The channel map
What each advertising channel is actually good at, and how intent and interruption differ.
Every channel sells attention, but not the same kind of attention. The single most useful split is whether the person was already looking for you.
Intent versus interruption
| Intent channels | Interruption channels | |
|---|---|---|
| Examples | Search, shopping feeds, marketplaces | Social feeds, short video, display, connected TV |
| The user is | Actively searching for a solution | Doing something else |
| Creative's job | Match the query and remove friction | Stop the scroll and create the want |
| Volume ceiling | Capped by search demand | Capped by audience size and creative supply |
| Typical cost curve | Cheap at low volume, expensive to expand | Expensive to learn, cheaper to scale |
Intent channels harvest demand that already exists. Interruption channels create it. An account that only harvests eventually runs out of harvest; an account that only creates demand leaves easy conversions on the table.
What each channel is for
| Channel | Strongest at | Weakest at |
|---|---|---|
| Search | Capturing existing demand, brand defence | Making anyone want something new |
| Shopping / product feeds | Considered purchases with visual comparison | Products needing explanation |
| Meta (Facebook, Instagram) | Broad prospecting, creative-led scaling | Very long or highly technical sales cycles |
| Short video (TikTok, Reels, Shorts) | Cheap reach, discovery, trend-fed hooks | Complex offers, high price points |
| YouTube | Explaining, demonstrating, building trust | Fast, low-effort test cycles |
| Display / native | Cheap reach, remarketing | Attention quality |
| Email and owned lists | Repeat purchase, margin | Reaching anyone new |
Picking the mix
A workable order of operations for a young account:
- Cover the intent you already have — brand and category search, product feeds
- Build one interruption channel properly before opening a second one
- Add a second channel only when the first has more than one winning creative
- Keep owned channels running in parallel; they cost the least per order
The common failure is opening four channels at once with one creative each. Every channel then sits in a permanent learning phase, and none of them produces a readable result.
Budget split is a consequence, not a decision. Fund the channel that is already returning above your target and starve the one that is not, rather than fixing percentages in advance.
Reading cost differences between channels
CPM differences between channels are not a quality ranking. A channel with a low CPM and a low conversion rate can cost more per order than an expensive one. Only compare channels at the level of cost per outcome, and only after measurement is trustworthy on both.