Brand or performance: where does the next marketing dollar go?

▶ Present these slides — opens the deck in your browser. Arrow keys advance; press S for speaker notes, F for fullscreen.

1. For the instructor

This is a ready-to-run 75-minute class you can teach cold, with no prep. Students meet an invented online brand whose founder wants to move the entire marketing budget into the channel that “returns four to one” on the dashboard. The catch they will discover: the number that looks like proof is partly borrowing its glory from spending the founder wants to cut — and an all-in bet on it quietly drains the demand it depends on. By the end, a table of students can look at a budget and say — in plain words, with arithmetic they can do in their heads — how much should go to selling now versus building demand for later, and why.

What’s in the pack: this run-of-show (section 4), the teaching points you need (section 5), a group exercise (section 6), a full facilitator answer key (section 7), debrief prompts (section 8), a printable student handout (section 9), a stretch task (section 10), and a slide outline you can present straight from (section 11).

Running it in a mixed room: every term is defined the first time it appears, and the only maths required is taking a percentage of a round number (e.g. 85% of $200,000). Put students in pairs or table groups of three to four so the confident ones explain to the rest — that talking-it-out is the learning. There is a low floor (every table can produce a split and one sentence defending it) and an optional ceiling (section 10) for fast tables.

One honest line to say out loud at the start: this is teaching material, not a certification and not a credit-bearing course. It is a session to practise a way of thinking, nothing more.

Further reading for you or keen students: the self-paced “Decide” course this builds on, brand-vs-performance-spend (“Brand or performance? Splitting the budget when the dashboard only sees one”).

A note on the clock. The run of show below is a tight 75 minutes with little slack — in a 30–40 student room budget ~85 minutes in practice (set-up, table-forming, and report-backs always run long). If you fall behind, cap report-backs at two tables and cover two of the discussion prompts. Protect the group exercise and the answer-key debrief — those are the session.

2. Session at a glance


Audience	Mixed-ability undergrads incl. non-business majors; basic arithmetic only
Class size	12-40
Total time	75 minutes
Materials	Printed handout (section 9); whiteboard or slides (section 11); no computers needed

By the end students can:

Tell whether a piece of marketing spend is harvesting demand that already exists (performance) or creating demand for later (brand) — and say why the two can’t be judged by the same number.
Explain why a strong performance return (ROAS) can be true and still not prove that brand spend is waste.
Set and defend a brand-versus-performance split for a budget, holding a floor of demand creation instead of chasing the dashboard to zero.

Further reading (a self-paced version of this idea): the Decide course brand-vs-performance-spend.

3. The case

Perch is a composite (invented) online brand that sells reusable water bottles and travel mugs directly through its own website. All figures below are illustrative — chosen for clean teaching, not measured from any real company.

Perch has a $200,000 marketing budget for the coming quarter. Today that money is split like this:

What Perch spends on	Split today	Dollars (illustrative)
Performance ads — the retargeting ad, the discount code, the “buy now” ad shown to people already searching	~85%	$170,000
Brand — a short film and sponsored posts that introduce Perch to people who have never heard of it	~15%	$30,000

The dashboard reports two numbers loudly:

Performance ROAS: 4.0x. For every $1 of performance spend, the dashboard traces $4 of sales.
Brand: ~0 traceable orders. The dashboard can’t tie a single order to the brand spend.

Two quieter numbers sit underneath: new-customer orders have been flat for three months, and branded searches (people typing “Perch” into a search bar) are down about 10%.

The founder has made a call and wants it done this week: “Brand is unmeasurable and performance returns four to one. Move the whole $200,000 into performance.” The room’s job is to decide whether that is the right call — and if not, what split is.

4. Run of show

Durations are written as **N min** and sum to 75 minutes.

