Opportunities

Find the opportunities hidden inside recurring problems.

Almost every worthwhile intervention in this ecosystem looks unremarkable from the outside: someone finally understood a problem well enough to act on it. These are the domains where that understanding is missing, and the hypotheses about what would follow if it were not.

Domains: 10
Hypotheses: 21
Evidence state: Untested

What this page is

Every entry below is a hypothesis. None is a validated business, a market estimate, a proven solution, or a claim that a gap has been confirmed to exist at a particular scale.

They are published in this form because a hypothesis that names its own evidence requirements, its alternatives, and its risks can be argued with — and arguing with them is the only way to find out which are worth testing.

The domains

Ten places where understanding something properly would change what is possible

Each domain names the problem as a hypothesis, who it lands on, what is missing, what could be tried, what evidence would be required, and what a real outcome would look like. None of these is a proposal to build a product — each is a description of what would need to be true first.

Production & Operations

Scheduling, approvals, procurement, payments, vendor coordination, and project learning.

The problem, as a hypothesis: Production workflows depend on a chain of approvals and handoffs, and the points where that chain stalls are rarely visible to anyone other than the people trapped inside it.
What is missing: There is no shared, comparable picture of where a production loses time or money, because each company's operational reality is observed privately and rarely aggregated in a usable form.
What could be tried: Aggregate anonymised workflow signals across productions to identify recurring bottlenecks, then test a targeted process change with the teams that own them.
Evidence required: Time-on-task and approval-cycle data from a range of productions; a documented baseline; a comparison design across at least two teams.
A real outcome: Measurable reduction in a specific, named bottleneck — measured against the pre-intervention baseline, not against an industry average.
Who it lands on: ProducersLine producersCoordinatorsAccountantsVendor managersCrew

Talent & Work

Discovery, professional identity, credits, skills, opportunity access, and career continuity.

The problem, as a hypothesis: Matching people to work depends on referrals and proximity, so capable people and suitable work fail to find each other even when both are available.
What is missing: Professional identity is assembled from inconsistent records across employers, so neither the person nor the hiring party can verify experience reliably or quickly.
What could be tried: Build a consented, portable work-and-credits record, and test matching approaches that surface candidates beyond existing referral networks.
Evidence required: Matching outcome data across participating productions; comparison of referred and non-referred candidate outcomes; consent and withdrawal rates.
A real outcome: Higher fill quality or shorter time-to-fill on matched work, with no reduction in retention or engagement.
Who it lands on: TechniciansPerformersCasting agenciesProducersLocation managers

Rights & Revenue

IP ownership, licensing, royalty visibility, metadata, attribution, and rights commercialisation.

The problem, as a hypothesis: Ownership chains, licensing terms, and royalty flows are opaque at every step, so value that is contractually owed is difficult to trace or claim.
What is missing: There is no shared, neutral representation of who owns what under which terms — each organisation holds its own partial truth, and no party is positioned to reconcile them.
What could be tried: Establish agreed rights metadata standards and a neutral dispute pathway, then test a lightweight clearance workflow with participating creators and rightsholders.
Evidence required: Chain-of-title reconstruction for a sample of works; royalty reconciliation against contracts; participant willingness to adopt a standard.
A real outcome: Reduced time and cost to clear a sampled work, with a documented reduction in unresolved royalty exceptions.
Who it lands on: Rights ownersPublishersLabelsCreatorsLicensing businessesPlatforms

Distribution & Discovery

Regional content distribution, independent film access, audience discovery, localisation, and cross-border reach.

The problem, as a hypothesis: Work that is finished and releasable frequently cannot find an audience, and the point at which it stops travelling is rarely identified or recorded.
What is missing: Because the journey of a title is not documented after release, the industry cannot distinguish a work that failed to travel from one that was never given the chance.
What could be tried: Instrument the distribution journey of a cohort of regional and independent titles, then test a targeted discovery and positioning intervention.
Evidence required: Title-by-title journey records with reasons for each step; comparable data for works that did and did not travel; exhibitor and platform willingness to participate.
A real outcome: A documented increase in reach for the treated cohort, with the mechanism for each increase understood well enough to explain it.
Who it lands on: Independent producersRegional producersDistributorsExhibitorsPlatformsAudiences

Creator Economy

Brand matching, platform dependence, campaign coordination, audience relationships, and sustainable monetisation.

The problem, as a hypothesis: Independent creators carry the risk of a single revenue relationship and rarely have a negotiating position, a rights position, or a plan for when their main platform changes.
What is missing: Creator and brand priorities are never expressed in a comparable form, so matching depends on whichever intermediary happened to know both sides.
What could be tried: Define a shared, minimal description of audience and intent, then test a matching mechanism that keeps both sides in control of their own data.
Evidence required: Matching outcomes across a cohort; creator-reported change in campaign alignment; evidence of whether concentration on one platform actually reduced.
A real outcome: More and better-matched partnerships per creator, and a measurable change in dependence concentration — not a claim that dependence is solved.
Who it lands on: Independent creatorsCreator managersBrand partnership teamsAgenciesPlatforms

Workforce & Education

Skills intelligence, training relevance, portfolio development, industry readiness, and employment outcomes.

