Performance Brand-Building Case Studies
The industry thinks what we’re doing is wrong. These six case studies prove it’s not — and that no one else is getting results like this because no one else is doing this.
Performance Brand-Building Drives Better Results in the Long Term AND the Short Term
Contribution margin (total revenue − product costs − ad spend) measures the dollars your business actually keeps. It’s the clearest measure of whether growth is profitable — and it’s how we hold ourselves accountable.
We're Operators, Not Theorists
- Losing money for six months while waiting for brand to "kick in" is not a luxury any business owner has
- In every case where we've taken over, we've seen immediate improvement — not loss
- We start brand at ~10% of spend and expand as results confirm
Fix DR First, Then Layer Brand
- Right-size retargeting and reduce over-serving to existing customers
- Orient spend toward incremental new customer growth
- Contribution dollars and margin growth have improved within the first month across our clients
Brand IS Better Performance
- These results come from proprietary performance brand-building campaigns — not traditional direct response
- Brand campaigns drive stronger incremental ROAS than anything on the DR side
- Marathon's proprietary tools measure brand campaign impact daily — not quarterly. Every campaign is optimized in real time, which is why the results are both large and repeatable
- The data below proves it across six brands, in both short and long term
Results at a Glance — by Brand
Every result below was driven by top-of-funnel brand-building campaigns — not traditional direct response. This is the performance brand-building playbook replicated across multiple brands and categories.
Short-Term Results
In six weeks: contribution margin grew from ~$140K/week to $209K/week. YoY comp accelerated from +22% to +68%. New customer revenue flipped from six consecutive weeks negative to positive by week four.
Pre-campaign avg 497/day → week of Mar 9 avg 883/day. Steady growth every month since campaign start. YoY reversed from -50% to +70%. Follower growth 341x.
~6K→~14K organic branded search impressions per day in under two months. CM +37% YoY. Revenue +20% YoY.
Long-Term Incrementality Test Results
6-month causal holdout testing strictly brand advertising and brand creative. Platform showed 0.91x blended ROAS for DR during the same period. Revenue lift over 2x higher than all of direct response. Brick-and-mortar lift 2x larger than e-commerce.
Causal holdout at major retailer proves brand building materially drives wholesale. 2.5x incremental ROAS in brick and mortar on brand media alone.
97% confidence across Shopify, Amazon, and TikTok Shop. 72% higher incremental return than Meta DR test, 47% higher than CTV — and this was brand media. Brand searches +129% YoY.
What We Heard From the Industry — and What the Data Actually Shows
What the Industry Says
- Brand investment means adding budget on top of what you’re already spending
- There will be months of losses before any payoff materializes — it’s a long-term play
- Performance media means direct response — top-of-funnel doesn’t drive measurable returns
- No one has proven this works at scale — results like these are a one-off that cannot be replicated
What the Data Actually Shows
- Top-of-funnel brand campaigns drive stronger incremental ROAS than direct response — proven across six brands by rigorous causal holdout tests
- Contribution margin improves in weeks because the least efficient DR dollars are reallocated first — no budget added
- Brand searches, new customer acquisition, and community growth respond in days and weeks, not months
- The results have been replicated across multiple brands, categories, and channels — it is not a one-off
Every incrementality result below was measured by direct response standards — the hardest standard there is. The brand perspective upside is additive: while driving incremental revenue, these campaigns simultaneously build audience, brand searches, and baseline. A 2x return on brand spend is fundamentally more valuable than a 2x return on DR spend — the brand version compounds. No other agency believes this. No other agency can prove it.
The Evidence, Brand by Brand
Every metric is sourced. Every claim is verified. See the Data Sources appendix below for full bibliographic provenance.
The Starting Point
Up and down week to week — inconsistent YoY comps with no clear growth trajectory
Significant negative comps on new customer acquisition — the engine of future growth was stalled
A lot of wasted ad spend — heavy investment across Google brand keywords, CTV, podcast, and Reddit with diminishing returns
No contribution margin visibility, no incrementality measurement, budget decisions made on platform ROAS
“It’s not formulaic. It’s a gut decision.” — VP Marketing, December 2025
Smart Reallocation, Not Added Budget
Increasing Reach to New Audiences
Right-sized retargeting and retention spend (diminishing returns) and reallocated into campaigns designed to reach new people who don't know the brand yet. More top-of-funnel reach, less recycling the same audience.
Brand-Centric Creative Across the Whole Funnel
Creative that reinforces who the brand is at every stage — not generic direct response ads that could have come from any brand.
