Two managers argue over where the money goes. One wants a TV-style brand campaign, big, broad, expensive, hard to measure. The other wants performance ads that track every click to a sale. They are really arguing about media channels: the routes a message travels to reach a person. And most of the labels they're using were drawn on a budget sheet seventy years ago.

The quick version

  • ATL (above the line) is broad, mass-reach media, TV, radio, print, outdoor. BTL (below the line) is targeted, direct, measurable, events, direct mail, search, email. TTL (through the line) blends both.
  • The "line" isn't strategy. It came from how Procter & Gamble split its advertising budget in the 1950s, agencies that earned media commission went above it, everything else below.
  • "Digital vs offline" cuts across all of it. Digital can do mass reach (connected TV, YouTube) or pin-point targeting (search, retargeting); the channel isn't the strategy, the job is.
  • The useful question isn't ATL or BTL. It's the reach-versus-targeting trade-off, and the evidence says you need both, roughly a 60/40 split between long-term brand building and short-term activation.

The idea in depth: the "line" was an accounting choice, not a strategy

The vocabulary feels technical, so people assume it encodes a real distinction about how marketing works. It doesn't. The conventional account, repeated in standard texts such as Michael Baker's The Marketing Book, traces "above the line" to Procter & Gamble in the 1950s. When P&G laid out its promotional budget, the costs handled by advertising agencies, the agencies earned a commission on media they bought, like TV, radio, press and outdoor, were listed above a literal line on the page. Everything paid for a different way, sampling, coupons, direct mail, events, sat below it.

So the original split was about who got paid how, not about what worked. That history matters because it explains why the labels are slippery today. There is no agency-commission line any more. Yet the terms survived and acquired a second meaning, which is the one worth keeping: above the line came to stand for broad, untargeted mass reach, and below the line for targeted, direct, response-driven activity you can measure to a person. Through the line (TTL) names a campaign that deliberately does both, a TV ad that drives people to a trackable landing page is the classic example.

So here's the shift worth making: stop sorting channels by their label and start sorting them by the job they do. Ask of every channel: is this here to build broad awareness and future demand (a reach job), or to convert demand that already exists into a sale this week (a targeting job)? That single question does more work than the entire ATL/BTL vocabulary.

An honest limitation: the reach/targeting framing is cleaner than reality. Plenty of channels do both at once. A well-known YouTube creator integration is mass and targeted; a billboard outside a specific store is "ATL" media doing a very local "BTL" job. Treat the categories as a lens for thinking, not boxes every line item must fit.

flowchart TB
  M(["Your message"]) --> R(["Reach job
build future demand"])
  M --> T(["Targeting job
convert demand now"])
  R --> A(["ATL / broad media
TV, radio, print, outdoor,
online video, social reach"])
  T --> B(["BTL / direct media
search, email, events,
direct mail, retargeting"])
  A --> TTL(["Through the line:
broad media that hands off
to a trackable response"])
  B --> TTL
Sort channels by the job, not the label. Reach builds demand; targeting harvests it. Leaders Loop

Digital vs offline is the wrong axis, and a trust gap hides inside it

The other split people reach for is digital versus offline. It's intuitive and almost useless as a planning tool, because "digital" spans channels that do opposite jobs. Connected TV and YouTube are mass-reach media wearing a digital coat; paid search and retargeting are precision-targeting media. Grouping them because they share a screen tells you nothing about what they'll achieve.

What the channel does change is credibility, and here there's real evidence. Nielsen's Global Trust in Advertising study (fieldwork February–March 2013, more than 29,000 respondents across 58 countries) found that 84% of people trusted recommendations from friends and family above every other source, and 68% trusted online consumer reviews. Paid formats, TV, print, online display, were trusted by far fewer. The lesson isn't "abandon paid media." It's that earned credibility (a recommendation, a review) carries weight that bought attention can't manufacture.

This is why the cleaner modern map isn't ATL/BTL at all. It's the PESO model, Paid, Earned, Shared, Owned, coined by Gini Dietrich in her 2014 book Spin Sucks. Paid is advertising you buy. Earned is coverage and word of mouth you can't buy directly. Shared is social. Owned is your site, app, email list and shopfront. PESO sorts channels by who controls the message and how trust flows, which, unlike the old line, actually predicts how an audience will receive it.

Practically, that means mapping your channels to PESO before you brief anyone. If your whole plan is Paid, you're renting attention and inheriting the trust gap Nielsen measured. The cheapest credibility upgrade is usually Owned (a genuinely useful website, a good email) and Earned (reviews, referrals, PR), channels that cost effort rather than media spend.

