Most India CTV buys are demographic buys. You tell the platform you want SEC A women 25–44 in metros. The platform tells you they have exactly that. You buy. What you actually get is anyone watching that content who roughly fits that cohort — with no way to verify, deduplicate, or suppress. This guide is for brand and agency teams who want to stop buying demographics and start buying audiences they actually know.
What this guide covers
1 Why audience data changes CTV economics
Audience data changes three things. First, targeting precision: instead of buying demographics you're buying people with a known relationship to your brand — site visitors, app users, CRM customers, lapsed buyers. Second, suppression: you stop wasting budget on existing customers when the goal is acquisition, or stop pestering customers who just bought. Third, measurement: when you know who you targeted, you can measure actual outcomes — not just VCR and impressions.
The CPM premium for audience-targeted CTV buys in India is real (Rs 50–150 incremental per thousand), but the efficiency gain is larger. A well-targeted campaign at Rs 250 CPM reaching the right 50,000 people outperforms a demographic buy at Rs 150 CPM reaching 200,000 people who may never buy your product.
Demographic buy CPM
Rs 150–200
Standard JioHotstar / SonyLIV direct deal. Broad age/gender/metro targeting.
Audience-targeted CPM
Rs 250–400
First-party or third-party segment activation. Higher cost, higher relevance, measurably better outcomes.
Suppression value
15–30%
Typical budget saved by suppressing existing customers or recent converters from acquisition campaigns.
Retargeting lift
2–4×
Typical conversion rate lift when retargeting known site/app visitors on CTV vs. cold demographic targeting.
2 What first-party data you already have
Before building anything new, audit what exists. Most brands with a digital presence have more first-party data than they realise — it's just not organised for CTV activation.
| Data source | What it contains | CTV use case | Activation readiness |
|---|---|---|---|
| Website visitors (GA4 / pixel) | Cookie/device IDs, pages viewed, time on site, product interest signals | Retargeting site visitors on YouTube CTV / Amazon DSP | High — Google and Amazon both accept website audiences directly |
| App users (Firebase / AppsFlyer) | Mobile device IDs (GAID/IDFA), in-app events, purchase history | Cross-device retargeting; find CTV households of mobile app users | Medium — requires identity resolution step for CTV match |
| CRM / customer database | Email addresses, phone numbers, purchase history, loyalty tier | Customer match audiences for YouTube CTV; suppression lists | High for YouTube (Customer Match); medium for other platforms |
| E-commerce transaction data | Purchase category, value, recency, frequency | Lapsed buyer re-engagement; upsell to high-value segments | Medium — requires upload to platform or DMP/CDP |
| Email subscriber list | Email addresses, engagement history, preferences | Customer Match on YouTube; custom audiences on Amazon DSP | High — direct upload supported by both platforms |
| Offline / in-store data | Phone numbers, purchase records, store location visits | Match to digital IDs for cross-channel retargeting | Low — requires data onboarding partner (LiveRamp, etc.) |
India reality check
Indian consumers are more likely to share phone numbers than email addresses — especially in regional markets. If your CRM is phone-number-heavy and email-light, your Customer Match rates on YouTube will be lower than global benchmarks. Build phone-number hashing into your data pipeline from the start.
3 How to collect more — and the right way
The best first-party data is collected with explicit value exchange. Don't collect for collection's sake — collect data you will actually use within 90 days, and give the user a clear reason to share it.
High-value collection points for India brands
- App registration and login: Mobile apps are the primary collection surface in India. Require registration before content unlock (even for free tiers) and capture email + phone at sign-up. SMS OTP verification is standard and gives you a verified phone number at near-zero friction.
- Loyalty and rewards programmes: India consumers are highly responsive to loyalty mechanics. A points programme tied to purchase history gives you transaction data with identity attached. FMCG brands are well-placed here.
- Gated content / tools: Calculators, planners, diagnostic tools, and downloadable guides generate leads with explicit email capture.
- Contest and sweepstake mechanics: High conversion for email + phone capture in Indian markets. Ensure DPDP Act compliance on consent capture.
- WhatsApp Business: WhatsApp opt-in via QR code or click-to-chat gives you a verified phone number. Increasingly used by D2C and consumer brands in India for first-party capture.
DPDP Act compliance (India)
India's Digital Personal Data Protection Act (2023) requires explicit, informed, purpose-specific consent for personal data collection. Your consent capture must specify what data is being collected, why, and how it will be used — including advertising use cases. Generic "I agree to T&Cs" does not constitute valid consent under DPDP. Review your consent flows before building a first-party data programme.