Systematic Brand Campaigns, Measured by Business Results
Launched systematic brand campaigns held accountable by contribution margin, brand searches, and new customer revenue — not in-platform ROAS. Weekly measurement from day one, with clear thresholds for scaling.
Same budget, better allocation. The top-of-funnel brand-building campaigns replaced the least efficient DR dollars — not supplemented them. And because these are performance brand-building campaigns, the returns don’t stop when the ad does. They compound through brand searches and baseline revenue that no direct response campaign can build.
$209K Weekly CM — Biggest Positive Comp on Record
From ~$140K to $209K in weekly contribution dollars — adding ~$70K/week. +68% YoY, the biggest positive comp on record.
From 18% (prior year) to 25.92% — the highest CM% for any week on record. CM% YoY growth went from roughly flat in January to +44% in March.
6 straight weeks negative YoY (Jan 19 – Feb 23). Flipped positive Mar 02 (+11% YoY, $226.53K) and Mar 09 (+4.7%, $220.34K) — both above $220K for the first time since December.
Every single week of DTC e-commerce revenue has been up YoY since Marathon took over. Last week was the biggest positive comp of the year.
$111K → $143K → $143K → $171K → $185K → $203K → $209K. No meaningful week-over-week decline across the entire engagement period.
Monthly average CM$ YoY growth accelerated from ~15% in January to ~65% in March — a 4x acceleration in growth rate
No trough. No revenue drop. These results came from top-of-funnel brand-building campaigns — not more direct response spend. From the first week, the negative new customer revenue comp started closing. By week four it was positive. Contribution margin has grown every single week — because the dollars reallocated into brand campaigns were already the least efficient in the account. Enabled by daily measurement and optimization — not a launch-and-hope approach. Every other agency said this doesn’t work. This is what actually happened.
Brand Search Up 78% Since January
Pre-campaign baseline (Jan 1–21): 497/day. Week of Mar 9: 883/day (+78%). Steady upward trend every week since campaign start. YoY reversed from -50% to +70%.
From ~2 new followers/day in January to 683/day by March
22 shares/day → 161 shares/day in two months
For the first time this year, new customer acquisition revenue is trending up — driven by brand search and direct traffic, not paid click attribution
Brand search is the leading indicator of a sustainable growth engine. These results came from top-of-funnel performance brand-building campaigns — the same proprietary playbook driving results across every brand in this deck. The brand was trending -50% YoY on branded search when we started. Ten weeks of brand-building campaigns later: +70%. Enabled by daily measurement and optimization — not a launch-and-hope approach. No other agency is running this playbook. No other agency is generating results like this. DR spend buys a click that disappears tomorrow. Performance brand-building builds compounding search volume that keeps lowering acquisition cost long after the campaign ends.
Brand Search More Than Doubled Since Mid-January
Organic branded search impressions grew from ~6K/day (mid-January) to ~14K/day (mid-March) — a 125% increase in under two months
Contribution margin dollars growing at +37% year-over-year while the brand campaigns scale
$1.49M weekly revenue, trending up every week since engagement start
This engagement started in mid-January 2026 — less than two months ago. These results came entirely from top-of-funnel brand-building campaigns. Every other agency said this takes months to show up. Organic branded search has more than doubled in under two months. Contribution margin is up 37% YoY. Enabled by daily measurement and optimization — not a launch-and-hope approach. These are current results — not projections. Unlike direct response, these brand search gains persist and compound. Every search is a customer who finds you without being targeted. This is the performance brand-building playbook. These numbers are why no one else is getting results like this.
$720K Incremental Revenue — Proven by Causal Holdout
Every $1 of brand spend returned $1.93 in proven incremental revenue. 90% CI: 1.45x – 2.41x — entire range above breakeven.
Brand incremental ROAS was 2.1x higher than the blended in-platform DR ROAS (0.91x) during the same period. Brand outperformed DR by every measure.
Brand is profitable in e-commerce, but twice as profitable in building wholesale. Incremental retail lift was 2x larger than DTC — revenue you cannot reach with direct response ads.
January 2026 — the full business grew nearly 30% year-over-year while brand campaigns scaled
This is a 6-month causal geo-holdout — the most rigorous test available. These were exclusively top-of-funnel brand-building campaigns. The platforms said 0.91x. The real return was 1.93x. And here’s what makes this fundamentally different from DR: while generating that 1.93x incremental return, these brand campaigns were also building brand searches, growing the audience, and lifting retail sales in channels with zero ad targeting. No direct response campaign does that. This is the performance brand-building playbook — and it outperformed everything on the DR side by every measure. This result alone would end the argument. But we have five more case studies showing the same thing.
