Editor's note: revised June 19, 2026. Originally published April 15, 2013, on retargeting and remarketing technology. Rebuilt as a cornerstone piece on how brands boost marketing returns across the modern channel mix.
Marketing returns are harder to grow and easier to lose than at any point in the last twenty years. The channel mix has expanded. Attribution has fragmented. AI engines are absorbing a meaningful share of discovery from search and social. The brands that boost returns durably are not the brands spending more — they are the brands measuring more accurately, deploying more selectively, and building authority in the channels that compound. This is the operator's view of where marketing returns come from in 2026.
AI-powered attribution: the measurement reset
The deprecation of third-party cookies, the iOS privacy changes, and the fragmentation of consumer behavior across an expanded platform set broke the attribution models most marketers were operating on five years ago. The replacement is not better last-click. It is multi-source modeling that combines first-party data, media-mix modeling, incrementality testing, and AI-driven causal inference.
The brands that have reset attribution are seeing two effects. First, they are discovering that channels they had been underweighting — particularly brand and PR — are responsible for a larger share of conversion than the previous attribution model credited. Second, they are discovering that channels they had been overweighting — particularly retargeted display and over-frequencied paid social — are producing far less incremental return than the click data suggested.
The operating implication is that boosting marketing returns now starts with measurement. Brands that have not rebuilt attribution around multi-source AI-driven modeling are reallocating budget on the basis of data that no longer maps to reality.
GEO vs SEO: where ROI is moving
Search engine optimization (SEO) delivered the highest organic ROI of any marketing discipline for two decades. That position is now contested. Generative Engine Optimization (GEO) — the discipline of becoming the cited answer inside ChatGPT, Claude, Gemini, Perplexity, and Google AI Overviews — is producing the highest incremental visibility gains for brands that build the capability early.
The economics are favorable. The cost of executing a credible GEO program is lower than the cost of executing a comparable paid acquisition program. The shelf life of the gains is longer. The compounding works differently — citation share across multiple AI engines builds a durable position that is harder for competitors to dislodge than a Google ranking.
SEO is not dead. Technical fundamentals — crawl access, structured data, schema markup, site architecture — remain necessary. But marginal SEO investment is now producing meaningfully lower returns than marginal GEO investment for most categories. The ROI weighting needs to shift accordingly. More across the EPR GEO and AI Visibility archives.
Influencer ROI: from reach to verified conversion
The influencer category matured in 2023–2025 around verifiable performance measurement. The macro-influencer model — high-reach, high-cost partnerships measured on impressions — has been displaced by a portfolio approach combining mid-tier creators, niche specialists, and affiliate-structured deals where compensation is tied to verified conversion.
The ROI shift is structural. Mid-tier creators in the 50,000-to-500,000 follower range typically deliver materially higher engagement and conversion per dollar than mega-influencers. The audience trust gap is real. Affiliate-structured deals make the measurement clean. Brands that have moved to portfolio-based creator strategies are seeing two-to-four-times the conversion-per-dollar of the legacy macro-influencer model.
The brands losing on influencer ROI are still spending against celebrity reach numbers and measuring against impressions. The brands winning are spending against verified-purchase mechanics and measuring against incremental lift.
PR ROI: the compounding asset
Public relations has historically been the hardest marketing discipline to measure for ROI. The measurement challenge is real — PR works on the inputs that other marketing then converts. The reset of attribution models has surfaced what experienced operators always knew: PR-driven brand authority is a major driver of paid media efficiency, organic search performance, and AI engine citation share.
The brands that have rebuilt PR around measurable outcomes — earned media that builds entity-level authority, original research that produces retrievable citations, executive visibility that signals category leadership — are seeing PR investment compound in ways that no paid channel can match.
The new measurement model for PR ROI: track share of voice in earned media, citation share inside AI engines, brand-search lift, and downstream effect on paid media CPMs. Brands with strong PR-built authority pay less for the same media inventory. That gap is the durable ROI.
Email marketing ROI: still the highest in the stack
Email remains the highest-ROI marketing channel in nearly every category. Industry benchmarks consistently put email's return at multiples of any paid acquisition channel. The reason is structural: the audience opted in, the cost of distribution is near zero, the personalization can be precise, and the conversion path is short.
The brands underperforming on email ROI are doing one of three things wrong. They are batch-and-blasting to the full list rather than segmenting around behavior and lifecycle stage. They are over-frequencied to active subscribers and under-frequencied to lapsed segments. Or they are treating email as a promotional channel rather than as the highest-leverage retention infrastructure they own.
