A request for proposal lands on a Monday morning. It is sixty pages long, the deadline is six weeks away, and within the hour someone has opened last year’s proposal template and started copying sections across. It feels productive. It is usually the most expensive mistake a bid team makes, because it turns urgency into activity before anyone has checked whether the opportunity is winnable, deliverable or worth the money.
No honest method guarantees every bid. Price, an incumbent supplier, procurement rules, mandatory qualifications, internal politics, timing and the client’s appetite for risk can all outweigh technical merit. The realistic goal is different and more useful: win more of the opportunities that suit you, stop weak pursuits early, and learn something from every result.
This series is about how to do that. It is written for solution architects, proposal writers, technical leads, sales teams and the executives who sign bids off, and it assumes no prior knowledge of the jargon.
What a strong proposal has to prove
Strip away the formatting and a proposal is an argument. A strong one proves four things:
- The bidder understands the client’s situation and the outcome the client actually wants.
- Evaluators can verify compliance and award points without guessing.
- The solution, the transition to it, and the way it will be run afterwards are believable.
- Choosing this bidder creates acceptable delivery and commercial risk for the client.
Architecture supports all four. It connects the business problem to the proposed change, the delivery plan, the way the service will operate and the price. It is not a decorative technical appendix that someone attaches at the end. If the diagram shows something the price does not include, or the plan cannot reach what the diagram promises, an experienced evaluator will notice.
The series in one table
| Part | Topic | What it produces |
|---|---|---|
| 1 | Win before you write | A bid/no-bid decision |
| 2 | Read the client and the room | A stakeholder and decision map |
| 3 | Discovery | An evidence and clarification register |
| 4 | Requirements and assumptions | A traceability matrix |
| 5 | Win themes and value | Win themes and a value case |
| 6 | TOGAF, CAF and WAF | Framework evidence, tailored to the bid |
| 7 | Architecture layers and views | A purposeful architecture pack |
| 8 | Flows and ownership | End-to-end operating evidence |
| 9 | Transition | As-is, bridge states and to-be |
| 10 | Estimate and price | Work packages and commercials |
| 11 | Write and present | An evaluator-led response |
| 12 | Red-team and learn | Assurance and improvement |
The running example: Rivermark Water
Advice about bids is easy to agree with and hard to apply, so every part of this series follows one pursuit from start to finish. Rivermark Water and Ridgeline Digital are fictional, and all figures are illustrative. Any resemblance to a real organisation is coincidental.
The client. Rivermark Water is a regional water services utility with about 380,000 residential and business customer accounts. Its contact centre has 60 agents across two shifts, and in peak periods callers wait around 14 minutes. Customers report leaks, bursts and outages by phone, email and a WhatsApp number that three agents monitor by hand. About 9,000 fault reports arrive in a normal month, rising to about 30,000 in a bad storm week, alongside roughly 14,000 billing queries. Duplicate reports are common and nobody gets a status update. Billing disputes take about 21 days, and an internal audit found that complaints were not tracked from start to finish.
Behind the scenes, a 20-year-old on-premises billing system does the heavy lifting. Its vendor support ends in 30 months, and replacing it is a separate procurement for another day. A tariff change on 1 July reliably triggers a spike in billing queries, and Rivermark wants self-service in place before then.
The request for proposal (RFP). Rivermark has issued an RFP for a Customer Self-Service and Case Management Platform: a mobile-friendly customer portal, a WhatsApp channel, case management for faults, billing queries and complaints, a console for agents, integration with the billing system and the field teams’ job app, notifications and reporting. It must be hosted in South Africa and comply with POPIA, the Protection of Personal Information Act. The contract runs for 36 months: about a year to build and transition, then two years of managed service. The indicative budget is about R36 million over three years. Bidders have six weeks, with a compulsory briefing session in week 1 and clarification questions closing at the end of week 3.
