See the stack behind the television screen
The CTV experience includes hardware, operating system, app, aggregator, identity layer, ad stack, and content service. Power often sits with the layer that controls the home screen and account relationship. The paper highlights Roku, Amazon Fire TV, Samsung, Apple, LG, Google TV, and Vizio as distinct ecosystems rather than interchangeable delivery pipes.
For a streaming service, platform reach is only the first question. Leaders must also ask who owns login data, who controls promotion, what subscription or advertising share applies, which ad formats are available, and how measurement can be reconciled. A large audience can still be strategically expensive if the service loses margin and customer visibility.
Balance direct relationships and aggregation
Direct distribution generally offers greater margin, product control, and first-party data, but the service must fund discovery and acquisition. Aggregators and bundles can lower CAC, simplify sign-up, and place content in front of high-intent viewers, while taking fees and controlling valuable usage signals. The correct answer is usually a managed mix, not an ideological choice.
Forecast direct and indirect cohorts separately. Compare acquisition cost, activation, engagement, churn, realized revenue, platform share, data access, and upsell ability. A bundle may create a stable cohort even with lower direct ARPU; an aggregator may become risky when it supplies nearly all additions. Portfolio limits can prevent convenient distribution from turning the service into a commodity supplier.
Evaluate platforms with a power matrix
The paper's device landscape shows how hardware, advertising, commerce, and data are converging. Roku combines operating-system reach with advertising and aggregation. Amazon connects devices, subscriptions, advertising, and purchase behavior. Walmart's Vizio acquisition illustrates how a retailer can connect television usage and commerce data. Smart-TV manufacturers also influence placement and automatic content recognition.
Use a consistent scorecard instead of choosing platforms from headline scale. Weight each dimension according to strategy: a new service may prioritize discoverability and lower CAC, while a mature service may value identity ownership and direct ad integrations. Revisit the score as contracts, platform policies, audience behavior, and measurement access change.
| Dimension | What to assess | Why it matters |
|---|---|---|
| Reach | Households, audience fit, duplication | Incremental access, not gross scale |
| Discovery | Home-screen, search, promotion | Organic acquisition and engagement |
| Economics | Fees, shares, paid placement | Contribution margin |
| Data | Login, viewing, export, clean-room access | Retention and measurement |
| Ad stack | Inventory rights, formats, integrations | Yield and control |
Plan creative and measurement for fragmentation
Device and platform fragmentation affects technical compatibility, ad serving, latency, interaction design, and measurement. A creative that works on one remote or operating system may not work identically elsewhere. Household identity can also duplicate viewers across devices and platforms, inflating apparent reach and weakening frequency control.
Create a platform test plan that covers playback quality, tracking, valid delivery, interactive behavior, ad completion, and reconciliation. Use deduplicated household reporting where the evidence supports it, and disclose modeled overlap. Preserve platform-level results long enough to identify whether performance differences reflect audience, inventory quality, placement, price, or measurement.
Govern concentration and dependency
A platform portfolio should have explicit dependency thresholds. Track the share of additions, viewing, revenue, ad inventory, and customer data controlled by each partner. Also record contract renewal dates, policy risks, data-portability limits, and operational alternatives. This makes platform power visible before a negotiation or outage makes it urgent.
Review build, buy, and partner choices for capabilities such as billing, identity, advertising, and measurement. Build when control is strategically differentiating and economically justified; buy when a mature capability accelerates execution; partner when scale or data cannot be recreated efficiently. The goal is resilient reach without surrendering the ability to understand and serve the customer.
- Measure incremental reach after platform overlap.
- Compare contribution and retention by acquisition source.
- Score discovery, data, ad rights, and measurement access.
- Set concentration limits for critical partners.
- Maintain alternatives for billing, identity, and ad operations.
Decision implication
CTV platform planning is a portfolio exercise in reach, economics, data, and strategic control. Services that compare direct and indirect cohorts, test technical delivery, and monitor partner concentration can use aggregation for growth without losing the customer relationship that underpins long-term value.
Compare US CTV platform scenarios and dependencies with PyxiVisio.
Use a PyxiVisio decision-intelligence model to connect assumptions, delivery, economics, risk and approval conditions.