Pillar 3 — Think

Industry Benchmarks.

"Is 120 days to a signed pilot good or bad?" There is no public benchmark for that question. Industry reports are anecdotal, analyst firms sell you three-figure PDFs with out-of-date numbers, and your competitors are not going to email you their pilot velocity. CORTEX Benchmarks is the answer — a continuously-updated peer comparison across eight partnership metrics, built from the anonymised aggregate of every TECH member, with privacy-preserving cohort minimums baked in.

Why it matters

Most partnership teams operate without reliable external reference points. The result is over-indexing on internal norms — "we're getting faster" when the industry is getting faster twice as quickly. Benchmarks gives you a factual basis for the quarterly business review: "we're in the top quartile for time-to-pilot and lagging on compliance cost per partnership."

The eight metrics

Time to Due Diligence

First contact → completed due diligence. Measures how fast your team moves from "interesting" to "approved to proceed."

Time to Pilot

First contact → pilot launch. The composite metric most banking leaders care about; anything over 180 days flags a process problem.

Time to Revenue

First contact → first dollar attributed to the partnership. Includes pilot and initial contract phases.

Success Rate

Pilots that convert to commercial contracts, as a percentage of pilots launched.

Average Deal Size

ACV of signed partnerships. Segmented by deal structure (rev share vs. flat fee).

Integration Time

Technical integration duration for API-based partnerships. Measured from kickoff to go-live in production.

Compliance Cost

Dollar cost of getting each partnership through legal and compliance review. Includes internal time + external counsel.

Revenue per Partnership

Annualised, computed after one full year. Separates BaaS from referral from co-brand.

How your position is computed

For each metric, TECH computes three percentile points over the relevant peer cohort: p25, p50 (median), and p75. Your own value is plotted against the distribution with a plain-English label:

Top quartile — your value is above p75 (on metrics where higher = better) or below p25 (where lower = better, like Time to Pilot).
Above median — you outperform the peer median.
At peer median — your value is between p25 and p75 and close to p50.
Below median — you lag the peer median.
Bottom quartile — flagged for attention; the biggest opportunity for improvement.

Peer cohort selection

"Peer" in Benchmarks is not a single pool — it is computed per request based on your organization's type and size bucket. The cohort rules:

Same organization type (bank vs. fintech vs. investor).
Similar stage bucket — for fintechs: seed/Series A, Series B, Series C+, public. For banks: regional, global, investment.
Similar employee count bucket (10–50 / 50–500 / 500–5,000 / 5,000+).
Optionally filtered by sector when you request a sector-specific view.

The Benchmarks UI always shows the cohort size used for each metric — so you can see "compared to 23 peers" and judge for yourself whether the sample is meaningful.

Privacy

Minimum cohort size: 5. TECH will not publish a benchmark for any metric until at least five organizations in the relevant cohort have data. For smaller cohorts, the UI displays a "contribute to unlock" placeholder instead of a misleading number. No individual organization's value is ever exposed, even in aggregate displays — the p25/p50/p75 are pure statistics, not joinable to any specific member.

Data sources

Every benchmark value is derived from platform-observable data:

Timing metrics (time to pilot, time to revenue) — computed from deal stage transitions in Pipeline, which are timestamped server-side and cannot be back-dated.
Outcome metrics (success rate) — from deal outcome transitions (won/lost).
Financial metrics (deal size, revenue per partnership) — from user-entered values in deal records, cross-validated against ROI Projection entries where available.
Cost metrics (compliance cost, integration time) — currently user-reported at deal milestone transitions with validation prompts.

Reporting cadence

Benchmarks are recomputed weekly (Monday 02:00 UTC) using data through the prior Sunday 23:59 UTC. Longer-horizon metrics (revenue per partnership, measured annualised) are computed monthly on the first of each month.

Industry insights

Alongside raw benchmarks, the Benchmarks tab surfaces 3–5 written insights per quarter. These are observations synthesised from the distribution shifts — e.g., "time to pilot for mid-market fintechs has improved 18% YoY across the TECH ecosystem, driven primarily by faster KYB turnarounds" — and are intended as context for your internal discussions, not as predictive claims.

Export for committee packs

Benchmarks exports as PDF in a format designed to drop into a board or partnership-committee deck: one page per metric showing your value against the peer distribution with executive-ready framing language. Professional and Enterprise plans include unlimited exports.

API surface

GET /cortex/benchmarks/

Your metrics vs peer medians + percentile rankings + cohort size per metric

GET /cortex/industry-benchmarks/

Raw industry distributions (p25/p50/p75) by metric, period, and cohort filter

Know where you stand

Open the Benchmarks tab in CORTEX.

The quickest read: scan the percentile labels down the page. Every bottom-quartile label is a concrete opportunity for your next quarterly planning cycle.

Open CORTEX →