Dynamic Creative Optimisation: 3x ROI vs the 67% Failure Trap

Eighty per cent of consumers worldwide say they are comfortable with personalised experiences. Sixty-seven per cent of them have walked away from a brand because of a personalised experience that went wrong. Both figures come from Boston Consulting Group research, and both define the situation senior marketers face when they sit down to evaluate dynamic creative optimisation.

Personalised offers generate three times the ROI of mass promotions, also per BCG. Retailers commit less than five per cent of their promotional budgets to personalisation. Ninety-nine per cent of agencies told Digiday and Clinch that DCO has become a significant factor in their work. The gap between widespread adoption and successful execution is the story this article tells.

Here’s what actually works — and where most DCO implementations break down. I’ve run enough personalisation campaigns across enough categories to know that the technology rarely fails on its own. The failure points are organisational, creative, and measurement-related. Solving them requires a clear-eyed view of what DCO can and cannot do under current privacy constraints, and a campaign playbook that respects both the technology’s potential and its real limitations.

What This Article Delivers

The Business Case: 3x ROI Against a 5% Budget Reality

The 3x Return Most Marketers Are Underfunding

The results speak for themselves: personalised offers return three times the ROI of mass promotions, according to Boston Consulting Group research. That figure shows up consistently across retail implementations, not as an outlier in a single case study. McKinsey research cited by Acxiom puts the marketing ROI lift from personalisation at 10 to 30 per cent. Personalisation leaders in retail achieve revenue growth ten percentage points higher than companies that lag, based on BCG’s Personalization Index.

Now look at where the money actually goes. Retailers allocate less than five per cent of promotional budgets to personalisation on average. The gap between proven performance and investment behaviour is one of the largest you’ll find in modern marketing. Companies acknowledge personalisation works. They keep funding mass promotions anyway.

The reason isn’t ignorance of the data. It’s the implementation challenge. Building DCO capability requires creative volume, workflow redesign, and measurement discipline that most marketing organisations haven’t yet developed. The 3x return is real. So is the difficulty of capturing it. The question every marketing leader needs to answer is whether their organisation has the operational capacity to convert the opportunity into actual returns.

The $570 Billion Window in First-Party Data

The numbers at the top of the BCG Personalization Index get bigger. Top retailers can capture $570 billion in incremental growth by 2030 through first-party data utilisation. Individual large retailers can generate over $100 million in topline impact from personalised offers at scale. These aren’t projections from a vendor pitch deck — they’re the addressable opportunity for organisations that solve the implementation problem.

The timing matters. Retail media advertising is growing at 25 per cent annually, fuelled by first-party data capabilities. As privacy regulations restrict third-party data, brands with robust first-party data assets and the technical capability to activate them gain disproportionate advantage. Commerce media is already extending past traditional retailers — Chase and American Airlines demonstrated at Cannes Lions 2024 how brands with customer journey data can build advertising businesses outside conventional e-commerce.

Industry analysts describe this as the third great wave of advertising, after the rise of digital and the rise of programmatic. First-party data combined with DCO lets brands deliver relevance without depending on the cookie-based tracking that powered earlier generations of personalisation. The $570 billion belongs to organisations that build that capability before competitors do. Everyone else watches from a shrinking position.

Eighty Per Cent Comfortable, Sixty-Seven Per Cent Burned

Four-fifths of surveyed consumers worldwide are comfortable with personalised experiences, per BCG. That establishes personalisation as a baseline expectation, not a novel feature. The top three benefits consumers want from personalisation are value (best price and offers), enjoyment (better experience), and convenience.

Generational comfort is even higher. Eighty-one per cent of Gen Z prefer personalised ads, compared with 57 per cent of millennials, based on research cited by AppsFlyer. As digital-native consumers represent larger market segments, comfort with personalisation will keep climbing. The demand side of this market is already in place.

The execution side is where it falls apart. Whilst 80 per cent express comfort with personalisation in principle, 67 per cent have had at least one negative personalised experience that caused them to break off interaction with a brand entirely. Read that again. Poor DCO implementation doesn’t simply underperform — it actively damages customer relationships. The high baseline expectation and low tolerance for poor execution creates a narrow window. Either you deliver genuinely relevant, well-timed creative that respects user intent, or you push customers away. There’s not much middle ground left.