The Performance Brand-Building Playbook Replicates Across Categories and Channels
Both results below came from top-of-funnel brand-building campaigns only — no DR. Both outperformed direct response by every measured standard. Two more data points proving this is not a one-off. More case studies are being added continuously.
Geo-holdout test measuring brand campaign impact on wholesale sell-through at a major outdoor retailer, excluding all DTC channels. Brand building materially drives wholesale — revenue you cannot reach with direct response.
6-month geo-holdout across Shopify, Amazon, and TikTok Shop. 97% statistical confidence that spend is profitable.
Marathon’s brand holdout generated a higher incremental return than every DR tactic they ran incrementality studies against. Four independent tests — Marathon holdout, Meta test, two CTV lift studies. Brand outperformed DR across the board.
Maximizing Impact in the First 60 Days
- Pixel & data audit — validate billing, tracking, and everything passing to ad platforms. No conflicts or under-firing.
- Validate and define all audience segments optimized for incremental new customer growth
- Campaign structure custom-tailored to your business category, stage, and channel mix
- Top & upper-mid funnel campaigns live across Meta, YouTube, and TikTok
- 30-day model trainer with holdout data completes the tuning of the Marathon brand model with causal data
- Brand-building creative across the whole funnel — memory structures and creative diversity from day one
- Marathon model goes live — fully populated with causal and statistical data, retraining weekly, updated daily
- Brand budgets start at ~10% and scale as they earn more budget — backed by the data to support every dollar
- Creative feedback loop on what drives incrementality and brand value at every funnel stage
- Scale what works — validated by volume growth, organic sessions, and behavioral confirmation
- Creative iteration with true diversity — more shots on goal, not just DR variations
- Larger funnel ready to capture demand — more people who already know and trust the brand
By day 60, you’re less dependent on the auction — because you’ve built a fundamentally larger body of people who proactively seek you out before they ever see a direct response ad. And the data is clear: this isn’t a sacrifice. Across every client, contribution margin improves in weeks, not months. The full funnel performs better than bottom-funnel spend alone — immediately.
The Full-Funnel Playbook
Optimize for likes, comments, follows, shares, and saves — the ideal first touchpoint with people not actively in market. They can't perform a shopping action yet, but they can opt in to the brand.
- Net new reach we weren't getting otherwise — lowering cost per thousand accounts reached
- Growing pool of brand-opted-in users dramatically more likely to take the next step
Optimize for a custom event: a session from organic search. Measured with incremental attribution to validate these dollars drive genuine branded search volume growth.
- Step-change growth in branded organic search — people seeking out your brand without clicking a DR ad
- Smooths the path from top-of-funnel brand action to active purchase consideration
Stronger brand creative at the bottom drives more conversion lift and incrementality — even if it doesn't show up in platform ROAS. Lower CPMs, higher dwell time, full-price brand buyers.
- Higher-margin conversions from buyers who purchase because of the brand, not a promotion
- Less dependence on the ever-increasingly expensive auction dynamic
This is better performance — not a tradeoff. The measured incrementality on brand campaigns consistently exceeds DR in every account we manage. No other agency is running this playbook. No other agency has the data to back it up. Creative diversity gives us auction advantages at every stage, brings in full-price brand buyers instead of discount shoppers, and builds a compounding demand engine that no direct response campaign can replicate. Your business performs better — today and long-term — when this proprietary approach is working.
What This Builds
What We Track Weekly
We are triangulators. That’s our job.
- Brand Value & Incrementality — iROAS on new audiences, conversion lift, holdout vs. exposed
- Reach & New Audiences — spend to new audiences, cost per thousand accounts reached, reach growth
- Brand Searches — branded organic search volume, the leading indicator of compounding demand
- Engagements Across the Funnel — brand actions, community growth, engagement rate by funnel stage
The Creative Feedback Loop
Immediate signal on what drives both incrementality and brand value. More toward brand value at the top, more toward demand capture at the bottom — but we want both working at every stage.
The Compounding Effect
Every dollar of brand investment creates people who come to you without ever seeing or clicking a direct response ad. They search the brand. They memorize the URL. They come directly.
This stream of revenue grows and compounds over time — reducing dependence on the auction, increasing optionality, and growing owned audiences.
One client saw contribution dollars up 60% in 80 days. Another hit seven consecutive weeks of CM$ growth with no decline. This isn't a long-term bet — it pays out in weeks.