8 min — Hook: read the founder’s order aloud (“move all $200k into performance”). Ask for a show of hands: who agrees? Note the split before anyone has heard the concepts.
10 min — Teach the two jobs of a marketing dollar (harvesting vs creating demand). Section 5, first heading. Keep it concrete with the October-film / December-purchase story.
12 min — Teach why the 4.0x ROAS flatters itself, and walk the five-step failure loop on the board. Section 5, second heading. This is the heart of the session.
8 min — Teach the 60/40 convention and the idea of a brand floor: tilt the dial, don’t remove it. Section 5, third heading.
20 min — Group exercise: tables set Perch’s split and write one sentence defending it. Section 6. Circulate with the answer key (section 7).
10 min — Debrief: two or three tables report their split; run the discussion prompts. Section 8. Reveal and weigh the options.
7 min — Wrap: state the one-sentence takeaway, point fast finishers at the stretch (section 10), and hand back the handout for keeping.

5. Teaching points

Teach only these three ideas — they are all the group exercise needs.

The two jobs of a marketing dollar

Every marketing dollar does one of two jobs.

Performance marketing (also called direct response) tries to turn demand that already exists into a sale now. It is the retargeting ad, the discount code, the “buy” button shown to someone who just searched your name. You pay for a measurable action — a click or an order — so it is easy to trace. Call this harvesting demand.
Brand marketing tries to create demand for later. It makes more people know, remember, and feel something about you, so months from now they choose you without being chased. Call this creating demand.

A tiny story that shows the difference: someone sees a warm Perch film in October and does nothing. In December they want a gift, remember the name, search “Perch”, and buy through a performance ad. The dashboard credits the December ad. But the October film planted the memory that made the click happen — and no dashboard shows that.

The one-line test for any spend: is it harvesting demand that already exists, or creating demand that does not yet?

Why the 4.0x ROAS flatters itself

ROAS means return on ad spend — sales the dashboard traces, divided by what you spent. Perch’s performance ROAS is 4.0x. Three things make “so move everything into performance” look right and be wrong.

Performance mostly harvests demand something else created. The people who click the “buy” ad often already knew Perch — they searched the name, they saw the October film. Many would have bought anyway. So a high ROAS partly measures how much demand the brand spend created, not how good performance is on its own.
Hard-to-measure is not the same as worthless. Brand shows ~0 traced orders because tracing tools follow clicks, and brand works by memory, not clicks. Reading the zero as “brand does nothing” confuses unmeasured with absent.
Piling more into one channel gets less back each time (this is called diminishing returns). The first performance dollars catch the easiest buyers cheaply; each extra dollar chases a colder, costlier one, so the return on the last dollar is well below the 4.0x average.

Put together, here is the failure loop to draw on the board:

Dashboard shows a great performance ROAS and a blank brand row.
Budget follows the number it can see, so brand is cut toward zero.
For a quarter or two, performance keeps posting a good ROAS — it is harvesting demand brand already created — so the call looks smart.
With nothing creating new demand, the pool of people who know and want Perch stops growing, then shrinks. New-customer orders flatten and fall. (Perch is already seeing this: flat new customers, branded search down 10%.)
Now even performance struggles — fewer warm people to harvest — so it costs more to get each new customer. The stall shows up a quarter after the cause, so it gets blamed on the ad platform, not the budget cut that caused it.

That is the efficiency trap: a budget that is efficient this quarter by starving the thing that makes next quarter possible.

The 60/40 convention and the brand floor

Is there a defensible split? The most-cited reference comes from an advertising-effectiveness analysis (Binet & Field, The Long and the Short of It, IPA, 2013) of close to a thousand case studies: across many consumer brands, the mix with the best combined short- and long-term results put roughly 60% into brand building and 40% into performance/activation.

Treat 60/40 as a convention, not a law:

It is an average across many brands, not a target for one — a useful starting reference, not a rule.
It shifts with the situation — a very early brand still proving people will buy at all tilts toward performance; a growing brand with something worth remembering tilts toward brand.
Its real job is to challenge an extreme split. Perch at ~85/15 has clearly abandoned demand creation; 60/40 tells you the direction it is wrong in, not an exact number to copy.

The key idea for the exercise: the stage can move the dial, but it should not remove it. Even a performance-tilted brand keeps a floor of demand creation, because a company with no one entering the top of its funnel has capped its own ceiling.