Inside the DCO Engine: A Four-Stage Operational Framework

DCO operates through a structured workflow that turns static creative production into automated, data-led personalisation. Understanding this framework clarifies both the capability requirements and the points where implementations break down most often.

Stage One: Modular Creative as Inputs, Not Outputs

DCO begins with creative assets prepared as modular components rather than finished advertisements. Instead of producing complete ads for each audience segment, creative teams build a library of interchangeable elements: headlines, images, calls-to-action, product shots, background designs, and copy variations. These components are designed to combine in many configurations whilst holding brand standards.

The template architecture defines how the elements can combine. A single template might accommodate five headline options, eight product images, four background colours, and three calls-to-action — generating hundreds of potential variations from a manageable component set. The discipline at this stage is making sure every possible combination still maintains brand standards and delivers a coherent message.

This represents a real shift in creative production methodology. Traditional campaigns produce finished assets that stay static after launch. DCO requires creative inputs that function as ingredients rather than finished dishes. For teams used to perfecting individual executions, this feels uncomfortable at first. It’s also non-negotiable if you want personalisation at scale.

Stage Two: The Data Management Platform Layer

All DCO data sources are managed through a centralised data management platform, according to AppsFlyer. The DMP is the integration layer that connects audience data from multiple sources: demographics, browsing history, location, device type, weather conditions, and CRM records. The platform normalises these inputs into a format the DCO system can process.

The data integration stage determines how sophisticated personalisation can actually get. Basic implementations might personalise on browsing history and location. Advanced applications incorporate real-time behavioural signals, predictive analytics, and cross-channel interaction data. Acxiom notes that DCO can use real-time data like weather conditions to automatically adjust creative — a swimwear ad in a rainy market gets replaced before it serves.

Privacy regulations fundamentally constrain DCO’s ability to use the granular user data that powers personalisation. Traditional DCO has been dramatically undercut by privacy regulations and technological changes that limit user tracking and data availability, making many promised DCO capabilities undeliverable. This technical reality means DCO promises often exceed what can be delivered legally and technically. Marketers need to reset expectations about personalisation scope before they evaluate platforms or build campaigns.

Stage Three: Real-Time Assembly and Machine Learning

DCO uses machine learning to orchestrate real-time optimisations to dynamic creative within campaigns, as Acxiom describes it. When a user triggers an ad impression, the system evaluates available data about that user against predefined rules and machine learning models. Within milliseconds, it picks the optimal combination of creative elements from the component library.

The optimisation logic evolves continuously. Early in a campaign, the system tests various combinations to establish performance baselines. As data accumulates, machine learning algorithms identify patterns: which headline performs best with which product image for which audience segment. The system shifts impression volume to higher-performing combinations whilst continuing to test new variations.

Automated assembly enables personalisation at scale with less manual burden on advertising teams, again per Acxiom. But most DCO platforms restrict marketers to predefined personalisation parameters and manual workflows, inhibiting rather than enabling the automated, real-time creative adaptation DCO promises. The gap between theoretical promise and practical reality means most implementations deliver limited value. Picking the right platform isn’t a technical preference — it’s the difference between a system that runs itself and one that demands constant intervention.

Stage Four: Millisecond Delivery and Element-Level Measurement

DCO originated in the early 2010s, and by 2024 platforms can process multiple data points simultaneously with real-time decisions in milliseconds, according to DevriX. That processing speed is what makes true real-time personalisation possible — creative that adapts to user context at the moment of impression, not days or weeks later.

When set up correctly, DCO ads work automatically with distribution channels like Google Ads and Meta, as Storyteq notes. The system optimises ad budget spending by using real-time data to determine placement, per AppsFlyer. Creative optimisation and media buying optimisation operate as a unified system rather than two separate functions running on different timelines.

Performance measurement happens simultaneously with delivery. The system tracks campaign-level metrics, but also element-level engagement rates, combination effectiveness scores, creative fatigue measurements, and segment-specific conversion rates. This granular feedback enables continuous refinement and helps identify when specific variations lose effectiveness over time. Without this measurement layer, the rest of the system runs blind.