The problem, as a hypothesis: Training providers cannot consistently demonstrate which programmes lead to work, because employment outcomes are not tracked back to the training that produced them.
What is missing: There is no agreed way to describe a person's working capability, so employers evaluate signals that are expensive to verify and students receive advice that is not evidence-based.
What could be tried: Agree a minimal outcome-tracking standard between providers and employers, and test whether publishing results changes programme behaviour.
Evidence required: Employment outcome data linked to cohorts; employer satisfaction measures; a control or comparison group of programmes.
A real outcome: A measurable improvement in placement or retention for tracked cohorts, with the limitations of the tracking method stated.
Who it lands on: Training providersAcademiesEmployersStudentsPolicy institutions

Live Entertainment

Venue discovery, event coordination, vendor capabilities, ticketing, accessibility, and event learning.

The problem, as a hypothesis: Small and mid-scale promoters make high-consequence decisions about venues, crews, and ticketing on incomplete information, and the consequences land on people who were not in the room.
What is missing: Operational reliability of a venue, crew, or ticketing provider is mostly known by reputation, which does not travel to a promoter working outside their usual circuit.
What could be tried: Publish a verified capability and reliability record for venues, crews, and ticketing providers, and test whether it improves event outcomes for new promoters.
Evidence required: Reliability outcomes for covered providers; promoter decision quality before and after access; event-level outcome data.
A real outcome: Fewer preventable event failures in the treated group, and a reduction in the cost of evaluating a provider from scratch.
Who it lands on: PromotersVenue operatorsCrewTicketing companiesAudiencesArtists

Audience & Community

Accessibility, affordability, language availability, representation, fan participation, and community value.

The problem, as a hypothesis: Barriers that prevent access are experienced directly by audiences but rarely reach the people who could remove them, because there is no usable channel carrying the detail.
What is missing: The people who notice a barrier are rarely the people with the authority to fix it, and the information does not currently travel between them in a structured way.
What could be tried: Build a consent-based access-obstruction reporting channel with defined routing to responsible parties, and track whether reported obstructions are resolved.
Evidence required: Reported obstruction volumes and resolution rates by responsible party; independent verification of a sample; participant trust measures.
A real outcome: A higher share of reported obstructions resolved, with a clear account of which obstructions remain unresolved and why.
Who it lands on: AudiencesAccessibility communitiesCultural organisationsExhibitorsPlatformsRegional audiences

Financing & Investment

Project readiness, information gaps, risk analysis, transparent assumptions, and access to relevant partners.

The problem, as a hypothesis: Capital is allocated against assumptions that are rarely tested, so the same expectation is repeatedly wrong in similar ways across unrelated projects.
What is missing: Project-level outcomes are rarely returned to the people who made the investment decision, so the industry repeats expensive errors instead of learning from them.
What could be tried: Establish a structured, confidential post-completion review standard and test whether it changes subsequent decision quality.
Evidence required: Pre- and post-project assumption records; comparable decision quality over time; financier willingness to participate.
A real outcome: A measurable reduction in the gap between documented assumption and observed outcome, with an explanation of what still cannot be predicted.
Who it lands on: InvestorsProducersFinanciersInsurersBanksRegional producers

Technology & Infrastructure

Workflow interoperability, AI adoption, secure collaboration, metadata standards, and underused capabilities.

The problem, as a hypothesis: Each company has invested heavily in tools that do not speak to each other, so the interoperability cost falls on the same production teams that are already under time pressure.
What is missing: The real obstacle to interoperability is rarely the software — it is that no shared convention exists for how production information should be expressed and exchanged.
What could be tried: Define a minimal shared metadata and exchange convention for a single workflow step, and test whether it removes meaningful friction before attempting broader standardisation.
Evidence required: Time spent on manual reconciliation for the targeted step; error rates; the willingness of competing vendors to conform to a shared convention.
A real outcome: Measurably less reconciliation time and fewer downstream errors for the targeted step, with a clear statement of what remains non-standard.
Who it lands on: Production teamsTechnology providersStudiosVendorsPlatforms

An honest note

These are not businesses. They are problems worth taking seriously.

Nothing on this page is a market size, a revenue forecast, a claim that a gap exists at a particular scale, or a validated business model. Where a number would be expected, there is a description of what would have to be measured to produce one.

The alternatives listed under each entry name the kind of approach already in use. The ones that exist as specific, named services — casting platforms, professional listings, audience measurement, the federations — are set out individually, with what each of them does and does not cover, on the landscape page.

A hypothesis here means nobody has yet established whether the problem is real, how common it is, or whether the proposed response would work.
The alternatives listed under each entry are real approaches already being used by people in the ecosystem. We have not evaluated which is best — that requires the evidence these entries lack.
The outcome measures are deliberately specific to a context and a baseline, because a result that cannot be compared with its own starting point is not a result.
Some of these will turn out to be wrong. Publishing them in this form is what makes that discoverable — and the first ones to be tested are likely to be the most obviously stated.

Share a problem Join the founding network