Inside the Scoring Room: What Evaluators Actually Reward in Technical Proposals

​The proposal isn't judged the way most firms think it is

​Engineering consultancies preparing technical proposals for World Bank, AfDB, or national government assignments in Africa tend to write for an imagined reader who weighs eloquence, comprehensiveness, and design ambition. The reality inside an actual evaluation panel is narrower and far more mechanical: a fixed set of weighted criteria, published in advance in the Request for Proposals (RFP), scored independently by multiple evaluators against a rubric the procuring entity is legally bound not to deviate from. Firms that understand this scoring architecture, along with the growing body of African case law testing its limits, write measurably stronger proposals than firms that don't.

​The Fixed Architecture: Technical First, Sealed, Then Financial

​Under the World Bank's Guidelines for Selection and Employment of Consultants, which underpin most donor-financed consulting assignments across Africa, Quality- and Cost-Based Selection (QCBS) proceeds in a strict sequence: technical proposals are evaluated and scored in full before financial proposals are even opened. The technical evaluation report must substantiate the scores assigned to each proposal, describing the relative strengths and weaknesses against each criterion. Where individual evaluators' scores diverge significantly on the same criterion, the evaluation report must document a justification for the discrepancy. Financial proposals cannot be opened until the Bank has given its no-objection to the technical evaluation report, and, critically, technical scores cannot be changed once the financial envelope is opened.

​This sequencing matters practically: a firm cannot "buy back" a weak technical score with an aggressive fee, and a client's evaluators cannot retroactively adjust a technical score once they see who priced lowest. The two stages are procedurally firewalled from each other.

​The technical-to-financial weighting itself is fixed in the RFP before any envelope is opened and cannot be renegotiated after submission. Depending on the complexity of the assignment and the funding agency's guidance, ratios commonly range from 70:30 to 90:10 in favour of technical quality, meaning on a typical infrastructure design or supervision assignment, technical merit can carry two to four times the weight of price. A firm that treats fee competitiveness as its primary lever is optimising the smaller half of the scoring formula.

​What the Technical Score is Actually Built From

​Within the technical envelope, evaluators are not scoring a single holistic impression. Instead, they are filling in a rubric against discrete, weighted sub-criteria set out in the RFP's evaluation grid. The recurring building blocks across World Bank and AfDB-financed consulting assignments are:

• ​Specific experience of the firm relevant to the assignment
• ​Adequacy of the proposed methodology, work plan, and approach
• ​Qualifications and competence of key experts proposed for the assignment
• ​Extent of participation by nationals among key experts, and associated knowledge transfer

​That last criterion is worth pausing on, because international firms routinely underweight it. World Bank guidance specifies that the participation of national experts, calculated as the ratio of key national experts' time in person-months to the total key experts' time proposed, can be given a higher weight to reflect its importance to the Borrower. In practice, this means a proposal's staffing plan is not simply a resourcing decision; it is itself a scored line item, and a firm that fields an entirely expatriate key-expert team is deliberately forfeiting points many RFPs make available for local capacity building.

​For rated technical criteria more broadly, World Bank guidance on implementing rated criteria illustrates how granular and assignment-specific the sub-weighting gets. A published example breaks a technical envelope into components such as skills and experience of key experts and proposed methodology accounting for a defined share of the total technical weighting, with the remainder allocated to other rated categories tailored to the specific risks and priorities of that assignment (for instance, logistics and phasing challenges on a difficult site, or approach to security and continuity of service). The evaluation grid is written to reward whatever the procuring entity has identified, through its own project planning, as the assignment's real risk points. This is precisely why generic, templated methodology sections that ignore the specific ToR consistently score lower than proposals that mirror the RFP's own stated priorities back at the evaluator.

​Two African Cases That Show Where the Scoring Can Be Challenged and Why

​Sintmond Group v Kenya Electricity Generating Company (2026)

​In a judgment delivered in July 2026, Kenya's Court of Appeal overturned a Public Procurement Administrative Review Board decision and ordered a fresh review of a disputed KenGen carbon credits tender (Tender No. KGN-SALE-005-2025). Sintmond Group had initially met a mandatory tender requirement, but KenGen introduced additional evaluation standards during the due diligence stage that followed technical scoring. The Court of Appeal ruled that procuring entities cannot introduce new evaluation criteria during due diligence, as due diligence exists to verify claims made against the criteria published in the RFP, not to add fresh scoring grounds after proposals have already been submitted and evaluated. For engineering and consulting firms, the case is a reminder that the evaluation criteria disclosed in the RFP is not just a drafting guide for your proposal; it is a legal ceiling on what the evaluator is permitted to score you against, and departures from it are challengeable.