Data you should NOT collect
Don't build a CRM data programme around third-party cookies — they are deprecated in Chrome and blocked in Safari/Firefox. Don't rely on device fingerprinting; it's unreliable on CTV and legally grey under DPDP. Don't purchase third-party audience lists from Indian data brokers — list quality is poor, consent provenance is unclear, and platform match rates are typically under 10%.
4 Identity resolution for CTV in India
The fundamental challenge of CTV audience targeting is that your first-party data lives in mobile and web environments (cookies, GAIDs, email addresses) while CTV screens are a different device with different identifiers. Identity resolution bridges these environments.
The practical flow: you upload hashed email addresses or phone numbers to a platform (YouTube, Amazon DSP). The platform matches against its own logged-in user graph. On YouTube, a user logged into their Google account on their smart TV is matched to their email — your CRM record connects to a CTV impression. Amazon does the same via Amazon.in account emails and Prime subscriber data.
| Platform | Matching mechanism | Typical India match rate | What affects match rate |
|---|---|---|---|
| YouTube (Customer Match) | Hashed email / phone matched to Google account | 30–55% for email lists; 40–65% for phone number lists | Email vs. phone mix; Gmail vs. non-Gmail; list recency |
| Amazon DSP | Hashed email matched to Amazon.in account | 20–40% | Amazon purchase history of your audience; Prime membership rate |
| JioHotstar (direct deal) | No self-serve audience upload; platform applies its own segments | Not applicable | Platform-defined segments only; no custom audience activation |
| Samsung Ads | ACR-based segments; no external audience upload | Not applicable | Samsung TV ownership; platform-defined ACR segments |
The India CTV identity reality
India CTV identity resolution is essentially a two-platform story: Google (YouTube CTV) and Amazon (Fire TV / Amazon DSP). JioHotstar, SonyLIV, and Zee5 do not currently support advertiser-side custom audience activation. If your data strategy depends on activating first-party audiences on JioHotstar's massive CTV reach, you are ahead of what the platform currently supports.
Cross-device identity for CTV households
On YouTube: if the same Google account is used on mobile and on the smart TV, your mobile audience segment automatically extends to CTV screens — no additional step needed. This is the cleanest cross-device path in India.
On Amazon DSP: Amazon's household graph connects Fire TV devices to Amazon.in accounts. Coverage is lower than YouTube's because it requires Fire TV device ownership. Beyond Google and Amazon: no reliable cross-device path exists in India currently. The third-party identity resolution infrastructure (LiveRamp, ID5) that enables this in the US has limited India scale.
5 Building segments that work on CTV
CTV is a lean-back, household, premium environment. Segments that perform well on mobile performance marketing often underperform on CTV — and vice versa.
| Segment type | How to build it | Why it works on CTV | Platform |
|---|---|---|---|
| High-intent site visitors (non-converters) | GA4 audience: visited product/pricing page, did not purchase, within 30 days | Upper-funnel CTV reinforces consideration for a known intent signal. 15–30s pre-roll at home extends the mobile research journey. | YouTube CTV via DV360 |
| Lapsed customers (90–365 days) | CRM export: last purchase 90–365 days ago, no recent activity | CTV's brand-building format re-activates dormant customers better than direct-response email or push notification. | YouTube CTV Customer Match; Amazon DSP |
| App installers (non-purchasers) | AppsFlyer/Firebase: installed app, never completed a purchase event | CTV ad for a mobile app is counterintuitive but effective — household awareness creates social purchase intent. | YouTube CTV (cross-device via Google account) |
| High-LTV customer lookalikes | Upload top-20% LTV customer list → Google Similar Audiences or Amazon Lookalike | Platform expands to statistically similar users — useful for prospecting without platform-only demographic targeting. | YouTube CTV (Similar Audiences); Amazon DSP (Lookalike) |
| Category in-market (third party) | Google Affinity / In-Market audience (auto finance, consumer electronics, etc.) | No first-party data required. Useful for new brands or categories where you have no CRM. Lower precision but broad scale. | YouTube CTV via DV360 |
Segments to avoid on CTV
- Retargeting windows under 7 days: CTV is not a direct-response channel. Showing a product someone browsed 2 hours ago on a family TV screen is intrusive. Use 7–30 day windows.
- Very small segments (<5,000 matched users): Build segments of 20,000+ matched users before activating on CTV. Small segments deliver inconsistently.