6. Group exercise

The task: You advise Perch. The founder wants all $200,000 in performance. In your table, decide the brand-versus-performance split for the coming quarter and write one sentence defending it. Use the handout (section 9); no computers needed.

Work through these steps:

Step 1 (about 4 min). On the handout, fill in the dollars for the founder’s plan (100% performance) and for the split today (85% performance / 15% brand). Reminder: 85% of $200,000 is $170,000; 15% is $30,000.
Step 2 (about 6 min). List, in your own words, what Perch would gain and lose by going to 100% performance. Point to the two quiet numbers in the case (flat new customers; branded search down 10%) — what are they warning you about?
Step 3 (about 6 min). Choose your split. Write the two percentages and the two dollar amounts. You do not have to match 60/40 — but be ready to say why yours fits Perch.
Step 4 (about 4 min). Write the one-sentence defence, naming the single thing that would make you shift more toward brand (or more toward performance).

Every table must leave with a split and a sentence. There is no single “correct” percentage — the reasoning is what’s graded by the room.

7. Facilitator answer key

The defensible call: do not go to 100% performance. Rebalance away from the ~85/15 extreme and hold a real brand floor — a split somewhere in the region of 65% performance / 35% brand is well-defensible for Perch, and any answer that lifts brand meaningfully above 15% while keeping performance the larger share is on the right track. The reasoning to reward: Perch’s 4.0x ROAS is partly harvesting demand the brand spend created, so cutting brand to zero flatters this quarter and starves the next — which is exactly what the two quiet numbers (flat new-customer orders, branded search down 10%) are already warning about.

Weigh the options out loud:

All $200k into performance (the founder’s plan) costs Perch its demand creation entirely; the ROAS may hold for a quarter while the funnel empties, then new-customer cost climbs and growth stalls — the efficiency trap.
Rigid 60/40 (60% brand) applied as a rule over-corrects: it treats a databank average as a law and could push a still-scaling DTC brand further from performance than its situation warrants.
A flat 50/50 is arbitrary — it hedges instead of reasoning from Perch’s actual signals (an extreme starting split and a funnel already showing strain).

Common wrong turns and how to steer. A table that sides with the founder is usually trusting the one number they can see — ask them what the 4.0x would look like if branded search keeps falling. A table paralysed by “there’s no right answer” should be told the exact percentage matters far less than holding a floor above zero and being able to defend the direction. A table that jumps to 60/40 by reflex should be asked why that number, for this company — and reminded it’s a starting reference, not a rule.

8. Discussion & debrief

Run three or four of these after tables report their splits:

Perch’s brand row shows ~0 traced orders. Does “we can’t measure it” mean “it isn’t working”? What could you look at instead of traced orders to tell if brand is doing its job?
The performance ROAS is 4.0x today. Whose work is that number partly measuring — and what would happen to it if the brand spend went to zero for two quarters?
The founder only trusts what the dashboard traces. In one plain sentence, how would you explain to them why a blank brand row is not proof of waste?
What single change to Perch’s situation (say, running out of cash, or a rival launching) would make you tilt more toward performance? What would make you tilt more toward brand?
Is there ever a case for zero brand spend? What does a company with no one entering the top of its funnel eventually run into?

One-sentence takeaway: A great performance number tells you what you can measure, not what works — so fund selling-now without starving the demand-creation that makes next quarter possible.

9. Student handout

Perch — where does the next marketing dollar go?

Perch is an invented online brand selling reusable water bottles and travel mugs. All figures are illustrative. Perch has $200,000 to spend on marketing next quarter.

Two jobs a marketing dollar can do:

Performance = turn demand that already exists into a sale now (traceable clicks and orders).
Brand = create demand for later (people who will remember and choose you). Hard to trace.

What the dashboard shows Perch:

Split today: ~85% performance ($170,000) / ~15% brand ($30,000)
Performance ROAS = 4.0x (each $1 traced to $4 of sales)
Brand: ~0 traceable orders
Quiet numbers: new-customer orders flat for 3 months; branded searches down ~10%

The founder’s order: “Move the whole $200,000 into performance.”