Why 67% of Personalised Experiences Fail

Privacy Regulation Has Hollowed Out Traditional DCO

Traditional DCO has been dramatically undercut by privacy regulations and technological changes that limit user tracking and data availability. The granular user tracking that powered first-generation DCO systems is no longer available in many markets and will become further constrained as third-party cookies deprecate completely.

This regulatory shift forces a fundamental architectural change. DCO systems must now operate primarily on first-party data, contextual signals, and consented user information rather than comprehensive third-party tracking. For brands with strong first-party data assets, this creates competitive advantage. For those without robust first-party data collection and activation capabilities, it severely limits personalisation scope.

The path forward needs investment in first-party data infrastructure and privacy-compliant personalisation methods. Commerce media is one viable model: retailers and other companies with direct customer relationships can personalise based on owned data. Contextual targeting offers another route, using page content and user context rather than individual tracking. Both methods deliver meaningful personalisation without depending on the surveillance infrastructure regulators are dismantling.

The Control Paradox Inside Most Platforms

Most DCO platforms only allow personalisation of predefined parameters rather than full creative customisation, according to Pixis analysis. Teams face a control paradox where platforms withhold data whilst restricting creative flexibility, leaving marketers without the iterations they want or the brand guardrails they need.

This platform limitation means the automation DCO promises often requires extensive manual setup and ongoing management. Instead of letting creative teams focus on strategic work, many implementations simply shift the burden from producing multiple finished ads to configuring and maintaining complex rule sets inside constrained platform interfaces. The labour didn’t disappear. It moved.

The core limitation is that meaningful personalisation outputs require high volumes of meaningful creative inputs, and most DCO technology fails to deliver that volume. Current DCO tools inhibit marketers more than they help, with manual workflows preventing real-time creative adaptation. When you’re choosing a DCO platform, evaluate the practical usability of the creative production and management interface alongside the sophistication of the personalisation engine. The fancier engine is worthless if your team can’t feed it inputs at the rate it needs them.

Creative Fatigue and the Over-Personalisation Trap

Excessive personalisation and algorithmic content bombardment can lead to social media fatigue, user exhaustion, and platform avoidance behaviour rather than increased engagement. Research published in PMC shows that content bombardment drives platform avoidance, which directly challenges the assumption that more personalisation always produces better outcomes.

Users feel exhausted by constant algorithmic content curation and information overload. When every brand attempts sophisticated personalisation simultaneously, the cumulative effect feels invasive instead of helpful. The line between relevant and intrusive shifts based on context, relationship stage, and user mood — variables most DCO systems cannot reliably measure.

DCO strategies must balance personalisation intensity with user cognitive limits to avoid diminishing returns or negative effects on brand perception and retention. The measurement challenge is detecting fatigue before it drives disengagement. Creative fatigue measurements are one of the key performance metrics effective DCO systems must track, monitoring when specific variations lose effectiveness over time. If you’re not measuring fatigue, you’re not running DCO — you’re running a creative pipeline that happens to use templates.

Manual Workflows That Cancel the Automation

Teams with limited bandwidth choose safer creative variations over deeply personalised content that would better increase engagement, according to Pixis analysis. This conservative behaviour stems from manual approval processes and quality control requirements that most organisations impose on personalised creative.

The tension between automation and control creates operational bottlenecks. Legal and brand teams want to review every possible creative combination before it goes live, but reviewing hundreds or thousands of variations defeats the point of automated personalisation. Finding the right balance — automated assembly inside defined parameters, with human review of templates and rules rather than individual executions — requires organisational change alongside technology implementation.

This organisational challenge explains why DCO adoption significantly outpaces successful DCO implementation. The technology exists to deliver sophisticated personalisation. The workflows, approval processes, and creative production methodologies surrounding the technology often prevent it from operating as designed. Implementation needs three things: solving the creative production problem, building workflows that allow speed, and putting measurement discipline in place to prevent fatigue. Skip any one of them and the 67 per cent failure rate is where you end up.

What Worked: Five Documented DCO Implementations

Fashion&Friends: 73% ROAS on Valentine’s Day

Fashion&Friends, a multi-brand concept store, implemented customised image templates to create engaging dynamic ad creatives for Valentine’s Day cyber sales. The goal was converting growing physical store traffic into online sales. The campaign delivered a 73 per cent boost in ROAS and a 50 per cent reduction in CPA.

