What is frequency capping in CTV?
Frequency capping limits the number of times a single viewer (or household) is shown the same ad within a defined time window — for example, no more than 3 exposures per 7 days. In CTV, frequency is managed at the device ID level: the DSP or ad server tracks impressions against a specific device identifier (GAID on Android TV, TIFA on Samsung Tizen, AFAI on Fire OS) and stops serving once the cap is reached. Frequency capping matters in CTV because the lean-back viewing environment means heavy streamers can accumulate dozens of exposures to the same ad in a single session if uncapped — which drives up frequency costs and damages brand perception.
Why is frequency capping harder in India CTV than on mobile?
Three reasons: (1) Fragmented device IDs — different CTV operating systems use different device identifiers (GAID, TIFA, AFAI, LGUDID). A DSP buying across multiple publishers may see the same household as different devices if they use different OS platforms, leading to undercounting frequency. (2) SSAI publishers — JioHotstar uses server-side ad insertion, which means the DSP does not have direct visibility into individual impression events. Frequency counting relies on the publisher’s reported data, which may not sync in real time with the DSP’s own cap. (3) No cross-publisher deduplication — a cap set in DV360 for JioHotstar does not communicate with a separate buy on Sony LIV. The viewer may be capped within one publisher but over-exposed across the whole campaign.
What frequency cap should I set for an India CTV campaign?
Industry starting point for India CTV: 3–5 exposures per household per week for a brand awareness campaign; 2–3 per week for a performance-focused campaign where over-exposure to unconverted viewers wastes budget. For high-impact formats like live cricket on JioHotstar, cap at 2 per session to avoid viewer irritation during high-frequency ad breaks. Review frequency distribution reports — not just average frequency — after the first 48 hours. If the 75th percentile of exposed households has already hit 5+ exposures, tighten the cap or add publisher diversity to spread reach.