Multi-touch attribution (MTA) distributes conversion credit across all the marketing touchpoints a customer encountered before converting, rather than giving all credit to the last one. Models include: Linear (equal credit to all touchpoints), Time-decay (more credit to touchpoints closer to conversion), Position-based (heavy credit to first and last touch), and Data-driven (algorithmic model trained on your own conversion data). MTA is better than last-touch for CTV because it can include the CTV impression in the credit allocation across the full journey.
The limitations of MTA for India CTV: identity resolution is required to link a CTV impression (device-level) to a subsequent conversion on mobile or desktop (user-level). Across different identity systems (Jio CTV impression to non-Google conversion), the linkage breaks. Data-driven MTA requires high conversion volumes (1,000+ monthly conversions minimum) to train a reliable model. Most India CTV campaigns targeting mid-market brands do not reach this threshold. MTA is worth using where identity resolution and conversion volume allow it. Where they do not, incrementality testing or media mix modelling is more appropriate for CTV.
Full guide
For a complete explanation, read: Multi-touch attribution for CTV: models, limitations, and when to use it in India