Deduplicated reach — also called unduplicated reach — means counting the number of unique people exposed to your advertising, regardless of how many times or on how many platforms they saw it. If the same person sees your ad on JioCinema CTV and again on YouTube CTV, deduplicated reach counts that as one person, not two. Without deduplication, adding together reach figures from different platforms double-counts audiences who use multiple platforms, inflating apparent campaign reach. In India, where audiences use several streaming services simultaneously, deduplication is a significant challenge — and an unsolved one for most campaigns.
Why deduplication is necessary
Consider a campaign running simultaneously on JioCinema CTV, YouTube CTV, and SonyLIV CTV:
- JioCinema reports: 4 million unique devices reached
- YouTube CTV reports: 3 million unique devices reached
- SonyLIV reports: 2 million unique devices reached
Summing these gives 9 million. But if 2 million of the JioCinema viewers also watched YouTube CTV, and 1 million watched SonyLIV, the true unduplicated reach is considerably less than 9 million. The exact overlap depends on how the platforms' audiences intersect — which no individual platform can tell you, because each platform only knows its own audience.
Without deduplication, campaign reach figures are inflated, effective frequency is understated, and CPM-per-unique-person comparisons are meaningless. For large multi-platform campaigns, the overlap problem is material — not a rounding error.
The identity infrastructure required for deduplication
Deduplication requires a common identity — a shared identifier that can link the same person or household across multiple platforms. Three types of identity are used:
Deterministic identity (login-based)
If Platform A and Platform B both have a shared user identifier — an email address, a phone number, or a universal ID — they can identify users who appear on both platforms and deduplicate. This requires either a single shared login (like a Google or Facebook account used to log in to both platforms) or a trusted third-party identity resolution service that both platforms have shared hashed data with.
Deterministic identity is highly accurate when available. The problem in India is that many CTV viewers are not logged in with a shared identity across platforms. A household watching JioCinema on their Jio account and YouTube CTV on their Google account will appear as completely separate identities to each platform.
Probabilistic identity (device graph matching)
When deterministic identity is unavailable, probabilistic matching uses device signals — IP addresses, viewing patterns, household geographic location — to infer that two different device IDs likely belong to the same household or person. An IP address is the most common household-level signal: if both the JioCinema CTV device and the YouTube CTV device share the same IP address, they probably belong to the same household.
Probabilistic matching is less precise than deterministic. It can over-deduplicate (incorrectly treating two different people at the same IP as the same person) or under-deduplicate (failing to connect the same person using different devices on different networks). The accuracy of probabilistic models varies by vendor and context.
Panel-based deduplication
A third approach uses a panel (recruited, consented households) to directly measure cross-platform exposure. Panel homes have measurement across all their devices, allowing direct observation of who was reached on Platform A and who was also reached on Platform B. The panel data is then statistically projected to the full population.
This is the most methodologically rigorous approach but requires a large enough panel to produce stable estimates — difficult when the measurement universe is a small number of CTV-specific households.
Vendors offering cross-publisher deduplication
Several measurement vendors offer cross-platform deduplication services. Their coverage in India varies:
- Nielsen ONE: In the US, Nielsen offers cross-platform deduplicated reach through its panel-plus-census methodology. India coverage is limited — Nielsen's India CTV data infrastructure is less developed than BARC's linear TV panel.
- Comscore: Active in India for digital audience measurement. Comscore provides cross-publisher digital reach measurement for web and some OTT platforms. CTV-specific deduplication across Indian platforms is not yet a standard Comscore product for India.
- DoubleVerify and IAS: These are primarily impression verification vendors, not reach deduplication vendors. They validate that an impression was delivered, not how many unique people were reached across platforms.
- Platform-provided cross-property measurement: Google/YouTube offers cross-property reach data within its ecosystem (YouTube CTV + YouTube mobile, for example). Meta offers similar within its properties. Cross-ecosystem deduplication (YouTube vs JioCinema) is not offered by either platform.
- Data clean rooms: Advertisers with sufficient data sophistication and platform relationships can set up data clean room arrangements — privacy-preserving environments where each platform contributes encrypted user data, and reach overlap is computed without either party exposing individual-level data. This approach is used by sophisticated global advertisers but requires significant technical investment and platform cooperation not yet widely available in India.
India-specific challenges for deduplication
India's CTV ecosystem faces several structural challenges that make cross-platform deduplication particularly difficult:
Fragmented identity landscape
JioCinema users are identified by Jio account or phone number. Hotstar (now JioHotstar post-merger) users have Disney+ or Hotstar account data. SonyLIV users have SonyLIV accounts. Zee5 users have Zee5 accounts. These identity systems do not interconnect. No universal advertising ID covers all platforms in India the way a Unified ID 2.0 or Google identity might aspire to in the US.
Low login rates on some platforms
Not all Indian CTV users are logged in to a platform when they watch. Guest viewing (where the app is used without an account) is more common on platforms without a strong subscription base. Guest views generate device-level impressions with no account identity attached — making cross-platform matching even harder.
Limited third-party measurement access
For cross-platform deduplication to work, each platform needs to either share data with a third-party measurement service or implement a common SDK. Indian platforms have been slow to implement consistent third-party measurement access compared to US platforms, partly due to competitive data concerns.
What planners should realistically expect
For India CTV campaigns in 2026, here is the honest assessment of what is possible:
- Within a single platform or ecosystem: Deduplication is generally available and reasonably accurate. JioCinema can tell you unduplicated reach within its own platform. YouTube can deduplicate across YouTube CTV and YouTube mobile.
- Across two owned-and-operated platforms of the same group: JioHotstar (post-merger of JioCinema and Hotstar) may be able to offer cross-property deduplication within its own platform. Watch for product announcements.
- Cross-publisher deduplication: Not reliably available in India today. If you run across JioCinema, SonyLIV, and YouTube CTV, you cannot get an audited unduplicated reach figure across all three.
- Practical approach: Report each platform's reach separately. If you need a combined reach estimate, use statistical modelling (market-level platform overlap estimates from research firms) and disclose the methodology and its limitations clearly.