” The example given is a Shell Select dealer in Boksburg who, unable to expand his store or install a bakery, added a food truck selling sandwiches and pap-and-steak targeted at local workers. Franchise range discipline versus catchment specificity is the central operational tension in the channel for 2026–2028.

7.5 Loyalty programmes and the role of fuel loyalty in driving store spend

The mechanism that matters. Loyalty programmes are no longer primarily about discounting fuel — they are about routing an existing loyalty base onto a site so that store spend follows. Trade Intelligence’s framing is precise: these collaborations are “strategic levers to attract footfall by tapping into an existing loyalty base to draw that spend onto their site.”

Implication for FMCG suppliers: a promotion funded at a forecourt that is also earnable/redeemable through a bank or retail loyalty scheme has a materially higher conversion than the same promotion run standalone. Negotiate tri-party mechanics (supplier + fuel brand + loyalty/bank partner).

8. Operations, formats and site economics

8.1 Store formats

Layout principles now standard in the channel:

• Impulse and cold-drink wall visible from the entrance and, where possible, from the pump island.
• Coffee and hot food at the front, not the back — they are the destination categories.
• Queue architecture that separates fuel payers from store-only customers (46% of store customers don’t buy fuel).
• Self-service where safe; staffed service for tobacco, vaping and alcohol-adjacent categories.
• Seating and, increasingly, plug points and Wi-Fi — dwell time is the new metric.
• Sightlines and CCTV coverage designed for shrink control, not just safety.

8.2 Operational considerations

Staffing. MIBCO Sector 5 (fuel retail) sets minimum wages for forecourt attendants, cashiers and “chars” (support staff). The 23 August 2025 multi-year wage settlement included a concession that the petrol retail margin rise by 6.1 c/l, from 299.5 c/l to 305.6 c/l, effective 1 October 2025, explicitly to help fund the wage increase — a clear signal of how tight site-level labour economics are. A typical site runs:

Skill requirements are shifting: foodservice (barista, hot-food prep, HACCP), fresh-food handling, POS/analytics literacy, cash and shrink management, and customer experience. Historically the sector employed ~58,000 pump attendants; a SARB review found forecourt employee numbers grew 19% and cashier numbers 52% between 2006 and 2022 while sales volumes stagnated — evidence of the shift from fuel labour to retail labour.

Supply chain and distribution.

Shrinkage, security and cash handling. This is the channel’s structural cost problem. Forecourts are 24-hour, cash-heavy, high-value, publicly accessible targets. Mitigations now standard: CCTV and analytics, cash-management services (armoured collection, smart safes, cash recyclers), digital-payment migration (South Africans completed 118+ card transactions per person in 2025; total card volume R2.9tn), stock-control discipline and cycle counting, controlled tobacco/vape cabinets, and staffing to visible levels at night. Trade commentators specifically recommend “surveillance cameras to fight shrinkage and shoplifting, digital payments acceptance” as profitability levers.

Load-shedding and power resilience. Load-shedding has been effectively suspended (no load-shedding since mid-May 2025; ~76% reduction in load-shedding energy in 2024 vs 2023). But backup capacity is now a permanent requirement:

Case evidence of the cost of not investing: one industrial case study recorded generator operating costs of R86,628 during load-shedding versus R3,249 without — a 27× differential. Refrigeration loss during outages is the more damaging risk for a food-led forecourt: it destroys the foodservice margin advantage that justifies the whole strategy.

8.3 Fuel margin structure (the regulated core)

Petrol price build-up, December 2025 (CEF Group official schedule; inland, 95 ULP; c/l):

Note: the December 2025 schedule above reflects the year-end price. The published retail price history shows inland 95 ULP at R21.41/l in Dec-2025 and R26.92/l by Sep-2026 — the CEF schedule cited is the component build-up as at the relevant adjustment, and readers should treat the components (not the total) as the stable reference. Consumers pay R5.59/l in indirect taxes on petrol and R5.19/l on diesel; the RAF levy alone is R1.98/l.