What Success Looks Like
A fundamentally larger funnel of people who already know and trust your brand — less dependence on the auction, more control over your own growth.
That means: More demand to capture, less cost to capture it, higher margins, and a growth engine that keeps compounding after the holiday.
The Playbook Works. The Proof Is in the Data.
Across six brands — denim, wellness, footwear, food & beverage, outdoor gear, beauty — the pattern is the same. Top-of-funnel performance brand-building campaigns drive stronger incremental ROAS than anything on the DR side. No short-term loss. No trough. Results in weeks. These results were not accidents — they were produced by daily optimization of brand campaigns using Marathon's proprietary measurement tools. The opposite of the traditional "launch and check back in six months" approach. More case studies are being added continuously — because the playbook keeps working.
Immediate Improvement
Every brand saw contribution margin, revenue, and new customer acquisition improve from the first weeks — driven by top-of-funnel brand campaigns, not more DR spend
No Short-Term Loss
Zero brands experienced a revenue drop or contribution margin decline. Not one. This is what no other agency believes — and what this data proves.
Brand Outperforms DR — Every Time
1.93x–2.39x incremental return on brand campaigns alone — higher than every DR tactic tested in the same accounts, proven by 6-month causal holdout tests at 90–97% confidence
Replicated Across Six Brands — More Coming
This is not a one-off. The performance brand-building playbook has been proven across multiple categories and channels. More case studies are being added continuously.
Data Sources
Every quantitative claim on this page has a source row below. Source badge legend: Verified = Marathon data warehouse or direct measurement; Platform = ad platform or search console; Internal = Marathon internal reporting.
| Claim | Source | Detail |
|---|---|---|
| Weekly CM$: ~$140K → $209K (+68% YoY) | Verified | Marathon KPI Deep Dive (Shopify Contribution Breakdown). Week of Mar 09, 2026: $208.94K CM vs. $124.06K prior year = +68.4% YoY. Prior year baseline weeks ranged $88K–$131K. $209K is the biggest positive CM$ comp on record. |
| CM%: 18% (LY) → 25.92% (highest on record, +44%) | Verified | Marathon KPI Deep Dive. Week of Mar 09, 2026: 25.92% CM% vs. 18% prior year = +44% YoY. Highest CM% for any week on record. CM% YoY growth went from roughly flat in January (negative first two weeks, then +5.1%, +9.7%) to +40–48% in March. |
| CM$ YoY growth: ~15% (Jan avg) → ~65% (Mar avg) | Verified | Marathon KPI Deep Dive. January weekly CM$ YoY growth averaged ~15% (+15%, +2.5%, +18.5%, +22%). March weekly CM$ YoY growth: +61.9% (Mar 02), +68.4% (Mar 09), averaging ~65%. Monthly averages show clean 4x acceleration with no dips. |
| New customer revenue: 6 weeks negative → positive Mar 02 | Verified | Marathon KPI Deep Dive (Ecomm New Customer Revenue). Jan 19 – Feb 23: six consecutive weeks negative YoY (−5.6%, −8.1%, −8.2%, −7.0%, −0.2%, −5.2%). Mar 02: +11.0% YoY ($226.53K). Mar 09: +4.7% YoY ($220.34K). Both March weeks above $220K — no week above $220K since December. |
| DTC revenue up every week since takeover | Verified | Marathon KPI Deep Dive (Shopify Customer Revenue Total). Every week from Jan 26 through Mar 09, 2026 shows positive YoY comp: +11.2%, +9.1%, +21.5%, +20.2%, +7.4%, +15.9%, +13.8%. Last week was biggest positive total revenue comp of the year. |
| CM$ has not declined WoW since takeover | Verified | Marathon KPI Deep Dive. Weekly CM$ sequence: $111K, $143K, $143K, $171K, $185K, $203K, $209K. No week-over-week decline across the entire engagement period. |
| Claim | Source | Detail |
|---|---|---|
| Organic branded search impressions: pre-campaign avg 497/day (Jan 1–21) → 883/day (week of Mar 9) (+78%) | Verified | Cube.js query, metrics.google_search_organic_branded_search_impressions. Pre-campaign baseline (Jan 1–21, 21 days): avg 497/day. Week of Mar 9 (6 days, Mar 9–14): avg 883/day. +78% growth. Daily: Mar 9: 806, Mar 10: 696, Mar 11: 808, Mar 12: 790, Mar 13: 634, Mar 14: 1561. Weekly trend: Dec 29 289 → Jan 5 475 → Jan 19 635 → Feb 2 703 → Mar 2 735 → Mar 9 883. Verified 2026-03-18. |