How it will be judged. This RFP evaluates in stages. First come mandatory gates: registration on the government supplier database with tax compliance, a B-BBEE certificate, signed standard bidding forms, an information security policy, and two reference letters for a case management implementation serving at least 100,000 customers. Bids that pass are scored for functionality out of 100, and must reach 70 to go further. Remaining bids are then scored on price (80 points) and specific goals (20 points). Shortlisted bidders may be invited to a 45-minute presentation.
The bidder. Ridgeline Digital is a 45-person Johannesburg delivery partner, strong in integration, cloud platforms and service design, and it has run a managed service before. It has one qualifying reference, a case management rollout for 140,000 customers, and another for 60,000, which does not meet the gate. Remember that detail.
The first decision is whether to bid
A pursuit is worth serious investment only when three conditions overlap:
- The client has a real problem, with an owner, a path to funding and a decision process.
- The bidder has a credible right to win: people, references, assets, partnerships or insight that make it believable.
- The work can be delivered at acceptable risk and return.
Think of buying a house. You do not make an offer because you like the kitchen. You check that the seller genuinely wants to sell, that your offer can compete, and that you can actually get the finance and afford the repayments. Miss any one of those and the effort of making an offer is wasted, or worse, it succeeds and hurts you.
For Rivermark, the first and third circles were strong. The need is explicit, urgent, owned by the Head of Customer Services and driven by a date. Ridgeline has delivered similar integration work and run a managed service. The weak circle was the second one, because of that reference gate.
The qualification questions
Qualification means testing the opportunity before committing effort to it. Ridgeline’s bid owner worked through ten areas in a short meeting with the lead solution architect, the delivery lead and the commercial lead. The right-hand column shows what they found.
| Area | Questions to ask | What Ridgeline found |
|---|---|---|
| Problem | Is the need explicit, urgent and owned? What happens if nothing changes? | Yes. Queries spike on 1 July, and the audit finding will not go away. |
| Decision | Who evaluates, influences, approves and blocks? Is the scoring method known? | Scoring is published. The evaluation panel is not named. |
| Funding | Is the budget approved, indicative or still being sought? | Indicative, about R36 million. Approval status to be clarified. |
| Access | Have we spoken to the client? Can material questions be clarified? | The briefing session and a clarification window give real access. |
| Fit | Which people, references, assets or insights make us credible? | Strong integration and managed service experience. One qualifying reference, not two. |
| Competition | Is there an incumbent or a specification shaped around another bidder? | The old ticketing tool’s supplier may bid. Nothing looks tailored to them. |
| Delivery | Can the right people mobilise? Are dates and dependencies plausible? | Yes, if billing integration is simpler than it looks. To be tested in discovery. |
| Commercial | Can we price competitively without removing necessary work? | Probably, but price carries 80 of the final 100 points. |
| Compliance | Can every mandatory condition be met exactly and evidenced? | No. The reference gate fails as things stand. |
| Reputation | Would winning strengthen us, or create an unsafe commitment? | A public utility reference would strengthen the business. |
Hard gates come before weighted scores
There are two kinds of test in that table, and they must not be mixed.
A hard gate is pass or fail. A missing mandatory certificate, an impossible start date or a contract condition you cannot accept disqualifies the bid, however good everything else is. A weighted score is a judgement of how attractive the opportunity is, such as fit, access or delivery confidence, added up into a total.
Teams often score first and check gates later, and the attractive total quietly overrides the fatal flaw. Ridgeline’s weighted self-assessment came to 78%, which looks like an obvious bid. It was irrelevant, because one reference letter short means the proposal is never scored at all.
Four possible decisions
A qualification meeting should end with one of four decisions, recorded in writing:
- Bid. There is a credible route to win and deliver.
- Conditional bid. Proceed only if named issues are resolved by a named deadline. If they are not, stop.
- Partner. Another organisation closes a material gap in capability, access or evidence.
- No-bid. Protect proposal capacity for better opportunities and record why.
Ridgeline chose partner, made conditional. The case management software Ridgeline planned to propose has a local implementation partner, and that partner holds a qualifying reference for a large customer base. Ridgeline proposed a teaming agreement: a contract between the two firms setting out who leads the bid, who does which work, who carries which risk and how the reference will be presented honestly. The condition was that the agreement had to be signed by the end of week 1, or Ridgeline would withdraw. It was signed on the Thursday, and the bid went ahead.