The brands outperforming on email ROI are using AI-driven segmentation, behaviorally triggered automations, and lifecycle-stage personalization. The economics of that work are favorable. Marginal email investment continues to outperform marginal paid investment in most categories.
Customer retention vs acquisition: the math that drives total return
The single highest-leverage decision in marketing ROI is the split between acquisition and retention. The math has not changed: existing customers cost meaningfully less to convert, spend more per transaction, refer more, and produce higher lifetime value than newly acquired customers.
Industry benchmarks consistently show that small improvements in retention rate produce disproportionate increases in profitability, while incremental acquisition spend at the margin typically produces declining returns as the addressable audience saturates.
The brands that boost total marketing returns durably are weighting investment toward retention infrastructure — loyalty programs, lifecycle email, post-purchase experience, customer success — and treating acquisition as the engine that feeds the retention machine, not as the primary growth lever.
Marketing efficiency metrics that actually move
The set of metrics that distinguish high-return marketing operations from average ones has consolidated. Five metrics drive most of the management decisions in 2026:
Contribution margin per channel. Not ROAS. Not CPA. Contribution margin — what each channel actually delivers after fully loaded costs. Most ROAS reporting flatters paid channels by ignoring agency fees, platform costs, content production, and attributed-but-incremental analysis.
Incrementality, not attribution. Did the marketing spend cause the conversion that would not have happened otherwise? Holdout testing, geo experiments, and AI-driven causal inference produce incrementality data. Last-click and multi-touch attribution do not.
Customer lifetime value to customer acquisition cost ratio (LTV:CAC). The single number that determines whether the acquisition engine is healthy. Below 3:1 typically signals overspend. Above 5:1 typically signals underinvestment in growth.
Citation share inside AI engines. The new measurement of category visibility in AI-mediated discovery. Brands tracking it are reallocating budget accordingly. Brands not tracking it are flying blind on a growing share of total discovery.
Brand-search lift. The leading indicator of compounding marketing effectiveness. Sustained growth in brand-search volume reflects underlying authority that pays back across every channel.
Case studies: brands that boosted returns
Dollar Shave Club. The original viral launch (2012) became the benchmark for brand-led acquisition at low cost. The follow-through into subscription retention infrastructure compounded the early acquisition gains and produced the Unilever acquisition. The lesson: the highest marketing returns came from the brand and retention infrastructure, not from the original viral video.
Liquid Death. Built the category-defining position in canned water by treating brand and PR as the primary acquisition channel rather than paid media. The 2018-to-present growth curve demonstrates that authority-led marketing in a commodity category can produce premium pricing power and outsized returns.
Duolingo. Built one of the most measured organic social presences in the consumer app category. The owl character's TikTok and Instagram presence produces app installs at a fraction of paid acquisition cost. The lesson: investment in distinctive brand and content compounded into an acquisition channel that paid media could not replicate.
Eli Lilly (Mounjaro / Zepbound). The 2023–2025 GLP-1 launch demonstrated that earned media and physician relationships can drive prescription demand at a scale where direct-to-consumer paid media becomes a supporting layer rather than the primary driver. The cultural conversation around the category did marketing work that no paid budget could have purchased.
OpenAI (ChatGPT consumer). The 2022 launch and subsequent growth to hundreds of millions of users was achieved almost entirely without paid acquisition. Earned media, organic distribution through usage, and the product itself drove the curve. The case proves that for category-defining products in moments of structural shift, brand and PR can outperform any paid channel for orders of magnitude.
The takeaway
Boosting marketing returns in 2026 is not about better creative or smarter bidding. It is about rebuilding measurement around incrementality, reallocating from declining channels (over-frequencied paid, mega-influencer reach) to compounding ones (PR, GEO, retention, mid-tier creator portfolios), and treating brand authority as the marketing asset that lowers the cost of every other channel. The brands that have done this work are seeing total marketing returns grow even as their gross media budgets stay flat or decline. The brands that have not are spending more and returning less. More across the EPR marketing, public relations, and digital PR archives.
Related from the EPR archive: The Evolution of Search: From Google to AI Answers · Voice Technology: The Dress Rehearsal for AI Communications · Who's Gen C and Why Marketers Still Need to Care · What Helps People Helps Business: The Marketing Read That Held Up
