Want to personalize more? Then hire more people and plan for longer nights at the office.
That’s the hidden assumption behind many personalization programs. More segments mean more audience rules to build. More products mean more selections to make. More channels mean more campaigns to coordinate. And when the work is still manual, “personalization at scale” can start to look like scaling the workload instead.
But should better personalization really require a bigger team and longer hours?
Marketers already know what relevant experiences should look like. The problem is producing enough of them. In Salesforce’s 2026 State of Marketing report, 78% of marketers said they need more personalized content than they’re currently able to produce. The research surveyed 4,450 marketing professionals worldwide.
So Why Marketers Can’t Deliver Enough of It?
Marketers don’t need convincing that personalization matters. The harder question is whether the systems they rely on can actually deliver it at scale, across every customer touchpoint, without turning personalization into another manual workload.
That sounds less like a failure of imagination and more like a failure of execution. Marketers know what relevant experiences should look like. But too often, the tools (mostly marketing automation and CDP platforms) and processes meant to enable them create more work instead.
When the personalization tools become the workload
In e-commerce, marketing automation (MA) and customer data platform (CDP) systems can provide valuable capabilities. But having the platforms is not the same as delivering genuinely individual experiences.
Teams may still have to build segments by hand, define campaign rules, select products for each audience, and coordinate separate versions of a message across channels. The result can be a complex system that helps marketers send more campaigns, but still relies on people to decide who sees what, where, and when.
Is that personalization at scale? Or is it manual campaign production with more technology around it?
A segment such as “customers who bought running shoes in the past six months” may be useful. But it still groups many people together. One shopper may be looking for winter trail shoes; another may have stopped running. If marketers have to anticipate every meaningful difference and build a segment or rule for it, the number of exceptions quickly outgrows the team’s capacity.
Personalization should follow the customer, not the channel plan
A customer doesn’t experience a brand as a collection of marketing channels. They might browse a product page, leave without buying, open an email later, and return through an ad. If each channel is planned and optimized separately, the experience can feel fragmented, even when every individual interaction was technically “personalized.”
Salesforce’s report describes marketers operating across an average of 10 channels. It also finds that website experiences lag behind some other channels in individual personalization. That raises an important question: why should a customer’s experience become less relevant the moment they leave their inbox?
Personalization should be able to shape the experience wherever the customer interacts with the brand: on the website, in email and mobile messages, in product recommendations, and in other channels. The specific product, content, or offer should reflect that individual’s current context–not just a marketer’s pre-built segment.
From manually managed campaigns to autonomous decisions
The next step is not simply adding AI to the existing workflow. It is changing who or what does the repetitive decision-making.
An autonomous personalization system should be able to use available customer and product signals to decide, in real time, what is most relevant for each person and which channel or moment is appropriate. In e-commerce, that mean choosing which products to feature on a page, adjusting recommendations as someone browses, and coordinating those choices with follow-up communications.
The goal should be to eliminate recurring manual work such as building endless segments, choosing product lists for every campaign, and maintaining separate rules for every channel. Marketers should set the objectives, brand boundaries, and business rules; the system should handle the individual decisions and execution within those guardrails.
That isn’t the same as removing human responsibility. Teams still need to define what good outcomes look like, protect the customer experience, and monitor performance. But they shouldn’t have to manually orchestrate every customer-level interaction just to make personalization happen.
6 questions marketers should be asking
The 78% production gap is a reminder that marketers already see the need for more relevance. The challenge is building an operating model that can deliver it without requiring a new round of manual work for every segment, product selection, and channel. 6 questions marketers should be asking:
- Are we personalizing for individuals, or simply dividing customers into more and more segments?
- How much time does the team spend choosing audiences, products, and campaign variations by hand?
- Can our systems make coordinated decisions across the website, email, mobile, and other channels?
- Does our CDP or MA platform reduce operational effort or does it mostly give us more rules and workflows to manage?
- Can we measure whether the experience was relevant to each customer, not just whether a campaign performed well on average?
- What would our team do with the time freed up if routine personalization decisions were automated?
The real test isn’t whether a company has AI, a CDP, or marketing automation. It’s whether those systems can work together to make useful, timely decisions for each customer and do so across the whole experience. That’s the genuine difference between CDP or MA system and AI recommendation engine.
Discover Quarticon’s recommendation engine for e-commerce that works autonomously across channels, including web, apps, email, and more.
