Here is the uncomfortable truth most enterprise L&D functions won't say out loud: the RFP process you rely on to select training vendors is not designed to find the vendor who will produce the best learning outcomes. It is designed to produce a defensible paper trail. Those are not the same objective, and the gap between them is where most enterprise training budgets quietly go to die.
Walk into any procurement-led vendor selection for a capability programme and you'll find the same scoring rubric wearing different logos: proposal clarity, case studies, references, pricing tiers, brand recognition, response time to the RFP itself. Somewhere in a footnote there might be a line for "instructional design methodology," but it's scored by people who have never designed instruction and can't meaningfully differentiate a genuinely adaptive curriculum from twelve slides claiming to be one. The rubric rewards the vendor who writes the best proposal, not the vendor who runs the best programme. Those two skills overlap far less than anyone in the room wants to admit.
Consider what an RFP actually asks a vendor to demonstrate. It asks for a polished narrative, a list of recognisable clients, a modular curriculum outline, and a price that fits within a pre-approved band. None of that is illegitimate to ask for. But none of it correlates with what actually determines whether a cohort of engineers or managers changes behaviour six weeks after the workshop ends. That outcome depends on facilitator quality on the specific day of delivery, on whether the content was adapted to your actual tooling and constraints rather than a generic template, on whether assessment was built into the design rather than bolted on as a satisfaction survey, and on whether the vendor has any mechanism at all for finding out what happened after the invoice was paid. Almost none of that is visible in a proposal document, and even less of it is scored.
The result is a selection process that is rigorous in form and almost accidental in substance. Buyers walk away confident they ran a fair, thorough process — multiple bidders, weighted scorecards, stakeholder sign-off — while having selected largely on the basis of who presented best in a ninety-minute pitch meeting. Presentation skill is a real and valuable skill. It is also almost entirely uncorrelated with delivery skill, and enterprise buyers routinely mistake one for the other because the pitch is the only moment of the entire relationship they get to observe directly.
The second force quietly at work in every RFP is familiarity bias, and it is far more powerful than any scoring matrix. Incumbent vendors win renewal RFPs at rates that have nothing to do with performance, because the switching cost of onboarding a new provider — new contracts, new points of contact, new risk assessment — is itself treated as a cost the scorecard should minimise. Capability directors will tell you, off the record, that they've kept a mediocre vendor for three cycles running because replacing them meant reopening a procurement fight nobody had appetite for. That's not a failure of judgement. It's a rational response to an incentive structure that punishes disruption and never quite gets around to auditing outcomes.
Brand recognition plays the same role for new vendor selection. A name the steering committee has heard of clears a psychological bar that a smaller, sharper, more current provider simply cannot clear on reputation alone — regardless of what their last three cohorts actually achieved. This is particularly costly in fast-moving technical domains, where the vendor with the most enterprise logos on their homepage is frequently the vendor whose curriculum was last substantially updated two product cycles ago. A generative AI training programme built and delivered eighteen months ago is, in practical terms, a different subject entirely from one refreshed this quarter — and yet the RFP rubric has no field for "how recently was this content actually rebuilt," because that's not the kind of thing procurement templates are built to ask.
The assumption that a rigorous procurement process produces a rigorous vendor is one of the more expensive fictions in enterprise L&D — rigour in selection and rigour in delivery are governed by entirely different disciplines, and optimising for the first tells you almost nothing about the second.
Price survives every other flaw in the process because it is the one number that appears identical to every stakeholder in the room. Where instructional quality is subjective, contested, and hard to defend in a post-decision audit, price is a clean, comparable figure that closes arguments quickly. This is precisely why RFPs gravitate toward it as the tie-breaker of last resort — and precisely why so many enterprise training decisions ultimately reduce to per-seat cost, dressed up in language about "value" and "total cost of ownership" that rarely survives contact with the actual spreadsheet. A vendor who has built genuine adaptive capability into their delivery — modular pacing, live troubleshooting, post-programme reinforcement — costs more to run than one who hasn't, for entirely legitimate reasons. An RFP scoring model built around normalized pricing bands cannot see that difference; it only sees two numbers, and the lower one wins unless something else in the scorecard is weighted heavily enough to overrule it, which it almost never is.
The vendors who lose most consistently under this system are not the weak ones. They are the ones whose value is structurally difficult to present in proposal format: the smaller specialist shop whose facilitators are also practitioners, the provider who insists on a discovery call before quoting because they actually customise the content, the team whose real differentiator is what happens in week three of a cohort rather than what appears on a title slide. These providers frequently decline to bid at all once they understand how the scoring works, because they've learned that RFPs select for polish, not substance, and their competitive advantage doesn't live in either of those places. What's left in the pool, disproportionately, are vendors optimised for winning RFPs rather than for running programmes — a selection pressure that compounds with every renewal cycle.
None of this means procurement discipline should be abandoned. It means the discipline is currently pointed at the wrong target. A buyer serious about effectiveness would weight post-programme measurement design as heavily as proposal narrative, would ask to sit in on a live session before signing rather than after, and would treat facilitator continuity as a contractual term rather than a nice-to-have. Most don't, not from negligence, but because the RFP template was inherited from procurement functions built for software licensing and office furniture, where presentation and price genuinely are the right proxies for quality. Training isn't furniture. It doesn't behave the same way, and scoring it as though it does is how capable organisations keep signing multi-year contracts with vendors who were simply better at answering the questionnaire. For those willing to look past the template, the more useful signal is often the least glamorous one: whether a provider is running live, current cohorts you can actually observe — a quick look at any credible upcoming training batches page tells you more about operational reality than most twenty-page proposals ever will.
We run live, instructor-led programmes on Orbit. Browse our upcoming batches.