Retail margin history and the wage linkage:

How the retail margin is actually split at site level (industry convention, varies by lease):

At 300,000 l/month (the FRA’s stated viability threshold), a mixed petrol/diesel site therefore generates roughly R280k–R450k/month of fuel gross margin before any retail contribution.

8.4 Site economics — three worked archetypes

Modelled — all figures are monthly, ZAR, before tax and before owner remuneration unless stated. Built from published anchors: FRA viability threshold (>300,000 l/month), Kalibrate provincial throughputs (Gauteng 282 kl/mth, Limpopo 206 kl/mth), Trade Intelligence convenience revenue per litre (R1.36), FreshStop GP >30%, live site-sale listings (§8.5), MIBCO wage settlement, CEF margin schedules, and Absa’s capex guidance.

Archetype A — Urban metro, branded express store

Archetype B — Highway / inter-city, large format + QSR

Archetype C — Township / rural, small independent store

• Break-even on fuel alone sits at roughly 250,000–300,000 l/month for a modern site with full labour and compliance costs. The FRA’s stated benchmark is “>300,000 litres upwards… the majority are actually below that benchmark.”
• A convenience store at R800k/month adds ~R240k of gross profit — about 65% as much as the entire fuel business. That is the whole strategic argument in one number.
• Non-fuel contributes 27–47% of site gross margin depending on format, and the share rises with store size and foodservice depth.
• Labour is the single largest controllable cost (55–60% of opex). Foodservice increases labour but increases margin faster — provided waste and productivity are managed.

8.5 Capex benchmarks

8.6 Revenue per litre and revenue per site

8.7 Profitability drivers and break-even

9. Regulatory, macro and risk environment

9.1 Regulatory framework

9.2 Macro environment

Growth and inflation

Sources: National Treasury 2026 Budget Review; SARB MPC statements (May, July, September 2026); Deloitte Africa Economic Outlook.

The 2026 fuel-price shock. Middle East conflict and Strait of Hormuz disruption (the strait carries ~20% of global oil supply) pushed Brent above $100/barrel; the rand weakened from R16.00 to R16.64/USD in the review period. April 2026 delivered the largest fuel price increases in South African history: inland 93 petrol from R20.19 to R23.25/l, 95 from R20.30 to R23.36/l, diesel 0.05% from R18.53 to R25.90/l and 0.005% to R26.11/l (records). National Treasury temporarily cut the general fuel levy by R3.00/l to cushion motorists. By September 2026 inland 95 ULP stood at R26.92/l. SARB reported an average under-recovery of R2.83/l at the time of the September MPC.

Why this matters for forecourt retail: it is simultaneously (a) a volume destroyer, (b) a disposable-income destroyer for store baskets, (c) an inflation driver that lifts store prices and hence nominal store revenue, and (d) a trigger for trip-chaining and closer, more frequent, smaller shops — which favours forecourt convenience. Net effect on our forecast: upside to nominal convenience revenue in 2026–27, downside to real volumes. We have kept our base case inside the published 4.5–5.4% band rather than upgrading, because the disposable-income effect offsets the price effect.

Fuel demand and vehicle parc

Exchange rate and import costs. SA is now a net importer of road transport fuels. The basic fuel price is set off Mediterranean/Arab Gulf/Singapore product prices plus freight, insurance, cargo dues and storage, converted at the four-bank selling rate averaged over the 26th-to-25th window. Rand weakness therefore transmits directly and immediately to the pump — and to consumer wallets. A weaker rand simultaneously raises imported confectionery and novelty range costs for operators such as Circle K.

Electricity. Load-shedding is effectively over (no load-shedding since mid-May 2025; ~76% reduction in load-shedding energy in 2024 vs 2023; 281 consecutive days without load-shedding by 31 Dec 2024). Eskom’s winter 2026 outlook projected none. But tariffs keep rising and site-level outages persist, so backup power remains a mandatory opex line.

