What are the 15 readiness questions?

1. Can you name the decision, owner and deadline?

• 0 points: We are exploring products without a defined buying decision.
• 1 point: We have a possible supplier or renewal in mind, but ownership or timing is unclear.
• 2 points: We have a written decision, named football lead, budget owner and deadline.
• 3 points: Those people have agreed the scope, approval route, notice dates and what happens if no supplier passes.

Evidence to attach: a one-page decision brief. Include “retain the current arrangement” as an available outcome.

2. Do you know what the existing scouting stack already answers?

• 0 points: We cannot reliably list our subscriptions and entitlements.
• 1 point: We have product names and fees, but not a task-level capability map.
• 2 points: We have documented seats, modules, competitions, seasons, exports and the tools used for our priority tasks.
• 3 points: We have completed the proposed trial tasks using the incumbent stack and recorded answers, gaps, effort and dependencies.

Evidence to attach: a Scouting Subscription Overlap Audit with an existing-stack baseline.

Why refresh it? Hudl’s 18 June 2026 announcement describes broadcast tracking integrated with Wyscout and Statsbomb event data in Insight. Treat that as a reason to check current capabilities and entitlements—not proof that your club already licenses the integration. (hudl.com)

3. Can you specify what additional value you are testing?

• 0 points: Our aim is simply to obtain more data or use AI.
• 1 point: We have a broad pain, such as slow shortlisting or incomplete coverage.
• 2 points: We have one to three recruitment questions and a proposed improvement over the baseline.
• 3 points: Football and budget owners have agreed observable acceptance criteria, including what would constitute no meaningful improvement.

Evidence to attach: a testable statement: “For this role and market, we need this evidence that the current stack cannot provide adequately.”

Distinguish new evidence, better interpretation, faster retrieval and additional assessment capacity. Do not label all four as newly sourced intelligence.

4. Have you defined the exact competition-and-season requirements?

• 0 points: We rely on a supplier saying it has global or women’s coverage.
• 1 point: We have named leagues, but not seasons, historical depth or required evidence.
• 2 points: We have a competition-season matrix specifying required video, fields, history and availability timing.
• 3 points: Each priority row has an acceptance threshold, proposed sample and owner; women’s recruitment is tested directly wherever it is in scope.

Evidence to attach: an MLS coverage matrix or NWSL Recruitment Coverage Gap Map.

Hudl’s Wyscout FAQ explicitly says coverage and historical depth vary by competition and package. A catalogue entry should therefore trigger verification, not receive an automatic pass. (hudl.com)

5. Can you test whether the required evidence is usable, not merely listed?

• 0 points: A competition listing or highlight reel is sufficient for us.
• 1 point: We intend to open a few matches, without a documented completeness test.
• 2 points: We have a plan to check required footage, fields, player history, timestamps and delivery delays against an expected-match inventory.
• 3 points: The test distinguishes unavailable, late, incomplete and unusable material, and specifies which failures block the intended recruitment task.

Evidence to attach: a coverage-verification sheet with a dated denominator. Count completed matches within the agreed scope; record postponed and future fixtures separately.

For tracking-derived products, specify the metric and conditions being evaluated. FIFA’s EPTS programme assesses tracking-system accuracy, while a 2026 study of broadcast-derived auto-events found results sensitive to feed type and underlying tracking quality; that study examined one match, not every recruitment setting. (ipt.fifa.com)

6. Can suppliers be tested on the same recruitment work?

• 0 points: Each supplier will choose its own demonstration examples.
• 1 point: We will share a few players, but questions and deliverables differ.
• 2 points: We have a common, authorised shortlist and identical tasks, deadlines and output requirements.
• 3 points: We have frozen the baseline, included difficult cases, agreed how assistance is recorded and defined how failed or untestable tasks will be reported.

Evidence to attach: a Same-Shortlist Supplier Trial Kit.

A practical starting design is five to ten player cases, including known examples, incomplete records and unfamiliar markets. This is a proposed diagnostic sample—not a statistically representative coverage audit. For discovery tools, add an identical role-and-market search task so that the test does not exclude their discovery function.

7. Can the output be joined to your club’s player records?

• 0 points: We assume a matching name identifies the correct player.
• 1 point: We expect analysts to resolve discrepancies manually, without logging them.
• 2 points: We have defined identifier mapping, duplicate handling, field definitions and an exception process—or an equivalent controlled manual hand-off.
• 3 points: An authorised sample has passed through the proposed workflow, with mismatches, unresolved records and reconciliation time recorded.

Evidence to attach: a Player Data Joinability Check.

For API purchases, request the actual schema, versioning and operational documentation. Wyscout’s API documentation, for example, specifies identifiers, rate limits, versions and coordinate conventions; “API available” is not a complete integration specification. (apidocs.wyscout.com)

8. Can you challenge role-fit and comparison assumptions?

• 0 points: We accept player rankings or similarity scores at face value.
• 1 point: We know the headline metrics but not their comparison population or limitations.
• 2 points: We have specified questions about role, minutes, competition context, missingness and any league-translation assumptions.
• 3 points: We have relevant challenge cases, requested validation evidence and an agreed rule for when a model or assessment must abstain rather than rank confidently.