The implementation focused on connecting product catalogue data to dynamic templates that could showcase relevant items based on user browsing behaviour and preferences. Rather than running generic seasonal promotions, the system served personalised product recommendations within a consistent Valentine’s Day creative framework. The campaign held cohesion whilst delivering individual relevance.

The case demonstrates that even seasonal campaigns benefit from dynamic personalisation. The 73 per cent ROAS improvement suggests relevant product selection significantly outperforms generic creative, even when the seasonal messaging stays constant. The 50 per cent CPA reduction reflects improved conversion efficiency: users who see products aligned with their interests need less convincing to purchase.

Telenor’s 3x Return on Paid Social

Telenor implemented creative automation to design personalised customer journeys, using DCO to dynamically show the right phone and plan to each customer instead of serving generic ads. The telecommunications company achieved a 3x return on paid social advertising.

The implementation addressed a common challenge in telecoms marketing: the proliferation of device and plan combinations makes static ads for every option impossible to produce. By connecting product catalogues to dynamic templates, Telenor could serve relevant device-and-plan combinations based on user data and browsing behaviour.

The 3x return demonstrates the value of product-level personalisation in considered purchase categories. Telecoms contracts represent significant commitments and users research extensively before converting. Showing relevant options early in that research process improves campaign efficiency by attracting more qualified prospects and reducing wasted impressions on irrelevant offerings. In high-consideration categories, that early relevance compounds.

Wavemaker: 50% Lower CPA Through Dynamic Product Ads

Wavemaker used highly personalised ads powered by Dynamic Product Ads (DPA) for a telecommunications client to surpass expectations with faster creative testing. The implementation delivered a 50 per cent decrease in cost per acquisition and produced the best year-over-year results.

Dynamic Product Ads represent an accessible entry point to DCO capability because they use existing product catalogue infrastructure that retailers and telecoms operators already maintain for e-commerce. Rather than requiring entirely new creative production workflows, DPA implementations connect existing product data to standardised templates.

The 50 per cent CPA reduction Wavemaker’s client achieved demonstrates that even relatively basic DCO implementations can deliver substantial performance improvements when they solve the fundamental problem: showing relevant products to interested users. The year-over-year improvement suggests these gains sustain over time rather than representing a temporary boost. Sustained gains, not seasonal spikes, are what justify the investment.

Chase, American Airlines, and Commerce Media’s Expansion

Chase and American Airlines demonstrated at Cannes Lions 2024 how brands with customer journey data can build advertising businesses outside traditional retail environments. The expansion reflects growing recognition that commerce media capabilities aren’t limited to companies that directly sell products online.

Financial services companies hold extensive data about customer spending patterns, life events, and financial goals. Airlines track travel behaviour, destination preferences, and booking patterns. Both data sets enable sophisticated personalisation when activated through DCO systems. A banking customer who frequently travels to specific destinations might receive relevant airline partnership offers. An airline passenger booking business travel might see credit card promotions with business travel benefits.

This commerce media expansion is the practical application of first-party data strategy. As third-party cookies deprecate, companies with direct customer relationships and robust first-party data can build personalisation capabilities competitors without those assets cannot replicate. The fact that non-retail sectors are investing in DCO infrastructure shows how broadly applicable these methods are beyond traditional e-commerce.

The Surprise-and-Delight Layer: Marriott and Beyond

Marriott used personalisation to create surprise and delight moments during hotel stays, factoring customer emotional engagement into their personalisation strategy. The case shows how experience personalisation extends well past price optimisation. A stay personalisation programme that reads as thoughtful builds loyalty differently from a discount offer that reads as transactional.

The list of brands shifting from mass promotions to personalised offers keeps growing. Multiple retailers and restaurants — McDonald’s, Sweetgreen, Woolworth’s, Sobey’s, Target, Sephora, Boots, and Ulta Beauty — have launched personalised offers programmes, moving investment from undifferentiated promotions to personalisation focused on delivering better value. These implementations are practical applications across diverse retail categories.

The pattern across all of these is the same. Brands that already have direct customer relationships and structured first-party data are building personalisation capability now, before privacy constraints tighten further. Brands without those assets are facing harder decisions about how to compete in a market where relevance increasingly determines who wins the impression.