​Arprim Consultants / Nyange Integrated Consultants / Peng Limited (JV): PPARB Review

​In a review before Kenya's Public Procurement Administrative Review Board, a joint venture was recommended for award as the firm with the highest combined technical and financial score, at a financial proposal sum in excess of KSh 369 million. Part of the underlying dispute concerned Regulation 30(a) of Kenya's Public Procurement and Asset Disposal Regulations, 2020, which requires evaluation committee members to conduct technical and financial evaluation in strict adherence to the compliance and evaluation criteria set out in the RFP document, no more and no less. The regulation exists precisely because evaluators are not free to weigh proposals however seems reasonable to them in the room; they are bound to the published grid, sub-criterion by sub-criterion.

​Together, these cases establish a pattern that engineering firms can use offensively as well as defensively: if a losing proposal was marked down against a criterion, standard, or threshold that does not appear in the published RFP evaluation grid, or that only appeared after submission at the due diligence or negotiation stage, that is a specific, documented ground for a request for review, not merely a general grievance about being unsuccessful.

​What This Means for How a Technical Proposal Should Actually Be Written

• ​Mirror the evaluation grid's own language and structure, section by section. If the RFP lists "adequacy of methodology," "experience of the firm," and "qualifications of key experts" as separate rated criteria, your proposal's headings and emphasis should map directly onto those categories in that order, rather than a narrative structure of your own choosing. Evaluators scoring against a rubric reward proposals that make their job easy; a proposal an evaluator has to search through to find rubric-relevant content scores worse than one that hands the answer over on the page it's expected.
• ​Treat the staffing plan as a scored technical input, not an HR afterthought. Where national expert participation is a rated criterion, as it commonly is under World Bank guidance, the ratio of local to expatriate key-expert person-months is a number an evaluator will calculate and score. Firms bidding into Africa without a credible plan for national expert involvement are leaving points on the table that a competitor with a genuine local staffing structure will pick up.
• ​Read the risk-specific rated criteria, not just the standard categories, as the real signal of what will win. Beyond the standard experience, methodology, and staffing buckets, RFPs increasingly attach assignment-specific rated criteria tied to whatever the procuring entity's own project planning has flagged as the key delivery risk (such as site access, security, environmental sensitivity, or phasing around an operating facility). A generic methodology section that doesn't visibly respond to that specific, named risk is competing against proposals that do.
• ​Keep a documented record of exactly what criteria the RFP disclosed. Given the Sintmond precedent, firms should retain the published evaluation grid and any subsequent addenda as part of their bid file. If a debrief or award notification suggests scoring against something outside that disclosed grid, such as a standard introduced only at due diligence, negotiation, or contract award stage, that is now Court of Appeal-tested ground for a request for review in Kenya. This reflects a principle that criteria must be fixed and disclosed in advance, running through World Bank and AfDB procurement regulations as well.

​The Underlying Principle

​Technical proposal evaluation on African donor-financed and public-sector assignments is not a subjective judgment of overall quality. It is a scored exercise against a rubric fixed and disclosed before submission, sequenced strictly ahead of price, and increasingly subject to judicial and administrative review when procuring entities depart from it. Engineering and consulting firms that write proposals as if they are addressing that rubric directly, treating staffing composition, methodology structure, and assignment-specific risk response as scored inputs rather than narrative flourishes, are playing the game evaluators are actually required to score. Firms that write a well-crafted general capability statement instead are, in effect, submitting an answer to a question the evaluator was never permitted to ask.

​Sources: World Bank, Guidelines: Selection and Employment of Consultants by World Bank Borrowers; World Bank, "Rated Criteria and Evaluation Examples" and "How to Implement Rated Criteria" briefing (thedocs.worldbank.org); World Bank, Sample Form of Evaluation Report; Arched, "QCBS vs QBS: Key Differences in Government Tenders Explained"; Kenya Public Procurement Administrative Review Board decision on tender involving M/s Arprim Consultants, M/s Nyange Integrated Consultants Limited and M/s Peng Limited (JV) (ppra.go.ke); AllAfrica, "Kenya: Court Orders Fresh Review of Kengen Carbon Credits Tender" (July 2026), reporting on the Court of Appeal judgment in the Sintmond Group Limited matter concerning Tender No. KGN-SALE-005-2025; Kenya Public Procurement and Asset Disposal Regulations, 2020, Regulation 30(a).

​Master the Art of Winning African Tenders

​Gain an edge in your technical proposals with actionable procurement intelligence, evaluation breakdowns, and regulatory updates across Africa.

Subscribe to

​ aectenderlink.com. to get high-value tender insights delivered straight to your inbox!