Worth a caveat: the Nielsen trust figures are a snapshot from one (large, credible) survey, and trust isn't the same as sales impact, people buy plenty of things from ads they say they distrust. Use it to argue for balance across PESO, not to defund paid media on principle.

flowchart LR
  P(["Paid
ads you buy"]) --> AUD(["Your audience"])
  E(["Earned
PR, reviews, word of mouth"]) --> AUD
  S(["Shared
social, communities"]) --> AUD
  O(["Owned
site, app, email, store"]) --> AUD
  AUD --> RESULT(["Trust rises L to R:
bought attention is cheapest,
earned credibility is hardest to fake"])
The PESO model sorts channels by control and trust, a more useful map than the accounting "line." Leaders Loop

The split that the data actually supports: roughly 60/40

If channels are just routes for two jobs, building demand and converting it, the real strategic question is how much to spend on each. The most-cited answer comes from Les Binet and Peter Field, whose IPA study The Long and the Short of It (2013) analysed 996 campaigns entered into the IPA Effectiveness Awards between 1980 and 2010. Their headline finding: the most effective campaigns split spend roughly 60% to long-term brand building (broad-reach, emotional, the "reach job") and 40% to short-term sales activation (targeted, rational, the "targeting job"). Tilt too far either way and effectiveness drops.

"Reach builds the demand. Targeting harvests it. Spend everything on the harvest and there's nothing left to reap."

So before you debate specific channels, set the balance first. Earmark roughly 60% of working media for broad reach that builds the brand over months, and 40% for response channels that capture demand now. Then choose channels to fit each pot, and notice that "digital" and "offline" both show up on each side. The 60/40 is a starting heuristic to calibrate against your own results, not a law of physics.

And the limits of that number: the 60/40 figure is an average across the IPA dataset, weighted toward larger consumer brands and pre-2010 campaigns. A young start-up with no demand to harvest may need to lean harder into brand building; a pure-play e-commerce business with strong demand may justifiably run hotter on activation. Use the split as a default to argue from, then let your own data move it.

A worked example: a regional gym chain plans its year

The figures below are illustrative, to show the method, not benchmarks to copy.

Picture "Northside Fitness," a chain of six gyms with a modest annual marketing budget, say £120,000 of working media. The temptation, because it's measurable, is to pour everything into paid search and social ads: every pound tracks to a sign-up, and the dashboard looks great. Within a few months sign-ups plateau, costs per acquisition creep up, and the team is bidding against itself for the same small pool of people already searching "gym near me."

Here's the same budget planned by job. First, the split: about 60% (≈£72k) to reach, 40% (≈£48k) to targeting. On the reach side, building demand among people not yet searching, they run local radio and a sponsored community 10k run (ATL/offline), broad-reach video on YouTube and Instagram (ATL doing a digital job), and they invest in Owned: a genuinely useful website. On the targeting side they keep paid search, retargeting and a referral offer (BTL, mostly digital), and add a simple Earned play, asking happy members for Google reviews, which Nielsen's data suggests carries outsized trust.

What changed isn't the cleverness of any single ad. The plan now feeds its own funnel: reach channels create demand, targeting channels convert it, and Owned and Earned channels lift trust at the point of decision. The dashboard looks slightly worse for a quarter (reach is hard to attribute) and the business looks better within a year, exactly the lag the Binet and Field data predicts, and exactly why teams under monthly pressure keep starving the reach job.

Frequently asked questions

Is ATL/BTL outdated?

The accounting origin is dead, there's no agency-commission line to sit above or below. But the second meaning (broad-reach versus targeted-direct) is still useful shorthand, as long as you remember it's a lens, not a filing system. Many planners now prefer the PESO model (Paid, Earned, Shared, Owned) because it sorts channels by trust and control rather than a defunct budget convention.

Is digital always cheaper and better than offline?

No. Digital is easier to measure, which isn't the same as more effective. Cheap, trackable response channels mostly harvest demand that already exists; broad media, including offline TV, radio and outdoor, is what creates new demand in the first place. The Binet and Field analysis is a caution against judging everything by short-term, easily-attributed metrics.

How should a small business split its budget?

Start from the 60/40 brand-versus-activation heuristic, then adjust for your situation. If you have steady inbound demand, you can run hotter on activation; if nobody knows you exist, you need more reach. Owned (a useful website, email) and Earned (reviews, referrals) are the highest-trust, lowest-media-cost channels, so begin there before buying much paid media.

What's "through the line" (TTL)?

A campaign deliberately designed to do both jobs at once, a broad-reach ad that hands off to a trackable, response-driven channel. A radio spot driving people to a unique landing page, or a TV campaign with a QR code, is TTL. It's less a separate media type than an instruction to connect your reach and targeting so they reinforce each other.

How do media channels relate to the marketing mix?

Channels are the "Promotion" P doing its delivery work, but they only pay off if the rest of the mix is sound. The best channel plan can't sell a weak product, a wrong price, or a muddled position. See the marketing mix and marketing strategy & STP for where channel choice sits in the wider picture.

Related in the Toolkit

Where to go next