- Ultra-granular demographic overlays: Stacking age + gender + income + city on top of a first-party audience reduces scale to the point where delivery fails. Let the first-party data do the targeting.
6 Activating on India CTV platforms
YouTube CTV via DV360
This is the most capable audience activation path for India CTV. DV360 lets you activate Google Audiences (Customer Match, Similar Audiences, In-Market, Affinity) with a Connected TV device type filter — ensuring your budget delivers to TV screens only, not mobile or desktop.
- Upload your hashed email/phone list to Google Ads as a Customer Match audience
- Wait 24–48 hours for the audience to populate (minimum 1,000 matched users required)
- In DV360, create a new insertion order with YouTube inventory
- Set device targeting: Connected TV only
- Apply your Customer Match audience (or Similar Audience built from it)
- Set frequency cap: 3–5 per week per household for awareness; 1–2 per week for retargeting
Note: Google Ads Customer Match requires an account spend history of $50,000 (approximately Rs 40 lakh) for eligibility. New advertisers without this history should use In-Market or Affinity audiences as a bridge.
Amazon DSP
Amazon DSP is the right choice when your audience has a purchase intent signal that aligns with Amazon's data advantage — consumer electronics, FMCG, apparel, home goods. Upload hashed emails to Amazon Ads as a customer list, then use Amazon DSP to target those users across Fire TV, Amazon Freevee, and third-party inventory. Access requires either Amazon managed service (minimum Rs 10–15 lakh) or a certified partner agency.
Direct-deal platforms (JioHotstar, SonyLIV, Zee5)
Custom audience activation is not available on these platforms. Alternatives:
- Content adjacency targeting: Buy inventory adjacent to content your audience is likely watching. If you're targeting cricket fans, buy IPL. If you're targeting South Indian consumers, buy SonyLIV regional content.
- Platform-defined interest segments: JioHotstar offers interest and genre segments (sports enthusiasts, auto intenders, etc.) built from their own viewing data — more precise than pure demographics.
Frequency management across platforms
India CTV buyers running YouTube CTV and JioHotstar simultaneously have no cross-platform frequency cap. A viewer can receive your ad 4 times from YouTube CTV and 4 times from JioHotstar in the same week. Practical solution: stagger flights (run JioHotstar in weeks 1–2, YouTube in weeks 3–4) rather than running both simultaneously.
7 Closing the loop: measurement
Exposed vs. control brand lift
Google Brand Lift is available on YouTube CTV campaigns spending Rs 15–20 lakh+. It shows a post-ad survey to people who were exposed vs. a holdout group, measuring awareness, recall, consideration, and purchase intent lift. This tells you which creative drove the most consideration among which audience segment.
Website / app conversion measurement
For Customer Match campaigns on YouTube CTV, Google can attribute site conversions (GA4 goals) back to CTV exposure. The attribution is probabilistic for cross-device paths but reasonably accurate for same-device paths. Set up GA4 goal imports into DV360 before the campaign launches.
CRM match-back
After the campaign, take your converted customers and match them back against your exposed audience list. What percentage of converters were in your targeted segment? This post-hoc analysis gives you a directional read on whether your CTV targeting is finding the right people — no platform tool required.
8 Your 90-day data roadmap
Days 1–30: Audit and organise
- Export your CRM data and assess volume by identifier type (email / phone / both)
- Check Google Ads account for Customer Match eligibility (requires Rs 40 lakh+ historical spend)
- Review DPDP Act consent capture on all data collection touchpoints
- Set up GA4 audience segments: site visitors, product page viewers, cart abandoners
- Create a hashed upload of your email list and test in Google Ads (check match rate)
Days 31–60: Build and activate
- Upload CRM segments (lapsed customers, high-LTV) to Google Ads Customer Match
- Build 2–3 GA4 audiences with 30-day lookback windows (site visitors, category browsers)
- Launch a YouTube CTV test campaign via DV360: Customer Match + Connected TV device filter
- Set up Google Brand Lift study if spend threshold is met
- Brief creative team: 15s and 30s versions optimised for lean-back TV viewing
Days 61–90: Measure and iterate
- Pull Brand Lift results: which audience segment showed highest consideration lift?
- Run CRM match-back: what % of converters were in your targeted audience?
- Compare VCR and view-through rate by audience segment — identify best-performing segments
- Build Similar Audiences from your highest-performing first-party segments
- Plan suppression integration: upload recent converters as exclusion list for next campaign