Channel Strategy and Media Intelligence

The POEM model: paid, owned, earned, media

Before you map a single channel, you need a frame for what kinds of channels exist and what each one is for. POEM is the practitioner shorthand. Paid, Owned, Earned, Media. It is the same set of distinctions that the IPA databank measures separately when it studies effectiveness, but POEM is what working teams say out loud in a planning meeting. Each layer plays a distinct role and the mix only works when all four are deliberate.

• Paid. Channels you rent. Search ads, social ads, sponsorships, out-of-home, programmatic display, paid podcast spots. Paid buys reach into audiences who would not have arrived on their own. It is the only layer where you fully control timing and volume, and the only one where you can scale supply on demand.
• Owned. Channels you control. Your website, blog, newsletter, app, podcast, retail surface, customer community. Owned controls the narrative because nobody else is in the room. It compounds over time as content stacks and lists grow, and its cost per touch falls as the asset base widens.
• Earned. Channels you cannot buy. Press mentions, analyst coverage, organic word of mouth, customer-led Slack threads, unpaid reviews, conference stage invites. Earned generates third-party credibility that paid and owned cannot. It is the slowest layer to build and the most fragile to lose.
• Media. The integration layer that sits across the other three. The shared metric (reach, frequency, share of voice) that lets you compare a paid impression, an owned email open, and an earned analyst quote on the same page. Without a media layer the other three become silos and the team optimizes each one in isolation.

Worked example. A B2B SaaS at twelve million in annual revenue uses Paid (LinkedIn account-based marketing against twenty named target accounts) to plant a single thought in the buying committee. Owned (a weekly newsletter from the founder and a deep product blog) nurtures the same accounts between paid touches, so when a buyer searches for the category they land on the company's own narrative rather than a competitor's. Earned (one analyst mention, two customer-led Slack channels recommending the product) validates the claim that paid and owned have been making. The Media layer reports all three against the same twenty-account list. Paid drives the discovery curve, owned controls what the buyer reads next, earned converts the cautious buyers who would not have trusted paid or owned alone. Remove any layer and the other three lose effectiveness. A passive earned mention with no paid or owned wrapped around it is incidental, not a strategy.

The channel strategy mistake

The most common channel strategy error is starting with channels instead of audiences. "We should be on TikTok" is not a channel strategy. It is a platform preference. The correct starting point is always the audience: who are they, where are they, in what mindset, at what stage of the purchase journey? Channels are the answer to those questions, not the question itself.

Audience-first channel mapping

For each audience segment, map:

• Where they spend media time (platform + context)
• What mindset they are in when consuming that channel (entertainment, research, utility, social)
• What stage of the journey they are in when they encounter you there (unaware, aware, considering, decided)
• What job they need done in that moment

This mapping tells you which channels deserve investment and what job each one should do. It also tells you which channels to avoid: the ones where your audience is, but in the wrong mindset for what you need to communicate.

Binet's 60/40 rule

Les Binet's analysis of the IPA Databank, the largest database of advertising effectiveness evidence in the world, established one of the most important principles in media planning: approximately 60% of marketing budget should go to broad-reach brand building (emotional, memorable, reach-focused) and 40% to short-term activation (targeted, rational, conversion-focused).

The ratio is not rigid, it varies by category, brand maturity, and market context. But the principle is: brands that over-invest in activation at the expense of brand building see short-term sales improvement followed by long-term brand and margin erosion. The 60/40 split protects against this.

The false dichotomy of digital vs. traditional

Digital channels are not better than traditional channels. Traditional channels are not more credible than digital. The relevant questions are attention quality, context, and reach. A well-placed out-of-home placement in the right context can deliver stronger brand recall than a programmatic impression in a cluttered feed. The format is not the variable. The quality of attention is.

Channel roles

• Social media is a brand channel, not a sales channel. Its primary job is brand personality expression and community building. Treating it as a direct response channel consistently underperforms.
• Search is a demand capture channel. It captures intent that already exists. It cannot create demand on its own. Never confuse search investment with demand generation.
• Email is a relationship channel. For existing customers, it consistently outperforms every other digital channel on ROI. For acquisition, it is nearly useless.
• Content should follow the hub model: one anchor piece (deep, substantial, ownable) distributed across six or more channels in adapted formats.

The attribution trap

Last-click attribution is destroying brand budgets. When you credit the last touchpoint before conversion, you systematically undercredit brand-building channels (which work early in the journey and over long time horizons) and overcredit direct-response channels (which capture demand built by the brand). The result is a budget that keeps shifting toward activation until the brand equity that was making the activation efficient has been eroded.

The solution is not better attribution. It is accepting that brand effects are hard to attribute precisely and using a combination of econometrics, brand tracking, and share-of-voice analysis to make investment decisions.