Your task — fill this in:

Dollars for the founder’s plan (100% performance): brand $______ / performance $______
Dollars for today’s split (85/15): brand $______ / performance $______
What Perch gains by going all-performance: ____________________________________
What Perch loses by going all-performance: ____________________________________
Our split: __% performance ($) / __% brand ($)
One-sentence defence (and the single thing that would make us shift):

Reminder: 10% of $200,000 = $20,000. To take any percentage of $200,000, multiply $2,000 by the number of percent (e.g. 35% → 35 × $2,000 = $70,000).

10. Stretch

For tables that finish early:

Perch’s brand row reads ~0 orders because tracing follows clicks. Name two things you could measure instead to tell whether the brand floor is working — and say which you’d trust more and why. (Hint: think about branded search volume, direct website visits, or asking a sample of people whether they’ve heard of Perch.)
Suppose a rival launches next month and drives the cost of every performance click up sharply. Does that make Perch’s brand floor more or less important? Defend your answer in two sentences — this is the one that separates real understanding from pattern-matching.
The hard one: 60/40 is an average across many brands. Design the cheapest honest test Perch could run over one quarter to find its own right split instead of borrowing the average — what would you change, for whom, and what would you watch?

11. Slides

Present straight from these. Bold labels mark each slide.

Slide 1 — Brand or performance: where does the next dollar go?

Perch has $200,000 for next quarter.
The founder: “Move it all into performance — it returns four to one.”
Today’s question: is that right?
Presenter note: take a show of hands before teaching anything.

Slide 2 — Two jobs a marketing dollar can do

Performance = harvest demand that exists → sell now (traceable).
Brand = create demand for later (remembered, not clicked).
The test: harvesting, or creating?
Presenter note: tell the October-film / December-purchase story here.

Slide 3 — The dashboard’s story

Performance ROAS = 4.0x. Brand = ~0 traced orders.
Quiet numbers: new customers flat 3 months; branded search down ~10%.
The number that looks like proof is the one the founder wants to keep.

Slide 4 — Why the 4.0x flatters itself

Performance mostly harvests demand brand created.
Hard-to-measure ≠ worthless (brand works by memory, not clicks).
More into one channel returns less each time (diminishing returns).

Slide 5 — The efficiency trap (the failure loop)

Cut brand → ROAS holds a quarter → funnel empties → new customers fall → performance costs climb.
The stall shows up a quarter after the cause.
Efficient this quarter by starving next quarter.

Slide 6 — 60/40 is a reference, not a rule

A databank average (Binet & Field, IPA 2013): ~60% brand / 40% performance.
Shifts with the company’s stage.
Its job: challenge an extreme split like 85/15 — hold a brand floor above zero.

Slide 7 — Your call: set Perch’s split

Decide the split; write one sentence defending it.
No single right percentage — hold a floor, defend the direction.
Name the one thing that would make you shift.

Slide 8 — Debrief: the takeaway

A great performance number tells you what you can measure, not what works.
Fund selling-now without starving demand-creation.
Presenter note: land the one-sentence takeaway from section 8.

12. Sources & license

Full provenance is in SOURCES.md.

Honest data line: Perch is a composite — an invented brand built for clean classroom teaching. Every dollar figure, the ROAS, the split percentages, and the branded-search movement are illustrative numbers chosen to make the reasoning clear. Nothing here is drawn from, or is a claim about, any real company, brand, or campaign.

Classroom-use license & disclaimer: this Teach-hall material is offered for free classroom use under the license and notice maintained in company/legal/classroom-license.md — read it before you teach from this. In short: it is provided “as is,” with no warranty, and it is not a certification and not a credit-bearing course; completing it certifies nothing on its own, and it is for skill practice and general education, not professional advice.

Instructor teaching material, provided as-is. Not accredited, not a certification, and not affiliated with or endorsed by any university. Uses composite (invented) companies and illustrative figures.