The Metrics That Matter Beyond Click-Through Rates

Element-Level Engagement Rates

Element-level engagement rates track the performance of individual creative components rather than complete ad variations. This granular measurement reveals which headlines drive highest engagement, which product images generate clicks, which calls-to-action convert best, and which background designs support or distract from the message.

Component-level data enables iterative improvement of the creative library. Underperforming elements get retired or redesigned. High-performing components inform future creative development. Over time, the library evolves toward increasingly effective components, lifting campaign performance even when the overall creative strategy stays constant.

Measuring individual element performance requires sufficient volume to achieve statistical significance. Each element needs exposure across various combinations and audience segments before patterns become clear. The volume requirement means element-level optimisation works best for campaigns with substantial scale. Smaller campaigns may need to focus on combination-level rather than element-level measurement until they reach the impression count where the statistics actually mean something.

Combination Effectiveness Scores

Combination effectiveness scores measure how well different element combinations perform together. A headline that works well with one product image might underperform with another. A call-to-action that converts on one background colour might fail on another. These interaction effects matter as much as individual element performance, sometimes more.

Measuring combination effectiveness requires tracking performance at the variation level whilst controlling for audience segment differences. The aim is identifying combinations that consistently outperform others across similar audience segments. Winning combinations should receive higher impression share. Underperforming combinations get retired or refined.

The challenge is balancing exploration and exploitation. Systems must keep testing new combinations to discover better options, but they should concentrate impressions on known high performers to maximise current performance. The right balance depends on campaign objectives, timeline, and budget. Brand awareness campaigns with longer timeframes can afford more exploration. Direct response campaigns with short windows need to exploit known winners quickly.

Creative Fatigue Measurements

Creative fatigue measurements monitor when specific variations lose effectiveness over time. A combination that performs well initially may decline in impact as users see it repeatedly. Detecting this fatigue early enables proactive refresh before performance degrades.

Fatigue manifests differently across metrics. Click-through rates typically decline first as users develop banner blindness. Conversion rates may hold steady longer if the offer stays relevant. Engagement time can increase if fatigued users spend longer evaluating whether to click. Tracking multiple metrics gives you a more complete picture of how fatigue is progressing.

The refresh strategy depends on fatigue severity and causes. Sometimes rotating in new creative elements solves the problem. Other situations need more fundamental redesign. The point is detecting fatigue through measurement rather than waiting for obvious performance decline. Proactive refresh maintains performance. Reactive refresh tries to recover from damage already done.

Segment-Specific Conversion Rates

Segment-specific conversion rates assess performance across different audience segments. A creative variation that performs well with one demographic might fail with another. Product images that resonate in one geographic market might seem irrelevant in another. Measuring these segment-level differences enables more sophisticated optimisation.

This measurement reveals both creative effectiveness and audience composition patterns. Segments with consistently lower conversion rates might indicate poor creative fit — or they might represent lower-intent audiences that need a different campaign approach. Distinguishing between these explanations requires combining conversion data with other engagement metrics.

The practical application is segment-specific optimisation rules. Rather than serving the same winning combination to all users, the system serves segment-appropriate variations that account for known preference patterns. This raises overall campaign effectiveness by matching creative to audience, but it requires sufficient volume in each segment to support reliable measurement. Without that volume, segment-specific rules become guesses dressed up as data.

Three Steps That Determine Whether DCO Pays Back

Step One: Generate Quality Creative at Volume

The core limitation is that meaningful personalisation outputs require high volumes of meaningful creative inputs, which most DCO technology fails to deliver. This is the fundamental implementation challenge. You cannot personalise at scale without creative components at scale. No platform solves the problem if you don’t feed it enough inputs.

Generating quality creative at volume requires different production processes than traditional campaign development. Instead of perfecting individual finished ads, teams must produce libraries of components that work together in many combinations. This shift favours systematic design frameworks, clear brand guidelines, and template-based production over bespoke creative development. The shift is uncomfortable for creative teams trained on hero asset perfection.

The production challenge extends past design work. Every component needs strategic consideration of how it fits the broader message framework, which audience segments it serves, and how it combines with other available elements. Quality at volume means holding standards whilst increasing output — a balance that requires both process improvement and realistic scope definition. Don’t promise sixty templates when your team has the bandwidth for ten.