Consumer spending. Household consumption growth moderated to ~1.8% in 2026 and is forecast to average 2.0% from 2026 to 2028, supported by rising wages, lower inflation, gradual rate easing, stable power supply and improved sentiment. Consumers remain “highly price-sensitive” with “margins staying tight” (Nedbank). Card usage is a bright spot: 118+ transactions per person in 2025, R2.9tn total card volume.

9.3 Risk register

10. Technology and innovation

10.1 Current state in South Africa

10.2 Emerging and adjacent

• Kitchen and foodservice technology — the “foodvenience” model requires demand forecasting, waste management and store-level food analytics that most SA forecourt operators do not yet have. Global Convenience is explicit: “The technology infrastructure to support this sophistication, from field execution tools to performance analytics, has become non-negotiable.”
• Field execution and retail audit tools for franchise networks — critical when 346 of 350 FreshStop sites, all 188 PnP Express sites and all Foodstops are franchisee-owned.
• Scan-and-go / self-checkout — proven internationally; constrained in SA by shrink risk and cash prevalence, but viable in premium, well-staffed formats (Relish, Pantry, FreshStop flagships).
• Dynamic daypart pricing — already emerging informally (“daypart-based energy drink discounts”). Technology-enabled versions are the next step.
• Computer vision for pump-to-store conversion analytics — tracking how many fuel customers enter the store, and what they buy.
• Energy management — solar + battery orchestration, refrigeration telemetry, generator run-time optimisation.
• Digital advertising / retail media on forecourt — canopy screens, pump screens, in-store digital signage. Untapped in SA; a genuine high-margin revenue line given the ~10 visits/month frequency.

10.3 Global best practice adaptable to South Africa

10.4 The strategic technology insight

Trade Intelligence’s framing is the right one: “Every square metre of the forecourt is now an opportunity for attention.” With a 3.5-minute average in-store dwell and ~10 monthly site visits, the forecourt is one of the highest-frequency, lowest-attention retail environments in South Africa. Technology’s job here is not checkout speed — it is (a) extending dwell (pre-order, seating, Wi-Fi, services), (b) extending reach (delivery apps, pre-order), and (c) extending knowledge (site-level analytics that let a franchisee localise faster than the brand can).

11. Opportunities and strategic recommendations

11.1 Opportunity map, prioritised

Scored 1–5 on Impact (revenue/profit potential) and Feasibility (capex, complexity, time to implement). Priority = Impact × Feasibility.

Scroll horizontally — the row label stays fixed.

11.2 For fuel retailers wanting to grow non-fuel income

Format and site selection

1. Set a non-fuel gross-profit target as a board KPI. Move from “non-fuel % of revenue” (~12%) to “non-fuel % of gross profit” (~34%, target 45%+ by 2030). The revenue metric systematically understates the strategic importance of the shop.
2. Triage the estate into four tiers by throughput and catchment: (a) >500 kl/month → destination/multi-brand convenience centre; (b) 300–500 kl → branded express with foodservice; (c) 150–300 kl → standard c-store with coffee; (d) <150 kl → micro kiosk, consider exit or franchise-out.
3. Stop approving new sites on fuel volume alone. Require a catchment retail-demand model (Kalibrate-type micro-market analysis) alongside the fuel model. Gauteng’s 19 competing brands and 68% top-four concentration, with throughput down from 304 to 282 kl/month, is the cautionary tale.
4. Prioritise the 24/7 urban and township nodes. Midnight convenience stops run at twice the rate of fuel stops; this is a daypart with almost no formal competition.

Assortment and localisation

1. Give franchisees a localisation allowance. Define a core range (70–80%) and a catchment-discretionary range (20–30%). The Boksburg food-truck example — a Select-store dealer who couldn’t build a bakery so he bought a food truck and sold pap and steak to local workers — is exactly the behaviour the system should reward, not penalise.
2. Build the food first, the grocery second. Foodservice is 25% of store sales and 44% of store gross profit. Coffee is the highest-return single investment on a forecourt.
3. Fix availability before range. Cold drinks are bought by 75% of shoppers; an out-of-stock cold cabinet is the most expensive failure in the channel.