Evidence to attach: a model-and-assumptions test sheet. For human scouting, use the same principle to examine the assessor’s observation base and comparison standard.

For women’s recruitment, ask explicitly about women-specific benchmarks. SkillCorner’s 21 October 2026 webinar page describes dedicated women’s physical thresholds; that establishes a supplier-described approach, not independent proof of suitability for your shortlist. (skillcorner.com)

9. Can you trace important claims back to evidence?

• 0 points: A convincing report or AI answer is enough.
• 1 point: Sources are mentioned generally, without claim-level references.
• 2 points: The trial requires source records, dates, match references or clip timestamps, with inference separated from observation.
• 3 points: A reviewer and checking protocol are assigned, including tests for unsupported claims, conflicting records and missing evidence.

Evidence to attach: an AI Scouting Report Evidence Check—or the same evidence ledger for human reports.

NIST identifies confidently presented false content as a generative-AI risk and recommends verifying sources and citations in outputs. A citation should be checked for whether it actually supports the claim. (tsapps.nist.gov)

10. Are the intended trial uses authorised?

• 0 points: We assume access or a download button permits the intended use.
• 1 point: We have discussed rights, but lack the applicable documents or written clarification.
• 2 points: Trial access, processing and sharing are authorised; unresolved production uses are recorded separately.
• 3 points: We have a reviewed workflow-to-clause matrix covering relevant entities, users, exports, storage, modelling, external AI processing, sharing and termination.

Evidence to attach: a Recruitment Data Licence Stress Test and named legal-review owner.

Do not infer multi-club rights from common ownership. As a concrete historical example, Hudl’s published 27 September 2024 master agreement limits affiliate access unless the order form or authorised-user provisions permit it. Your applicable agreement and amendments—not that public example—must determine your rights. (static.hudl.com)

11. Can you run the evaluation without unnecessary disclosure?

• 0 points: We would send unrestricted shortlists, contracts or player records.
• 1 point: We would redact some material, but recipients, retention or onward processing are unclear.
• 2 points: A responsible reviewer has approved the minimum fields, recipients, access controls, retention and deletion process for the trial.
• 3 points: The controls have been checked, including relevant subprocessors, confidential-output handling and agreed restrictions on reuse or model training.

Evidence to attach: a trial data-handling sheet. Prefer pseudonymous shortlist references where identity is unnecessary; exclude medical, biometric and unnecessary minors’ information from this procurement exercise.

The FTC’s business guidance recommends limiting unnecessary information, controlling access and checking service providers’ security practices rather than relying only on assurances. These are useful procurement principles, not a determination of which privacy laws apply to your club. (ftc.gov)

12. Can you compare the full cost of the alternatives?

• 0 points: We only know a headline subscription price.
• 1 point: We have rough fees, but scopes or contract periods are not comparable.
• 2 points: We have aligned scope assumptions and identified implementation, support, training, overlap and exit costs.
• 3 points: We have a documented first-year and renewal-year comparison, scoped commercial evidence and a budget owner’s review of material uncertainties.

Evidence to attach: a Scouting Renewal Cost Calculator using your own quotes and assumptions.

Keep three lines separate: cash expenditure, internal staff capacity and unquantified decision benefits. In this assessment, staff hours released are not counted as cash savings unless a corresponding expenditure reduction is evidenced. Do not insert an assumed value for a future successful signing.

13. Will the people who use the intelligence actually test it?

• 0 points: Nobody has time assigned to the evaluation.
• 1 point: A champion will watch demonstrations, but ordinary users are not involved.
• 2 points: Relevant scouts, analysts and decision-makers have agreed tasks and protected review time.
• 3 points: The schedule includes onboarding, independent task completion, usability feedback, support escalation and an adoption owner.

Evidence to attach: a named evaluator rota and workflow test.

For context—not as evidence about every club—Nashville SC’s public Data Analyst role combines recruitment questions, reproducible workflows and competing short-turnaround responsibilities. Test the product within the working conditions your evaluators actually face. (teamworkonline.com)

14. Have you agreed how trial results will be measured?

• 0 points: We will decide whether the product feels impressive.
• 1 point: We have preferences, but no common measurement rules.
• 2 points: We have defined incremental findings, answer time, coverage exceptions, identity issues and reviewer effort for the relevant tasks.
• 3 points: Definitions, denominators, hard gates and any preference weights are agreed before results arrive; failed and untested tasks remain visible.

Evidence to attach: a scouting software evaluation scorecard with an exception register.

Measure additional supported answers and usable workflow improvements—not the number of dashboard screens, generated sentences or recommendations. Record supplier assistance and review time so that a heavily supported demonstration is not confused with independent everyday use.

15. Is there a decision process after the trial?

• 0 points: A successful demonstration will probably lead directly to purchase.
• 1 point: We expect an informal discussion, without a documented decision or transition plan.
• 2 points: We have a decision meeting, accountable approvers and a memo template covering evidence, cost, limitations and alternatives.
• 3 points: We have also defined sign-off conditions, conflict disclosures, migration and exit checks, plus a post-implementation review owner.

Evidence to attach: a retain/add/replace/consolidate/build/buy-nothing decision memo.

For replacement or consolidation, require a task-level explanation of what disappears when the incumbent is removed. Do not approve cancellation solely because two products occupy the same category.