CTV measurement setup for India: what to track, what to ignore, and how to report it

CTV measurement in India has a structural problem that most campaign reports paper over: the platform with the largest share of India CTV viewing — JioHotstar — uses server-side ad insertion (SSAI), which prevents client-side tracking pixels from firing. This means the standard digital measurement playbook — drop an IAS or DoubleVerify tag, fire quartile pixels, count verified impressions — does not work as expected on India's most important CTV inventory. This guide explains what you can actually measure, how to set it up correctly, and how to build a post-campaign report that reflects reality rather than platform-flattering self-reported data.

Understanding the SSAI tracking gap

SSAI (server-side ad insertion) stitches the ad video into the content stream at the server level before it reaches the viewer's device. From the video player's perspective, there is no ad break — just a continuous stream. This solves buffering and ad blocking problems for the publisher. It creates a measurement problem for the advertiser.

The standard VAST tracking mechanism works like this: the video player on the viewer's device parses the VAST XML, downloads the creative, fires impression and quartile pixels as playback progresses. All of this happens client-side — on the viewer's device. With SSAI, the ad is already stitched into the stream before the device receives it. The device's video player never parses a VAST tag. Client-side pixels never fire.

The consequence: your third-party verification vendor's impression pixel cannot fire. Your IAS or DoubleVerify tag cannot confirm a viewable impression. Your MOAT attention signal cannot run. What you receive instead is the platform's own server-side reporting — impressions counted when the ad was stitched into the stream, not when it was verified as delivered to a human viewer.

This is not a reason to avoid JioHotstar. It is a reason to calibrate your measurement expectations correctly before the campaign runs — and to stop benchmarking it against display viewability standards it structurally cannot meet.

What to measure and what to ignore

Measure these

Treat with caution

• Viewability rate from third-party vendors on SSAI platforms — if IAS/DV reports viewability on JioHotstar, understand how those numbers are derived. They are modelled estimates, not client-side pixel data.
• Click-through rate (CTR) — CTV CTR is near-zero (0.03–0.1%) because most viewers do not use a remote to click through. Do not use CTR as a CTV quality signal.
• Platform-reported brand lift — platforms have an incentive to show positive lift. Treat proprietary brand lift studies as directionally useful but not as rigorous as independently fielded surveys.

Ignore on CTV

• Bounce rate from CTV-sourced traffic — the small number of CTV click-throughs go to a second screen (phone) after a QR scan, not to a browser on the TV. Bounce rate is not meaningful.
• Cookie-based audience retargeting confirmation — CTV does not set cookies. You cannot confirm cookie-based retargeting pool membership from CTV exposure.
• Viewable CPM (vCPM) benchmarked against display — CTV viewability is structurally different from display. A 100% in-view completion rate on non-skippable CTV pre-roll is not comparable to a display vCPM.

India CTV measurement benchmarks (2025–26)

Setting up third-party verification

Third-party verification on India CTV requires understanding what each vendor can and cannot do on each platform before you include their tags.

YouTube CTV

YouTube CTV supports third-party verification tags from IAS, DoubleVerify, and MOAT. Tags fire client-side within the YouTube app. Setup: add the verification vendor's VAST wrapper tag to your DV360 or Google Ads campaign. Confirm with your verification vendor that they have an active YouTube CTV integration — not all vendors' YouTube integrations cover the CTV app environment equally.

JioHotstar

Standard third-party client-side pixel verification does not work on SSAI-served inventory. What you can do:

• Request platform-side server log reporting — impressions counted when the ad is stitched into the stream with device type, content category, and geo breakdowns
• Ask whether the platform supports SSAI-compatible server-to-server (S2S) ping for impression tracking — some platforms are building this capability
• Negotiate impression audit clauses into the IO: the platform agrees to allow post-campaign log-level data review for large buys (Rs 25 lakh+)
• Accept that independent third-party viewability is not available and price the campaign accordingly — you are trusting platform-reported data

Amazon DSP / Fire TV

Amazon DSP supports some third-party verification through its measurement partner programme. IAS has an Amazon DSP integration. Confirm with your verification vendor whether their Amazon DSP integration covers Fire TV in-app placements specifically.