Pricing and promotion

1. Move from “convenience surcharge” to “value promotion.” 82% of existing shoppers rate pricing fair-or-better; the barrier is perception among non-shoppers. Fund visible value deals to convert them.
2. Use diesel pricing discretion. It is the only unregulated fuel margin and can differ by ~R1/l between neighbouring sites. Use it to fund store promotions at diesel-heavy sites.
3. Engineer combos around dayparts — breakfast (coffee + pie), lunch (meal deal), afternoon (energy drink discount), late night (snack + drink). Operators already report these tactics; systematise them.

Partnerships

1. Pick one supermarket partner per tier and commit for a decade. FreshStop/Astron’s newly renegotiated 10-year agreement and the Woolworths/Engen partnership (26 years, 101 sites, doubling planned) both show that the returns compound over long horizons — FreshStop reports >60% compound retail sales growth over 10 years of trading.
2. Do not run a fuel-branded store where a supermarket partner is viable. Very few fuel-branded stores are growing consistently; half of consumers strongly prefer a supermarket brand.
3. Broker tri-party loyalty deals. Bank + fuel brand + retailer is the highest-leverage structure available. Astron/UCount’s +83% traffic is the benchmark.
4. Prepare for the Shell estate to change hands. ~591 sites, ~10% of the network, ~US$1bn. Whoever acquires it inherits SPAR Express, Vida e Caffè (Shoprite-owned) counters and a Wild Bean legacy — and will re-cut convenience partnerships. Have a position ready.

Technology and loyalty

1. Integrate the till with the pump. FreshPos is the direction of travel; a unified transaction record is the precondition for any serious loyalty or conversion analytics.
2. Turn on delivery from every viable store. Woolies After Dark delivers ~20% of sales at participating sites with cold-chain integrity to midnight. This is available technology, not R&D.
3. Instrument dwell and conversion. Measure pump-to-store conversion, dwell time, daypart mix and basket composition per site. Most operators cannot.

11.3 For convenience operators considering forecourt expansion

1. Target the ~2,150 unbranded stores. 849 forecourts have a supermarket-branded store out of ~3,000+ with any store. That is the addressable pool.
2. Choose your fuel partner by what’s left, not by what’s biggest. Engen→Woolworths, Astron→FreshStop, bp→Pick n Pay, TotalEnergies/Puma/Sasol→OK Express/Circle K, Shell→SPAR Express (in play). The realistic openings are: the Shell estate post-divestment, Sasol (Mark-V redesign creates a natural conversion window), and the ~10% independent/unbranded network.
3. Match format to catchment, not to brand. 40 m² kiosks to 450 m² flagships. FreshStop’s three-tier size concept is the right architecture. Do not force a 300 m² store onto a 150 kl/month site.
4. Franchise, don’t own. 346 of 350 FreshStop sites are franchised; PnP Express, SPAR Express and OK Express are franchise models. Franchising gives capital efficiency, local ownership motivation and faster rollout — and it lets the forecourt owner share in the upside, which is what makes the partnership stick.
5. Bring your own supply chain or don’t come. The reason Food Lover’s/FreshStop can offer fresh food nationally is 10 DCs, 5 production facilities and 2 meat-packing plants. This is the moat. Competitors without it should partner rather than attempt fresh.
6. Use the forecourt as a dark-store node. 24/7 sites, parking, security and existing cold chain make forecourts the cheapest last-mile fulfilment real estate in the country. Woolies After Dark is the proof.
7. Consider the coffee adjacency seriously. Shoprite’s ~R1bn acquisition of Vida e Caffè (~400 outlets, forecourt among its formats) tells you what the largest retailer in the country thinks coffee is worth.

