Most platform decisions are won by the loudest voice in the room
Walk into any enterprise mid-shortlist and you'll see the same scene — three VPs, three vendors, three slide decks, and a decision that drifts another month. The fix isn't more expertise. It's a defensible scoring model anyone can read in twenty minutes.
Walk into any enterprise mid-shortlist and you’ll see the same scene. Three VPs, three preferred vendors, three slide decks. The CFO is half-listening. Procurement is waiting for somebody to mention TCO. The CTO already has a favourite but won’t say so until the room shifts. Forty minutes in, the loudest voice — usually the one whose team is already running a pilot — gets the room to “lean” their way. Someone says “let’s regroup next week.” The decision drifts another month.
This is not a competence problem. The people in that room are experienced. They have done procurements before. The shortlists are reasonable. The vendors are credible.
It is a process problem. Nobody in the room can point to a number, an objective score, or a defensible reason why one platform is a better fit than another. They have impressions. Impressions are loud, and the loudest impression usually wins.
The cost of impression-based decisions
Gartner pegs the average enterprise technology decision cycle at 6–12 months. Sixty percent of buyers later regret their largest recent purchase. McKinsey/Oxford research across more than five thousand IT projects clocks the average cost overrun at 45%.
You can blame vendors for selling badly. You can blame teams for picking wrong. But after watching this pattern across logistics, banking, insurance, healthcare and the public sector — across MuleSoft and Salesforce and SAP and dozens of niche platforms — the actual cause is the same every time: the comparison was never structured in the first place.
What “structured” actually means
A defensible technology decision has three properties:
- Every option is scored against the same criteria. Not “we like Option B’s UI.” A weighted score on usability, with the weight agreed up front, applied uniformly to every shortlisted vendor.
- Every weight is visible. If integration ease is worth 25% of the decision and total cost of ownership is worth 15%, that is a deliberate choice. Write it down. Show it to the board. Defend it.
- Every score is reproducible. Anyone re-running the analysis with the same inputs should land at the same recommendation. Otherwise it is an opinion, not an evaluation.
None of this is hard mathematics. A weighted-criteria scoring matrix is undergraduate decision theory. What is hard is the discipline to actually do it — to capture the criteria before you have fallen in love with a vendor, to commit to the weights before you see the numbers, to publish the model so others can poke at it.
The reason most large consultancies don’t work this way is that the math, once visible, leaves them less room. A scoring matrix that says “Platform X wins because integration ease was 25% weighted and X scored 8.4 vs. Y’s 7.1” is harder to charge for than a fifty-slide narrative deck. The math is the deliverable. There is nowhere to hide.
What this looks like in practice
I run platform decisions on a five-phase rhythm:
- Rapid Assessment — frame the question, agree the criteria, lock the weights. 3–5 days.
- Deep Analysis — score every option against those criteria. Comparison matrices, TCO, risk profile. 2–3 weeks.
- POC / Validation — hands-on test of the top candidates. Flexible duration.
- Decision Workshop — walk stakeholders through the math, surface objections, capture the call. 1–2 days.
- Implementation Roadmap — translate the decision into milestones, dependencies, risks. 3–5 days.
End to end: five to eight weeks. Every phase is AI-powered now — vendor research, scoring, scheduling, transcription, reporting — which is what compresses what used to be six months into eight weeks without cutting corners.
The output is not a slide deck. It is a decision document that any board member can read in twenty minutes and ask sharp questions of: Why was security weighted at 18%? Why did Vendor C lose a point on TCO? Every answer is in the matrix.
The board test
If you can’t sit in front of your board and walk them through the math behind a major platform purchase — every criterion, every weight, every score — then the decision was not made yet. Someone in the room just decided loudest.
This is the work I do. If your next platform decision needs to survive that test, here is how the Rapid Decision Framework runs end-to-end.