Setting up brand lift measurement

Brand lift is the most actionable upper-funnel measurement for India CTV. It measures the change in brand awareness, consideration, or purchase intent among people exposed to your campaign vs a control group who was not. Setup varies by platform:

YouTube CTV Brand Lift (Google)

1. Minimum impression threshold: approximately 1 million impressions. Below this, survey sample sizes are too small for statistical significance.
2. Enable Brand Lift in your DV360 campaign settings or Google Ads campaign before the campaign starts — it cannot be added retroactively.
3. Define the survey question type: aided brand awareness, ad recall, consideration, or purchase intent. You can run multiple question types but each requires its own impression threshold.
4. Google automatically recruits survey respondents from exposed and unexposed YouTube users and fields the survey within the YouTube app.
5. Results available in Google Ads Brand Lift reporting or DV360 Insights within 2–4 weeks of campaign completion.

JioHotstar brand lift

JioHotstar offers proprietary brand lift studies as paid add-ons, typically priced at Rs 3–8 lakh on top of media spend. Methodology: platform-fielded surveys to exposed users within the app. The control group is constructed from unexposed users in the same content environment. Ask the platform for their methodology documentation — specifically how the control group is constructed and whether it controls for content affinity differences between exposed and unexposed groups.

Independent panel-based brand lift

For large campaigns (Rs 50 lakh+), independent brand lift studies fielded through a third-party research firm (Nielsen, Kantar, IMRB) provide the most defensible results. These are panel-based rather than in-app survey based, which reduces selection bias from in-app survey respondents. Cost: Rs 10–25 lakh for a properly designed study. Recommended for brand campaigns where proving CTV's contribution to brand metrics is a business objective rather than just a reporting requirement.

Attribution: what is realistic on India CTV

Direct attribution of business outcomes to India CTV exposure is limited by the same structural constraints as viewability measurement. What is achievable:

Search lift

On YouTube CTV campaigns via DV360, Google can measure whether exposed users showed higher search query volume for your brand in the weeks following exposure vs the control group. This is a clean signal of CTV's impact on consideration because it is measured on a deterministic platform (Google Search) using Google's login graph to connect CTV exposure to search behaviour.

App install attribution

If your campaign objective is app installs, QR codes on CTV creative provide a trackable bridge: the viewer scans the QR code on their phone, lands on the app store or a UTM-tagged landing page, and the install is attributed to the CTV QR source. This is the most direct attribution available for India CTV performance objectives.

Website traffic and UTM tracking

Add UTM parameters to all CTV click-through URLs. CTV click-through rates are low (0.03–0.1%) but the viewers who do click are high-intent. Tag these sessions separately in your analytics so the small CTV-originated traffic pool is not diluted into overall digital traffic.

Footfall and sales attribution (offline)

For brands with offline sales, geo-matched panel studies can correlate CTV campaign markets with sales lift. This requires a large enough campaign (typically Rs 1 crore+ in a single geography) to produce detectable signal. Nielsen, Kantar, and select India-specific measurement firms offer this methodology.

Building a post-campaign report

A CTV post-campaign report that accurately reflects performance has these sections:

1. Delivery summary

Impressions served vs contracted, completion rate overall and by quartile, unique household reach, average frequency, geographic and device breakdown. Source: platform-reported. Flag if data is SSAI server-side vs client-side verified.

2. Creative performance

Completion rate by creative version (if running multiple). Quartile falloff analysis — where did viewers drop relative to ad length? For skippable YouTube TrueView: view rate (% who watched past the skip point).

3. Audience delivery

Did the campaign deliver against the target audience? For programmatic campaigns via DV360: Google audience segment reach. For direct campaigns: platform-reported demographic delivery vs plan. Flag any significant deviation from planned targeting.

4. Brand impact (if measured)

Brand lift results: absolute lift in awareness/consideration/recall, percentage point change vs control, statistical significance level. Note the methodology and who fielded the study (platform proprietary vs independent).

5. Downstream signals

Search lift data (YouTube CTV), UTM-tagged CTV click-through sessions, QR code scan volume and downstream conversion rate, app installs attributed to CTV QR source.

6. Efficiency metrics

Cost per completed view (CPCV), effective CPM against target audience, cost per household reach point. Compare against pre-campaign benchmarks and plan.

7. Recommendations

What performed above benchmark and why. What underperformed and the likely cause. What to change in the next flight: creative length, platform mix, targeting, frequency cap, format. Be specific — "increase frequency cap from 3 to 5 on JioHotstar based on low delivery in week 2" is useful; "optimise the campaign" is not.

Measurement setup checklist

→ Use the CTV Measurement Audit tool to assess your current measurement capability against best practice before setting up your next campaign.