| YoY brand search: −50% → +70% | Platform | Google Search Console, same-period YoY comparison. January 2026 was −50% vs. January 2025. By early March 2026, trending +70% vs. same period 2025. |
| Follower growth 341x: ~2/day → 683/day | Platform | Meta Business Suite follower analytics. Jan 1–21, 2026: 44 total new followers (~2/day). Mar 1–10, 2026: 6,826 new followers (683/day). |
| Daily social shares +629%: 22 → 161/day | Platform | Meta content sharing metrics. January 2026 average: 22 shares/day. Mar 1–10, 2026: 161 shares/day. |
| New customer acquisition revenue now growing | Internal | Marathon internal reporting, Mar 2026. First positive new-customer revenue trend of the calendar year, driven by brand search and direct traffic growth. |
| Claim | Source | Detail |
|---|---|---|
| Organic branded search impressions 2.25x (6K/day → 14K/day) | Verified | Cube.js query, organic branded search impressions only. Measure: metrics.google_search_organic_branded_search_impressions. Jan 19-21 3-day avg: 6,145/day. Mar 12-14 3-day avg: 13,811/day. +125% growth (2.25x). Verified 2026-03-18. |
| Contribution Margin +37% YoY | Verified | Marathon Weekly Performance Report and Cube.js data warehouse. Week of Mar 6–12, 2026. CM calculated as Revenue − COGS − Ad Spend − Variable Costs, compared to same week prior year. |
| Revenue +20% YoY ($1.49M weekly) | Verified | Marathon daily briefing data, Mar 17, 2026. GA4 total revenue, 7-day trailing. $1.488M actual vs. prior year same period. Trending up every week since engagement start. |
| Claim | Source | Detail |
|---|---|---|
| 1.93x adjusted incremental ROAS | Verified | 6-month causal geo-holdout test, Jul 17, 2025 – Jan 13, 2026. Bayesian Synthetic Control methodology. Raw 1.59x adjusted to 1.93x after accounting for pre-test market differences. 90% CI: 1.45x – 2.41x. $454K total brand spend, $720K incremental revenue. |
| In-platform ROAS: 0.91x | Platform | Meta Ads Manager in-platform ROAS for same brand campaigns during test period. Meta's attribution model reported 0.91x; the causal holdout showed true return was 1.93x — a 2.1x gap between what the platform says and what actually happened. |
| Retailer revenue lift: 6.2% | Verified | Same geo-holdout test, retail channel (major national retailer) revenue comparison between test and control markets. 6.2% lift — nearly 2x the web DTC lift (3.3%). $141K incremental retail revenue. Brand campaigns had zero retail targeting. |
| Total business revenue +29% YoY (January) | Verified | Cube.js data warehouse, total revenue comparison Jan 2026 vs. Jan 2025. +28.7% YoY all-channel revenue growth. |
| Claim | Source | Detail |
|---|---|---|
| 2.5x causal revenue lift at major retailer | Verified | Geo-holdout test, Jan–Feb 2026 combined (30 days). Wholesale sell-through at a major outdoor retailer, excluding all DTC channels. Raw 1.76x, adjusted to 2.5x over 180-day window. Total spend: $26,200. 90% CI: 1.24x – 4.46x. |
| YouTube Brand ROAS: 12x | Internal | Marathon Brand Value model, week of Feb 23 – Mar 1, 2026. Early result expected to moderate as spend increases. |
| 12,600 new followers in 7 days, $0.17/follow | Platform | Meta + YouTube follower metrics, week of Feb 23 – Mar 1, 2026. 7,600 Facebook + 5,000 YouTube. $1,300 total spend. |
| Claim | Source | Detail |
|---|---|---|
| 2.39x adjusted incremental ROAS, 97% confidence | Verified | Marathon geo-holdout test through Dec 31, 2025. Bayesian Synthetic Control, 50/50 geo split. $172,619 total spend, $260,756 revenue lift. P(ROAS > 1.0) = 97%. |
| 4 independent tests confirm incrementality | Verified | (1) Marathon geo-holdout: 2.39x. (2) HAUS Meta test: 1.39x. (3) Vibe CTV: 1.63x, 95% confidence. (4) Universal Ads CTV: 0.86x, recommendation DEFUND. Three of four confirm positive return. |
| Brand searches +129% YoY | Platform | Google Search Console branded query volume, as of Oct 2025 vs. Oct 2024. |