Moment Architecture
Most media plans answer the wrong question — "where do we show up?" The right question: what behaviour state is your audience in, and what do they need from you in that exact moment?
How to plan a high-attention window — a tournament, a festival, a launch — so the brand enters the audience's behaviour instead of interrupting it. Reach tells you who could see you; the behaviour state decides what they can do with you.
The proof case is behaviour-adjacent: Swiggy already sells what match-watching amplifies. For brands with no natural role in the moment's behaviour states, this framework diagnoses but does not yet prescribe — that question is the research program behind The Embedding Line.
Five layers, multiplied — not five checkboxes.
Each layer is a question. A plan that can't answer all five isn't a moment plan — it's a reach buy wearing one.
State
What behaviour state is the audience in?
Surface
Which platform hosts that behaviour — not which has most reach?
Signal
What contextual trigger in the moment can the brand connect to? A six. A wicket. A break.
Message
What do you say in that state, on that surface, at that signal?
Measure
What outcome metric proves it worked?
Four tiers — each answers a harder question.
Reach, impressions, frequency. Table stakes — necessary, proves nothing.
VTR, completion, CTR, shares, saves. Did the message connect with the state?
In-moment conversion vs baseline; share of search vs competitors. Did behaviour move?
Query-type evolution — the shift from generic brand searches to brand-plus-moment searches during the window. When people stop searching your brand and start searching your brand + a moment, association is forming. An original Forethinker metric — a real-time proxy, not a replacement for brand tracking (limit stated below).
Two IPL seasons, two different layers of the same architecture.
Two separate activations in two separate years — kept distinct on purpose.
66% off, auto-triggered on every six, live for six minutes. That season produced 1,294 sixes = 1,294 fires. Result: 4× CTR vs benchmark and the highest CTV video-through rate Swiggy had recorded. The ad didn't interrupt the match — the match triggered the ad.
Food ordering inside the live stream — 690+ cities, 37M users. The behaviour (ordering food while watching) and the surface (the stream itself) merged; the distance between attention and action approached zero.
What this framework does not yet answer.
- The adjacency question — the big one. Swiggy's proof rides a built-in advantage: food delivery is already a match-watching behaviour. A brand with no natural role in any match-time state gets a diagnosis from this framework ("presence alone won't convert") but not yet a prescription. Whether such brands should Borrow, Build over seasons, or sit out is the open research front — carried into The Embedding Line program.
- Tier 4 is a proxy, not a verdict. Query-type evolution needs meaningful search volume, and true memory-level association still lives in brand-lift studies and tracking. Best real-time signal available; never the whole answer.
- One deep case, one category, one country. Two activations across two seasons make a strong single-brand proof — they do not make a law. Cross-category calibration is open.
- Number provenance: Sixes results are mCanvas/AFAQS-reported (vendor-linked); the 2026 integration figures are from a JioStar press release; recall uplift is AdCounty-sourced. All flagged, none independently audited.
Five questions before the big moment.
What behaviour state will your audience actually be in — discovering, coordinating, acting, or escaping?
Which surface hosts that behaviour — regardless of where the reach is cheapest?
What signal inside the moment can your brand legitimately attach to?
Does your message fit that state — or was it written for a person on a sofa who doesn't exist mid-match?
Which tier of the measurement stack will you accept as proof — decided before the window opens, not after?
Every claim, with its label.
Open the evidence ledger
| Source | Claim it supports | Tier | Label |
|---|---|---|---|
| mCanvas / AFAQS; ESPNcricinfo (sixes count) | Swiggy Sixes 2025: six-triggered offer, 1,294 fires, 4× CTR vs benchmark, record CTV VTR | B | Observed · vendor-linked |
| JioStar press release, May 2026 | JioHotstar × Swiggy in-stream ordering; 690+ cities, 37M users | B | Observed · company-reported |
| TAM Sports 2026; BCCI Clothing Regulations 2026 | 45+ brands per match (down from 65+); up to 10 sponsor logos per jersey | A/B | Observed |
| Industry reporting (AdCounty); category data | Food delivery spikes 2–3× during matches; powerplay/decisive-moment ads earn 30–40% higher recall | B | Supported · directional |
| Snap Inc., 2025 | 85% of Indian Gen Z follow the IPL (audience context) | B | Observed · vendor-cited |