In India’s digital used vehicle market, digital advertising is fiercely competitive. Auto-tech platforms spend heavily on Meta to reach buyers and sellers but most of that spend goes toward broad audiences that deliver inconsistent results. The bigger the platform, the harder it becomes to find new, high-intent audiences beyond the users already in its own database.
This case study examines how Persona 360, Kentrix’s audience intelligence platform for digital advertising, helped one of India’s largest auto-tech companies achieve a cost-per-click of ₹1.81 – 64% below the industry average. This was done by replacing catalogue-based retargeting with MAID-based lookalike audiences built on verified income and lifestyle data.
Challenge – Limitations of First Party Lookalikes
Large auto-tech platforms typically build their Meta lookalike audiences from their own customer catalogues – a pool of past buyers, registered users, or city-specific leads. While this approach works early on, it’s ineffective as the campaign gets old because the quality of the lookalike is capped by the quality of the seed audience.
For this platform, the existing Mumbai city-catalogue lookalike was becoming expensive and less efficient over time. CPCs were running between ₹5 and ₹10, and CPMs were sitting at ₹150–220, both in line with the Indian auto industry average, but leaving significant room for improvement.
The core question the team was trying to answer – could an external audience intelligence tool, built on richer data signals, outperform a first-party lookalike built on years of their own customer data?
Persona 360 Approach – MAID Based Lookalike Audiences with Verified Income Signals
Persona 360 is Kentrix’s audience intelligence tool built on 920M+ Indian consumer profiles and 545M+ digitally addressable MAIDs (Mobile Advertising IDs). Unlike first-party catalogue data, which reflects only past customers, Persona 360 draws on deterministic household-level signals, including verified income tier, lifestyle affinity, vehicle ownership history, and spend behaviour across 40+ categories.
For this auto-tech campaign, Kentrix built three custom MAID-based audiences – each seeded from a different income and lifestyle tier within Persona 360’s segmentation framework. These were then activated on Meta using an ABO (Ad-Set Budget Optimisation) structure, giving each audience cohort an independent budget so Meta’s algorithm could identify the best-performing segment organically.
The key differentiator is that Persona 360’s seed audiences are built from 545M MAIDs as the base universe, giving Meta’s algorithm significantly more signal to work with compared to a brand’s own first-party catalogue, which is typically limited to a few hundred thousand records at most. More signal means faster learning, tighter lookalikes, and lower CPMs from the first week of the campaign.
Why Did This Work?
Most lookalike audiences in India are built from app install data, purchase histories, or CRM exports. These are behavioural signals, they tell you what someone has done, but not who they are. Two people who both bought a used car last year could have very different income levels, lifestyle profiles, and likelihood of buying again.
Persona 360 approaches lookalike modelling differently. Its seed audiences are built on verified income tiers (ESI – Economic Segmentation India), psychographic archetypes (LSI – Lifestyle Affinity Segmentation India), and spend-behaviour signals across vehicle ownership, insurance, and travel categories. This means the MAID pool passed to Meta as a seed is not just large, it is structurally more qualified than a standard first-party catalogue.
The three-cohort ABO structure amplified this further. By running KENTRIX1, KENTRIX2, and KENTRIX3 as independent ad sets with separate budgets, Meta’s algorithm identified KENTRIX3 as the highest-performing segment and allocated spend accordingly, compressing CPM and CPC without any manual intervention.
This is the same ABO multi-cohort architecture that has delivered consistent results across Persona 360’s auto, BFSI, OTT, and D2C campaigns.
What This Means for Auto Tech Platforms & Dealers in India?
India’s used vehicle market is one of the most competitive digital advertising categories in the country. Auto OEMs, used car platforms, EV brands, and dealerships are all competing for the same high-intent buyers on Meta which drives up CPMs and makes efficiency gains harder.
This campaign demonstrates that MAID-based lookalike audiences built on verified income and lifestyle data can consistently outperform first-party catalogue lookalikes, even when those catalogues belong to large, data-rich platforms. The advantage is not marginal. A 2.4X improvement in cost-per-click against your own best-performing audience is a structural shift.
Key Takeaways
- Persona 360’s MAID-based lookalike audiences delivered ₹1.81 CPC, 64% below the Indian auto industry average of ₹5–10.
- The Persona 360 lookalike outperformed the client’s own first-party Mumbai catalogue lookalike by 2.4X on cost-per-click.
- Seed quality is the primary driver of lookalike quality. Persona 360’s 545M+ MAID universe, enriched with income, lifestyle, and spend signals produced structurally richer seeds than first-party app or CRM data.
- The three-cohort ABO architecture is replicable, this model works across auto, BFSI, OTT, D2C, and real estate campaigns.
FAQs
Can Persona 360 outperform a large platform’s own first-party data?
Yes, this campaign is the clearest evidence. One of India’s largest auto-tech platforms, with years of proprietary customer data, saw its own Mumbai catalogue lookalike outperform Persona 360 by 2.4× on cost-per-click.
The reason is structural: Persona 360’s seed universe of 545M+ MAIDs with income and lifestyle enrichment gives Meta’s algorithm more and better signal than any single brand’s first-party database can provide.
What is a MAID-based lookalike audience and how is it different from a standard Meta lookalike?
A standard Meta lookalike is built by uploading a brand’s own customer list – emails, phone numbers, past behaviour and asking Meta to find similar users.
A MAID-based lookalike, as used in Persona 360 campaigns, seeds the audience from a pre-verified pool of Mobile Advertising IDs enriched with offline signals: income tier, lifestyle segment, vehicle ownership history, and spend behaviour. Because the seed is richer and the MAID universe is larger (545M+ in Persona 360’s case), the resulting lookalike is more qualified and typically delivers a lower CPC and CPM from the start.
