The verdict in three sentences
A retail group with 45 stores launching omnichannel e-commerce is not buying a website, it is buying integration with its tills, store stock and logistics: that is where budget and risk sit. A well-built tender in Riyadh targets a 2026 order of magnitude of 320,000 to 670,000 USD in build cost, scores bids on a 60% technical and 40% price grid, and compares vendors on a 3-year TCO rather than upfront price. A realistic plan is 6 to 8 months to click and collect in every store.
Writing the specification: what drives budget
Bids for an e-commerce tender often vary by a factor of three. The gap almost always comes from how integrations are interpreted. A precise specification of flows (stock, prices, orders, customers, returns) narrows that gap and makes bids comparable.
| Lot | Expected scope | Indicative 2026 range | Main risk |
|---|---|---|---|
| E-commerce platform and front end | Catalog, product pages, search, account, mobile first, Arabic RTL | 95,000 to 190,000 USD | Mobile performance |
| POS integration | Sales sync, unified loyalty, in-store returns | 55,000 to 120,000 USD | Till software API |
| Real-time store stock | Availability per store, reservation, safety thresholds | 48,000 to 105,000 USD | Accuracy of book stock |
| Click and collect and ship from store | In-store picking, notifications, pickup | 40,000 to 95,000 USD | Store team adoption |
| Payment | mada, Apple Pay, cards, cash on delivery, buy now pay later | 22,000 to 48,000 USD | Decline rates and fraud |
| ERP and PIM connector | Prices, promotions, product master | 40,000 to 80,000 USD | Product data quality |
| Testing, training, change management | 45 stores, store manager training | 20,000 to 32,000 USD | Rollout timeline |
The sum of the low ends gives 320,000 USD, the high ends 670,000 USD. A bid far below 320,000 USD for this scope almost always signals an underestimated POS integration.
Scoring grid and 3-year TCO
A 60% technical and 40% price weighting protects the group from the cheapest but riskiest bid. Within the 60%, scoring should reward proof (omnichannel references, a demo on your data) over promises. In Riyadh, the shortlist typically opposes Salesforce Commerce Cloud and a custom or open-source build.
| Criterion | Weight | What is scored | Evidence required |
|---|---|---|---|
| Architecture and integrations | 20% | POS, stock, ERP flows | Detailed integration diagram |
| Comparable references | 12% | Multi-store groups, click and collect | 2 verifiable client contacts |
| Methodology and plan | 10% | Milestones, testing, pilot | Week-by-week plan |
| Proposed team | 10% | Project lead, architect, local presence | Named CVs |
| Security, performance, SLA | 8% | 99.9% uptime, load time, data residency | Contractual commitment |
| Build price | 20% | Fixed price per lot | Price schedule |
| 3-year run cost | 20% | Licences, hosting, maintenance | TCO detailed by year |
The 3-year TCO often changes the ranking. Salesforce Commerce Cloud usually charges a share of online gross merchandise value, often 1 to 3%, with minimum annual commitments that can reach 150,000 to 250,000 USD. A custom or open-source build costs more to build but 50,000 to 95,000 USD a year in hosting and maintenance.
Typical tender and project calendar
Weeks 1 to 4: specification and shortlist of 5 to 8 vendors. Weeks 5 to 8: bids and presentations. Weeks 9 to 10: negotiation and contract. Then 6 to 8 months of delivery: scoping (1 month), build (3 to 4 months), pilot in 5 Riyadh stores (1 month), rollout to all 45 stores (1 to 2 months).
Mini case study
Need a professional website?
Kolonell builds websites that attract clients, optimized for the Sénégalese market. Free quote in 2 minutes.
Faisal, digital director of a 45-store retail group in Riyadh, compares two finalists. Bid A, on Salesforce Commerce Cloud, costs 380,000 USD to build and 200,000 USD a year in licences and run, a 3-year TCO of 980,000 USD. Bid B, custom built, costs 560,000 USD to build and 80,000 USD a year, or 800,000 USD over 3 years. With a target of 4% of revenue online over 3 years, about 9.6 million USD of cumulative web sales, the 180,000 USD gap is 1.9 points of web margin. Faisal picks bid B on TCO, after checking that POS integration was priced as a fixed fee and that hosting meets Saudi data residency rules.
FAQ
Should the budget be published in the tender?
Giving a range, for example 320,000 to 670,000 USD, improves bid quality. Without one, 30 to 40% of bids received are off target and cost several weeks.
How many vendors should be consulted?
Five to eight on the shortlist, three in presentations. Beyond that, analysis costs the internal team more than 15 person-days without improving the choice.
Is cash on delivery still needed in Saudi Arabia?
Its share has fallen sharply with mada and Apple Pay, but it still covers 15 to 30% of orders in some categories. Click and collect with payment at pickup cuts parcel refusals to under 5%.
Who owns the code and data?
The specification must require it: group ownership of customer data and custom developments, priced reversibility, full export within 30 days.
Can we start with only some stores?
Yes, a 5-store pilot over 4 to 6 weeks is recommended to validate stock accuracy and in-store picking before extending to all 45 locations.
Let's scope your project. Share your store count, POS software and tender calendar: we will reply with a detailed bid per lot, a 3-year TCO and a 6 to 8 month plan. Detailed quote within 48 h. WhatsApp +221 77 596 93 33.
Mohamed Bah
Fondateur, Kolonell
Passionate about digital and entrepreneurship in Africa, Mohamed has been helping Sénégalese businesses with their digital transformation since 2020. Founder of Kolonell, he believes every SME deserves a professional and accessible online présence.
