Define the quality standard before work begins. Review the delivered concepts. Pay only for what you accept. Rework is on us. This is how we align incentives.
Acceptance criteria are defined at kickoff, before any work begins. The client and Model T team agree on specific, measurable standards that each product concept must meet. These typically include: the concept addresses a documented pain point (not a hypothetical one), the system architecture is technically feasible and reviewed by a domain expert, the BOM is based on real component pricing from authorized distributors, and the competitive positioning includes at least 3 named alternatives with specific differentiation points.
The acceptance review is a structured process, not a subjective judgment call. The client receives the delivered concepts and has 5 business days to review them against the agreed criteria. Each criterion is evaluated as met or not met. If all criteria are met, the concept is accepted and payment is triggered. If any criterion is not met, the specific gap is documented and the concept enters rework.
This definition eliminates the ambiguity that plagues traditional consulting engagements. There is no debate about whether the work was "good enough." The criteria are binary: either the BOM uses real pricing or it does not. Either the architecture was reviewed by a domain expert or it was not. Either 3 named competitors are analyzed or they are not. The client knows exactly what they are paying for before work begins.
The engagement follows a clear sequence. Step one: the client provides 5-7 target accounts and agrees on acceptance criteria during a kickoff meeting (typically 60-90 minutes). Step two: the Model T pipeline runs for 2-3 weeks, producing 2-3 product concepts per qualified account. Step three: concepts are delivered to the client for review. Step four: the client evaluates each concept against the agreed criteria within 5 business days.
For concepts that meet all criteria, the client confirms acceptance and payment is processed within standard terms (typically net 30). For concepts that do not meet one or more criteria, the client provides specific feedback documenting which criteria were not met. The Model T team then reworks the concept at no additional cost, addressing each documented gap. The reworked concept goes through the same acceptance review.
In practice, approximately 60% of delivered concepts are accepted on the first review. Another 25-30% are accepted after one round of rework (typically addressing BOM pricing accuracy or competitive positioning depth). Fewer than 10% require a second rework round, and these are almost always cases where the target company's situation changed during the engagement period.
The pay-on-acceptance model only works if the provider maintains a high acceptance rate. If acceptance drops below 50%, the rework cost exceeds the engagement margin and the model becomes unprofitable. Model T maintains a 60%+ first-round acceptance rate through the quality gates built into the 18-step pipeline.
The quality gates function as progressive filters. Of the 5-7 target accounts provided, typically 1-2 are disqualified at the profiling stage. Of the 5-7 insights generated per qualified account, only 2 survive scoring. Of the concepts created, each must pass product validation, technical validation, and business validation before delivery. By the time a concept reaches the client, it has already survived 5 internal quality checks.
This means the concepts that reach the acceptance review are the survivors of a rigorous internal process. The 60% first-round acceptance rate is not a gamble; it is the result of killing bad concepts early and investing concept creation resources only in opportunities with genuine potential. The pipeline is designed so that the provider bears the cost of quality control internally, rather than passing low-quality work to the client and hoping for acceptance.
Retainer models charge a fixed monthly fee regardless of output or quality. The client bears all the risk: if the deliverables are poor, the retainer is still due. The provider is incentivized to maintain the engagement duration, not to deliver results quickly. Retainer models work well for ongoing support relationships but create misaligned incentives for project-based deliverables.
Milestone models split the engagement into phases with payments tied to deliverable completion. This is better than retainers because payment requires demonstrated progress. But "completion" and "quality" are different things. A milestone can be technically complete (the document was delivered) without being useful (the architecture is not feasible). Milestone models reduce risk but do not eliminate it because quality assessment happens after payment, not before.
Pay-on-acceptance inverts the risk entirely. The provider bears the risk of producing substandard work, because substandard work is reworked at the provider's expense. The client bears no risk beyond the time invested in the kickoff meeting and concept review. This model works only when the provider has sufficient confidence in their methodology to accept the downside risk, which is why it requires a validated, repeatable pipeline rather than ad hoc consulting.
For procurement teams evaluating Model T, the pay-on-acceptance model simplifies vendor assessment. The standard concerns around scope creep, change orders, and quality disputes are eliminated by design. The scope is fixed (2-3 concepts per qualified account). There are no change orders (rework is the provider's cost). Quality disputes are resolved by reviewing deliverables against pre-agreed binary criteria.
Budget approval is simpler because the risk is bounded. The maximum exposure is the engagement price (from €15,000) multiplied by the number of accepted concepts. If no concepts are accepted, the client's only cost is the time invested in kickoff and review. This makes the approval process comparable to purchasing a defined product rather than commissioning open-ended services.
The payment structure also aligns with MDF (Market Development Fund) budgets used by semiconductor vendors and distributors. MDF-funded engagements require demonstrable ROI for channel program managers. Pay-on-acceptance provides a clean metric: the company paid for X accepted concepts targeting Y named accounts, resulting in Z meetings booked. This is significantly easier to report than the ROI of a retainer engagement or a generic marketing campaign.
If all concepts fail to meet the agreed acceptance criteria after rework, the client does not pay. This has not occurred in practice because the quality gates in the pipeline catch fundamental issues before concepts reach the client. However, the scenario is contractually covered: non-acceptance means non-payment. The provider absorbs the production cost.
No. Acceptance criteria are locked at kickoff to prevent scope creep. If the client's needs change during the 2-3 week engagement window, the new requirements are addressed in a subsequent engagement, not by retroactively changing the current acceptance standards. This protects both parties: the client gets what they agreed to, and the provider is not asked to hit a moving target.
Two rework rounds are included at no additional cost. In practice, fewer than 10% of concepts require a second round. If a concept still does not meet acceptance criteria after two rework rounds, the engagement terms include an option for the client to either reject the concept (no payment) or accept it at a reduced rate (negotiated on a case-by-case basis).
For clients who need continuous pipeline support (e.g., running concepts for new target accounts every month), Model T offers a sprint-based model where each sprint is priced independently and follows the same pay-on-acceptance terms. This provides the continuity of a retainer with the risk protection of pay-on-acceptance.