What Our CTO Taught Me About Personalization Engines (and where they’re headed next)

Personalization has become the beating heart of digital experiences. Customers expect websites to know what they want before they do, whether that’s recommending the next concert, surfacing a relevant blog post, or suggesting the perfect product add-on at checkout. 

Core dna personalization engine is designed for marketers who want speed without sacrificing control, and for businesses who need personalization baked directly into their CMS, commerce, and content workflows.

Recently, I sat down with our CTO to talk about a project we rolled out for Frontier, a leading live-entertainment brand. What he shared gave me a behind-the-scenes view into how the engine actually works, why it’s so fast, and where personalization is headed next in the age of generative AI.

A coffee break with our CTO and everything I learned about personalization

When I sat down with  Dmitry our CTO  to talk about personalization, the conversation started with Frontier Touring. We recently launched their new personalization engine that could keep pace with live entertainment. They needed  banners and recommendations that felt instantly relevant without slowing the site down and we delivered. 

Dmitry explained how the first attempt took nearly three seconds to load. That might not sound like much, but in digital time it feels endless. The issue was that the system was trying to rebuild records on the fly, page by page, and the process was just too heavy.

The fix was simple in principle but powerful in practice: move the heavy lifting to orchestration. Instead of trying to rebuild everything in real time, Core dna precomputes a slim metadata file that contains only the essentials,  an image, a link, an ID, and tags like location or genre. The result is personalization that renders in milliseconds.

And here’s why this mattered for Frontier: they deliberately avoided an out-of-the-box personalization tool. Their needs were very specific,  like giving marketers the ability to exclude certain tours or artists from recommendations, or to switch off entire sections when sensitivities demanded it. That kind of fine-grained control is rarely possible in off-the-shelf platforms. 

With Core dna, the framework could be built as low-code logic inside their existing environment, tailored exactly to their rules and workflows. For Frontier, not having to adopt a new platform but extending the one they already run on was priceless.

The approach has a few advantages:

• Any content can be personalized. For Frontier it was tours and banners, but the same method works for blog posts, products, FAQs, or landing pages.
• Marketers stay in control. Switches like “exclude from personalization” let teams respect brand sensitivities. At Frontier, some artists did not want cross-promotion on their pages, so the toggle handled that.
• User behavior is stored simply. We track activity in the browser’s localStorage, not cookies. If a user logs in, that history ties back to their user profile so the same recommendations follow them from laptop to phone.
• Caching is cost-smart. Personalization requests are passed as URLs with parameters. Identical parameters create identical URLs, which can be cached. This reduces uncached requests and helps control costs.

From a user’s perspective, none of this is visible, the page just loads, fast and relevant. Behind the scenes, the engine is caching URL-parameterized requests, storing browsing history in localStorage, and syncing across devices when a user logs in. That combination of speed, transparency, and cost-efficiency is what makes the engine practical in the real world.

How Core dna’s Personalization Engine Stacks Up Against the Market

Most personalization tools look great on a slide deck but often leave marketers frustrated when it comes to execution. Either they slow down the site, lock you into IT dependencies, or rack up costs when traffic scales.

As I talked this through with our CTO, we kept circling back to the same point: marketers want personalization that feels magical on the front end but is practical behind the scenes. That is where Core dna shines.

Here are the most common struggles we see in the market, and how our personalization engine tackles them head-on:

1. Personalization that slows down the site

We’ve all seen it: a page takes two or three seconds to swap in a “personalized” banner. By then the user has scrolled past, or worse, bounced. Tools that rely on real-time external calls just can’t keep up.

Core dna difference: The heavy work is done in orchestration, so the front end only has to render slim metadata. The result is personalization that appears in milliseconds, even on conversion-critical pages like the homepage.

2. Decisions hidden in a black box

Other platforms often make it hard to know why a certain item appeared. Was it a rule? An algorithm? A random guess? Marketers are left in the dark.

Core dna difference: Every campaign comes with decision logs and exclude toggles. You can see exactly why a banner appeared and switch it off if it doesn’t fit the context.

3. Costs that scale faster than traffic

With many SaaS vendors, every un-cached personalization request adds to your bill. Suddenly, your great campaign becomes a CFO headache.

Core dna difference: By embedding personalization in URL parameters, identical scenarios generate identical URLs. That means responses are cached at the edge, keeping costs predictable even at scale.

4. AI that feels like a gamble

AI-first personalization tools can be impressive, but they also make mistakes that damage brand trust. A rock artist tagged as “pop” is enough to cause headaches.

Core dna difference: AI enrichment is optional. You can lean on AI to auto-tag personas, topics, or audiences, or stick with manual rules when brand sensitivity matters.

5. Siloed personalization

Most platforms stop at the web layer. If you want personalization in email, eCommerce, or learning, you need extra tools or integrations.

Core dna difference: Our personalization engine is module-agnostic. The same metadata and rounds can drive banners, blog recs, product bundles, LMS lessons, even pop-ups.

6. IT bottlenecks

Some enterprise systems put IT in the driver’s seat, which means marketers wait weeks for a dev sprint just to test a new campaign.

Core dna difference: Round rules and toggles live inside the CMS interface marketers already use. IT sets the framework once, and marketing teams can run with it.

