Source material — Hydra Holdings, Payments group¶

Persona-redacted composite of the kind of material a consultant arrives with on day one. Deliberately messy: it mixes dates, authors, registers and confidence levels, and two of its claims contradict each other. It is the input for rework-recipes.md.

Nothing in here is in the narrative format. That is the point.

1 · Facilitator's typed-up workshop wall (2026-07-16, Delivery lead)¶

DOTS (52 cast, 9 participants)
  Skills / knowledge silos ............ 16
  Can't see delivery health ........... 11
  On-call knowledge in people's heads ... 9
  Internal platform friction ........... 8
  No time to improve anything .......... 5
  Test data ............................ 3

SME MATURITY (0-3, self-rated, whole group)
  tech capability ......... 1.7
  people & purpose ........ 1.9
  delivery metrics ........ 1.2
  incident management ..... 1.5
  career growth ........... 2.0

$ SPEND STACK (facilitator ranked all 41, parked undiscussed at #38 = $1)
  #1  tech capability        $340k
  #2  platform              $180k
  #3  incident mgmt          $95k
  ...
  #38 (30 needs) leftover     $1

Verbatim quotes captured on stickies:

"If Engineer C is on leave the pipeline work stops."
"We rebuild the delivery numbers by hand for every steering meeting."
"Nobody wants to be on call for quota-keeper."
"We use the platform where we have to and route around it where we can."

2 · Existing maturity self-assessment (2024-11, Engineering manager)¶

Overall we rate ourselves Level 3 — Defined against the internal maturity framework. Areas of strength: version control discipline, code review culture, CI coverage. Areas for development: observability tooling, formal incident process. We expect to reach Level 4 within two quarters given the platform investment now underway.

No evidence cited. No measurements. Two years old. The platform investment it refers to was cancelled in 2025.

3 · Retro notes, last four sprints (2026-05 → 2026-07, Tech lead)¶

S41  "review queue again" · "waiting on identity-broker team" · release slipped
S42  "review queue" · flaky pricing-engine tests · release slipped
S43  good sprint, one hotfix
S44  "review queue" · "3 approvers for a one-line change" · release slipped

Standing item carried forward each sprint: "Reduce review turnaround." Never actioned; no owner.

4 · SDO readout, screenshotted from the platform dashboard (2026-08-01)¶

Lead time for changes ....... 19h        (median, last 90d)
  of which review wait ...... 14h
Deployment frequency ........ 0.4 / day
Change failure rate ......... 0.12
MTTR ........................ 45
Sev1/2 incidents ............ 6 / month
Teams ....................... 3

Units are not labelled on the dashboard. 0.12 and 45 are both ambiguous.

5 · Prior discovery deck, slide notes (2026-06, external vendor)¶

Slide 7 — Key Findings - Engineering velocity is below industry benchmark for elite performers - Toolchain modernisation is the highest-leverage investment - Recommend a 6-month DevOps transformation programme - Culture is a blocker to change

Vendor had a tooling partnership. No measurements shown for any of the four claims.

6 · Existing markdown findings doc (2026-08-05, Head of Enablement, internal)¶

## Payments delivery — where we think we are

Our biggest problem is clearly tooling. Best practice says we should be
deploying daily and we are nowhere near that. The team has raised the review
queue repeatedly but honestly the bigger issue is that we have too few
engineers — we need 4 more heads to get anywhere.

Incident response is fine. We have a rota.

Written by the person who commissioned the assessment. Contradicts §3 on cause, and §1 on incident management.