Two cautions apply to this move. The partner’s reference must genuinely match what the RFP asks for, and the proposal must describe the partner’s real role rather than borrowing its credentials for work it will not do. Evaluators check, and a borrowed reference that falls apart at the presentation does more damage than a no-bid.
A no-bid is not a failure. Recording the reason builds a picture over time of where your right to win is weak, which is exactly the insight a business needs before deciding which partnerships or capabilities to invest in. There is more on the strategic side of this in a go/no-go gate is a strategy tool, and on automating the first pass in building a tender AI agent.
The pursuit brief
Once the decision is made, one page captures what the team believes about the opportunity. It is short on purpose. Everyone working on the bid reads it, and it is revisited when new facts arrive. The questions come first; Ridgeline’s answers after the teaming agreement are on the right.
| Question | Ridgeline’s answer for Rivermark |
|---|---|
| Client outcome: what must become better? | Customers can report faults, see progress and resolve billing queries without calling, before 1 July. |
| Decision date | Award expected about ten weeks after submission, according to the RFP timetable. |
| Evaluation model | Mandatory gates, functionality threshold of 70, then price 80 and specific goals 20. |
| Decision group | Head of Customer Services, CFO, IT manager, procurement, plus an unnamed evaluation panel. |
| Right to win | Integration and managed service experience, plus the partner’s large-scale reference. |
| Likely competitors | The current ticketing tool’s supplier and at least one large systems integrator. |
| Early win themes | Live before 1 July; one record of every fault; a service Rivermark can run. |
| Solution hypothesis | Case management with portal and WhatsApp, reading billing data without replacing it. Initial, not a commitment. |
| Commercial boundary | Must fit within about R36 million over three years with a healthy margin. |
| Top risks | Billing integration effort, storm-week volumes, messaging costs, data location. |
| Clarifications required | How billing data can be accessed; where WhatsApp messages are processed; peak volumes. |
| Bid owner | One accountable person: Ridgeline’s bid owner. |
Notice that the solution is labelled a hypothesis. At this stage it is a guess worth testing, and Part 3 shows how one clarification answer changed it substantially.
The last row matters more than it looks. A bid with two owners has none. One person decides priorities, resolves disagreements and signs off the submission.
Effort does not create entitlement
The hardest part of qualification is emotional. After a week of work a team feels it has earned the right to submit, and a bid owner who stops a pursuit can feel like the person who wasted everyone’s time. The opposite is true. Proposal capacity is finite, and a strong team spread across three weak bids will lose all three. Qualification turns that capacity into a deliberate investment.
What to take from this part
Decide whether to bid before anyone writes. Starting a template is not progress if the opportunity cannot be won or delivered.
Look for all three circles. A real client need, a credible right to win and acceptable delivery risk. Any two on their own lead to a predictable kind of failure.
Check hard gates before scoring. A failed mandatory condition beats a strong weighted total every time.
Use all four decisions. Bid, conditional bid, partner and no-bid are all legitimate, and a recorded no-bid is useful evidence.
Write a one-page pursuit brief with one owner. It keeps the team aligned and makes assumptions visible from day one.
Next in the series: There is more than one client: reading stakeholders, procurement and the room.
How CloudNala can help
We sit in on qualification meetings as an independent voice, usually for a single hour before any writing starts. We work through the gates, the three conditions and the pursuit brief with the bid team, and the most valuable output is often a clear, recorded reason to stop, or a specific condition that must be met before anyone continues.
Work with CloudNala
CloudNala helps organisations move from technology ambition to practical execution across cloud, AI, data, platform engineering and digital services.
Whether you are exploring AI, modernising your cloud environment, building a public-sector digital service, or turning an idea into a working MVP, we can help you shape the roadmap and deliver the next step.
Book a Bid Review or write to us at consult@cloudnala.co.za