11.4 For FMCG brands seeking to win in this channel

1. Design for 3.5 minutes. That is the average in-store dwell. Pack size, price point, shelf position and grab-ability matter more than range. Single-serve, immediate-consumption formats win.
2. Prioritise cold. Cold beverages are bought by 75% of forecourt shoppers. Cabinet space and fill rate are the two metrics to manage with the operator — negotiate them, don’t assume them.
3. Sell daypart solutions. Breakfast, lunch, afternoon lift, late night. Build promo mechanics around them (coffee + pastry, meal deal, energy-drink daypart discount).
4. Treat energy drinks as a strategic growth category, not a soft-drink sub-line. It is the fastest-growing beverage segment in the channel and skews young.
5. Build a franchise-capable route to market. You are selling to hundreds of independent P&Ls (346 FreshStop franchisees, 188 PnP Express franchisees, 128 OK Express, 101 Foodstops, plus fuel-brand and independent operators). Head-office agreements do not execute themselves.
6. Fund the value story. Price perception is the barrier for non-shoppers. Supplier-funded promotions convert them — and operators report actively seeking “creative options (including on specials and deals).”
7. Localise by catchment. Township, rural, highway and metro stores are different businesses. Upper-LSM stores are shifting to vapes; township stores are driven by airtime, prepaid electricity and small-pack staples; highway stores by meals and family packs.
8. Engage on foodservice ingredients and equipment, not just packaged goods. The channel’s growth is in prepared food — supply it (protein, bakery, dairy, sauces, packaging, equipment, training).
9. Negotiate tri-party promotions with the fuel brand and the loyalty/bank partner. The loyalty base is what drives site choice.
10. Prepare for tobacco regulation. If the Tobacco and EDS Bill passes with plain packaging and flavour restrictions, a 13%-of-sales category gets materially harder. Diversify your forecourt portfolio toward food, beverage and services now.
11. Priority account list for 2026–2028: Food Lover’s Market / FreshStop (350+ sites); Shoprite / OK Express (128) + OK Urban + Vida e Caffè (~400); Woolworths Foodstop (101, doubling); Pick n Pay Express (188); Engen retail (700+ Quickshop & Co, 365+ Café 365); Astron Energy retail (850 sites); SPAR Express / Shell estate (in play); TotalEnergies; Sasol (Mark-V rollout); Puma / Circle K; KAL Group / PEG Retail Operations (300+ stores).

11.5 A three-horizon roadmap

12. Appendix: data tables, assumptions and sources

Disclaimer and confidence statement

This report synthesises published trade research, official statistics, company disclosures and business media as at 30 September 2026. High confidence attaches to: the R40bn (2024) convenience anchor and R48bn (2028) forecast; the 4.5–5.4% growth band; branded-store counts (FreshStop 330–350+, PnP Express 188, OK Express 128, Woolworths Foodstop 101, Seattle 126, 849 supermarket-branded forecourts); fuel brand site counts; the −6.3% 2024 fuel consumption decline; shopper metrics (46% non-fuel, 3.5-minute dwell, 75%/58%/57% category incidence, +83% UCount traffic); macro data (repo 7.25%, CPI 4.4%, GDP 1.2% 2026, 2026 fuel-price shock); and NEV data.

Medium confidence attaches to: the pre-2023 convenience series (re-based by the publisher); retail fuel volumes and revenue (modelled); total non-fuel revenue including QSR and services; the site-count reconciliation; and category-level revenue splits.

Lower confidence / directional only attaches to: shopper archetype shares; the gross-profit split between fuel and non-fuel; site-level P&L archetypes; 2029–2030 forecasts; and township/rural revenue bands. These are labelled Modelled at every occurrence and the assumptions are registered in §12.5.

Where sources conflict, §12.4 states the conflict, the resolution and the figure used. Readers making capital commitments should commission primary site-level research (catchment modelling, store audits, franchisee P&L benchmarking) to validate the archetypes in §8.4 against their own estate.

Prepared 30 September 2026. Companion workbook: SA_Forecourt_Data_Appendix.xlsx.