Focused decision pilot · operating pack

Turn the pilot proposition into an executable club workflow.

This is the operational layer behind the sales page: what we need in week zero, who owns what, how each decision cycle runs and what evidence defines success.

Week 0

Onboarding before the first fixture decision.

01

Confirm scope

Choose the agreed 4–6 fixture or campaign window, primary commercial outcomes and named decision owners.

02

Connect minimum data

Fixture keys, ticket transactions, scans, campaign identifiers and a pseudonymous supporter key.

03

Set measurement rules

Agree attribution boundaries, privacy rules, baseline windows and what counts as a valid result.

04

Run the first dry cycle

Generate one opportunity, review assumptions, test the action plan and confirm the post-match scorecard.

Minimum data contract

Start narrow. Make the gaps explicit.

Build the integration before receiving club data

Fixture keysready

Shared across calendar, campaigns and post-match analysis.

Ticket-level grainready

One row per ticket prevents quantity and scan ambiguity.

Campaign attributiondemo

UTM and content keys work with synthetic records.

Scan and no-show logicdemo

Calculated in demo mode; requires gate data for actuals.

Pseudonymous supporter keyrequires-access

The club must generate the stable hash.

30/60/90-day retentionrequires-access

Requires records across multiple fixtures.

Data trust

Use supporter-level data to calculate. Store only what the decision needs.

The pilot workflow keeps authorised ticket-grain data outside this repository. Matching happens locally, and the product stores only fixture-level and repeat-cohort aggregates.

01 · SOURCEClub-controlled export

Raw CRM/ticketing files remain outside the product repository and are never used as demo content.

02 · MATCHPseudonymous local processing

A stable supporter hash can be used to calculate repeat behaviour without names, emails or phone numbers.

03 · STOREAggregate evidence only

Fixture summaries and cross-fixture cohort counts are the only club evidence written into the live product state.

Current connectionNo club CRM/ticketing aggregate connected

No authorised club CRM/ticketing aggregate is connected. Raw or pseudonymous supporter-level records are never stored in the repository.

Club roles

The product needs named humans around the decision.

Club sponsor

Owns the pilot outcome, removes blockers and approves material decisions.

CRM / marketing lead

Owns audiences, campaign execution and channel coordination.

Ticketing / data lead

Provides fixture, transaction, scan and product data.

Measurement lead

Defines attribution, scorecards and post-match learning.

Measurement design

Decide the strength of evidence before the campaign launches.

The pilot can report attribution with campaign identifiers, but incremental impact needs a counterfactual. Each decision cycle should choose the strongest practical design in advance.

StrongestRandomized holdout

Keep an eligible slice unexposed so incremental lift can be estimated against the same audience at the same fixture.

StrongMatched control or stepped rollout

Use a comparable audience, territory or timing window when randomization is not operationally acceptable.

DirectionalPre-agreed baseline

Compare against a defined historical or pre-campaign benchmark, while keeping the causal claim explicitly limited.

Causal claim ruleNo holdout, control or agreed counterfactual = no incremental-lift claim.

Observed purchases and campaign attribution remain useful decision evidence, but they stay descriptive.

Per-fixture cadence

One repeatable decision cycle.

T-10 → T-7Discover

Opportunity, audience hypothesis and evidence review.

T-7 → T-3Decide

Approve action, budget, owner, measurement and blockers.

T-3 → MatchdayAct

Execute with campaign and fixture identifiers intact.

T+1 → T+7Learn

Close result, attribution, attendance and next-decision implications.

Pilot success

Success is not “the dashboard looked useful”.

01Every pilot decision cycle has one explicit opportunity and decision owner.
02Every recommended action has a measurement plan before launch.
03Ticketing and CRM evidence can be matched locally with a pseudonymous supporter key and stored as aggregate evidence.
04Observed learning changes at least one subsequent decision.
05The club can explain what is measured, assumed and still missing.
Ready for a club conversation

The next step is to replace demo assumptions with one club’s real fixture and fan data.

Pilot Operating Pack | AVELA