Step Two: Build Workflows That Move at Automation Speed

Teams with limited bandwidth choose safer creative variations over deeply personalised content that would better increase engagement. This conservative pattern stems from organisational friction rather than capability limitations. Solving it requires workflow redesign alongside technology implementation. The technology runs at machine speed. Your approval process probably doesn’t.

The approval process is the most common bottleneck. Traditional creative approval reviews finished ads individually. DCO requires approving templates and rules that generate many variations automatically. This shift means reviewing the system logic and creative parameters rather than individual outputs. Stakeholders need to trust the framework rather than control every execution. That trust takes time to build, and you need to plan for that time deliberately.

Speed also requires clear ownership and decision rights. When creative production, media buying, and data teams operate independently, coordination overhead slows everything down. Successful DCO implementations often reorganise around customer segments or campaign objectives rather than functional specialities, reducing handoffs and accelerating decision-making. The org chart isn’t a sacred document. Redraw it when it’s costing you revenue.

Step Three: Measure Fatigue Before It Costs You Customers

DCO strategies must balance personalisation intensity with user cognitive limits to avoid diminishing returns or negative effects on brand perception and retention. Measurement discipline is what detects problems before they cause significant damage.

Frequency capping is the most basic fatigue prevention mechanism. Users should not see the same variation repeatedly inside short timeframes, regardless of how well that variation performs. Past frequency limits, monitoring engagement trends helps you detect fatigue as it develops. Declining click-through rates or rising bounce rates signal that creative has lost effectiveness.

Measurement discipline must extend past performance metrics to brand health indicators. Over-personalisation can improve short-term conversion whilst damaging long-term brand perception. Tracking brand metrics alongside campaign performance ensures personalisation enhances rather than erodes brand value. If brand consideration scores decline whilst conversion rates increase, the personalisation strategy is too aggressive and needs recalibration. The 67 per cent failure number cited at the start of this article is mostly composed of organisations that didn’t catch this signal in time.

The Maturity Path: From Product Catalogues to Automated Systems

Entry Point: Dynamic Product Catalogues

Dynamic Product Ads represent the most accessible entry point to DCO capability. These implementations connect existing product catalogue data to standardised ad templates, enabling product-level personalisation without extensive creative production. Dynamic creative optimisation lets brands show personalised products instead of generic ads, as HunchAds notes.

Feed-based personalisation works well for e-commerce, telecoms, travel, and other sectors with structured product catalogues. The implementation challenge is data quality and template design rather than sophisticated machine learning or complex rule configuration. Most major advertising platforms provide native Dynamic Product Ad capabilities, reducing technical integration requirements.

The limitation of this entry-level approach is that personalisation extends only to product selection, not to broader creative messaging, emotional tone, or brand positioning. The template stays static even as the product varies. For many campaigns, product-level personalisation delivers enough performance improvement to justify the investment. For others, it represents a first step toward more sophisticated approaches.

Intermediate: Multi-Element Optimisation

Intermediate DCO implementations personalise multiple creative elements simultaneously based on real-time data signals. Instead of just varying the product shown, these systems adjust headlines, calls-to-action, background designs, and messaging based on user data, behavioural signals, and contextual factors like location, device, time of day, and weather.

This stage requires more sophisticated creative production processes. Teams must develop component libraries with sufficient variety to enable meaningful personalisation across multiple dimensions. The template architecture has to accommodate these multiple variable elements whilst maintaining brand consistency and message coherence.

The data integration challenge also increases. Basic product catalogue feeds give way to audience data management, behavioural tracking, and real-time signal processing. All DCO data sources must be managed through a centralised data management platform, creating integration and data quality requirements that exceed the capabilities of many marketing organisations. If your data infrastructure isn’t ready, the creative infrastructure can’t help you.

Advanced: Cross-Channel Orchestration and Predictive Analytics

Advanced DCO implementations extend personalisation across multiple channels whilst maintaining consistent messaging and avoiding creative fatigue. A user who sees a particular creative variation on social media should not see conflicting messaging in display advertising or receive misaligned email communications. Cross-channel orchestration requires unified data platforms and coordinated campaign management.

Predictive analytics integration enables anticipatory personalisation rather than purely reactive approaches. Instead of responding to documented user behaviour, these systems predict likely needs based on patterns and life stage indicators. A customer whose transaction patterns suggest an upcoming purchase receives relevant creative before explicitly expressing intent.

Dynamic video optimisation is another advanced application, enabling personalisation of video creative through automated editing and assembly. Personalised video ads show 83 per cent higher emotional engagement, according to Unruly’s consumer data cited by DevriX. Video personalisation requires substantially more sophisticated production and rendering infrastructure than static display personalisation, so the bar to entry is higher and the payoff is correspondingly larger.

The Privacy-Compliant Future

The future of DCO necessarily operates within privacy-compliant frameworks that prioritise first-party data and contextual signals over comprehensive user tracking. The shift constrains certain personalisation methods whilst enabling others. Brands with strong first-party data collection and robust consent management can deliver sophisticated personalisation. Those without these capabilities face increasing limitations.

Contextual personalisation based on page content, user location, device type, time, and weather requires no personal data tracking. A user viewing winter sports content sees winter sports products. A user browsing from a mobile device sees mobile-optimised creative. These contextual signals enable relevant personalisation without privacy concerns and without depending on infrastructure that’s actively being dismantled.

The retail media model demonstrates one viable privacy-compliant path. Retailers personalise based on owned customer data and purchase history, delivering relevance without depending on third-party tracking. As commerce media expands past retail into financial services, airlines, telecoms, and other sectors with direct customer relationships, this first-party data approach becomes increasingly central to DCO strategy. The brands building these capabilities now are positioning themselves for a market that’s already emerging.

Frequently Asked Questions

Where DCO Goes From Here

Two figures from Boston Consulting Group define the situation: 80 per cent comfortable, 67 per cent burned. The market wants personalisation. The execution standard is unforgiving. That tension is what makes dynamic creative optimisation simultaneously the largest opportunity and the most operationally demanding capability in modern marketing. The 3x ROI is real. So is the failure rate that sits behind it.

The brands capturing the upside share three traits. They’ve built first-party data assets that operate inside privacy-compliant frameworks. They’ve redesigned creative production around modular component libraries rather than finished hero assets. They’ve put measurement discipline in place that catches fatigue before it damages brand health. None of these are technology decisions. They’re organisational decisions that determine whether the technology can do what it promises.

The $570 billion incremental growth opportunity that BCG attributes to top retailers on its Personalization Index by 2030 isn’t going to spread itself evenly. It’s going to concentrate among brands that solved the implementation problem before competitors did. The retail media expansion that Chase and American Airlines demonstrated at Cannes Lions 2024 indicates the addressable market extends well past traditional retail. The 99 per cent of agencies who told Digiday and Clinch that DCO is now a significant factor in their work are signalling where the next decade of marketing investment is going.

If you’re evaluating DCO right now, the most useful thing you can do is be honest about which side of the 67 per cent failure number your organisation is currently positioned for. The technology investment matters less than the operational readiness. Get the creative production, workflow, and measurement layers right, and the platform you choose mostly executes what you’ve already decided. Get them wrong, and no amount of platform sophistication compensates. No fluff: that’s the call you need to make this quarter, not next year.

Sources

Boston Consulting Group. “What Consumers Want from Personalization.” bcg.com

Acxiom. “What is Dynamic Creative Optimization (DCO)?” acxiom.com

DevriX. “Dynamic Creative Optimization Explained: What It Is & How It Works.” devrix.com

DP6. “Cannes Lions 2024 — Insights and Learnings on Data and Creative Effectiveness.” Medium. medium.com

Storyteq. “What is Dynamic Creative Optimization (DCO)?” storyteq.com

“Too much social media? Unveiling the effects of determinants in social media fatigue.” PMC. pmc.ncbi.nlm.nih.gov

AppsFlyer. “A comprehensive guide to dynamic creative optimization in 2024.” appsflyer.com

Boston Consulting Group. “Retail Spotlight: Personalization in Action.” bcg.com

HunchAds. “Dynamic Creative Optimization Case Study: Does DCO Work?” hunchads.com

Pixis. “How Dynamic Creative Optimization (DCO) Should Work.” pixis.ai

Disclosure: This article was produced using AI-assisted writing tools. The underlying research was gathered, analysed, and verified by human researchers. Final editorial review, fact-checking, and quality control were performed by human editors.