Bottom line: In most platforms, you’re forced to pick between speed, control, and cost. Core dna’s personalization engine gives you all three. That makes personalization practical for every marketer, not just those with unlimited budgets or patient IT teams.

Real-world applications: Best practices by industry

Because Core dna’s personalization engine is powered by orchestration and AI, it adapts to the personalization strategies that are proven to work best in each industry. Unlike most platforms, it is not limited to your internal content. 

Core dna can also tap into publicly available data — like upcoming concerts, local events, or trending news — to enrich recommendations even for anonymous visitors. That means the experience can feel personal even if we know nothing about the user yet.

B2B: Guiding the buyer journey

When it comes to B2B personalization, the best approach would be to help buyers quickly find the right resource at the right stage of the funnel. Prospects do not want to sift through irrelevant case studies or generic whitepapers when they are looking for something specific.

Core dna personalization engine gives your the ability to:

• Map content to personas and buying stages (technical, financial, executive) and surface the right mix.
• Adapt calls-to-action dynamically. For example, a first-time visitor sees a product overview, while a returning CTO sees an integration guide.
• Enrich content tags with AI-generated personas and topics so marketers do not have to label everything manually.
• Pull in industry news or market benchmarks from public sources to recommend thought leadership that builds credibility, even for anonymous visitors.

Media & entertainment: Instant, relevant, and dynamic

In media and entertainment, personalization is all about immediacy. Audiences should not feel that they are being tracked. The recommendations should make them pause and think, “is someone listening to me?”

Core dna personalization engine gives your the ability to:

• Serve personalized recommendations in milliseconds, so the experience feels natural and seamless.
• Use location, genre, and browsing history to tailor feeds without overstepping.
• Respect editorial control with switches and rules, ensuring sensitive content is excluded where necessary.
  
• Orchestrate external event feeds (for example, local concerts or cultural events) so even brand-new visitors get recommendations that feel surprisingly relevant.

Result: Higher engagement and repeat visits, with audiences experiencing personalization that feels magical rather than mechanical.

Franchise & multi-location brands: Local at scale

For franchises, personalization has a few more challenges as it must balance global brand consistency with local relevance. Customers expect a familiar brand experience, but they also want offers, content, and services that feel specific to their region or store.

Core dna personalization engine gives your the ability to:

• Target promotions by location without duplicating campaigns across hundreds of outlets.
  
• Push global brand assets while enabling local teams to adapt rules for their market.
  
• Connect personalization to local pricing, offers, and availability in commerce.
  
• Use publicly available regional data (like weather or community events) to trigger campaigns that feel timely and relevant.
  

Result: Customers enjoy a unified brand that still feels personal to their neighborhood, while franchise teams cut down on duplicated work and complexity.

Memberships & communities: Deepening engagement

For membership organizations, personalization is not just a nice-to-have, it is central to keeping members engaged and loyal. People join because they want to feel part of something bigger, but they stay when the experience feels tailored to their individual needs and interests.

Core dna personalization engine gives your the ability to:

• Personalize onboarding journeys so new members get helpful guidance, while long-term members are shown advanced opportunities or exclusive perks.
• Recommend resources based on past behavior and engagement scoring, such as courses completed, events attended, or articles read.
• Trigger re-engagement for inactive members with nudges tied to publicly available content like new research, industry updates, or local events that match their profile.
  

Result: Members feel recognized and supported at every stage of their journey, which strengthens loyalty and increases lifetime value.

The road ahead: Zero-shot to Head-shot personalization

The personalization engine we deliver today already transforms digital experiences by making them faster, smarter, and easier to manage. But the future is moving even further, into zero-shot to head-shot personalization in the age of generative AI.

The idea is simple: personalization should not be limited to what we already know about a user. With AI and orchestration, we can move from rules and history to real-time inference and proactive experiences.

Zero-shot: personalizing when we know nothing

A visitor arrives on your site for the very first time. No login, no history. In most systems, that would mean showing something generic. With Core dna, we can do better. By blending publicly available data (like local events, trending topics, or regional conditions) with contextual signals (such as geolocation), we can deliver recommendations that feel relevant from the very first click.

Few-shot: learning fast from small signals

As the user browses, the personalization engine starts collecting lightweight signals — the pages they visited, the tags those pages carry, the time they spent engaging. AI enrichment can then assign likely personas, topics, or audience profiles. Within just a few interactions, the experience shifts from generic to personal.

Head-shot: hyperpersonalization at scale

Over time, the personalization engine evolves into a personalization agent. It is no longer just responding to past actions. It is predicting next steps, orchestrating journeys across channels, and adapting experiences in real time. Examples include:

• Recommending an alternative concert the moment a preferred event sells out.
  
• Nudging a B2B buyer with a chatbot prompt when they appear to be comparing three similar products.
  
• Adapting membership onboarding based on both explicit preferences and inferred goals.
  

This is hyperpersonalization in practice, a system that feels as if it knows the user intimately, even when data is sparse.

Why Core dna is ready for this future

Because Core dna’s personalization engine is powered by orchestration, it is not tied to one model or channel. It can already integrate external data sources, enrich metadata with AI, and apply flexible rules. That foundation means our clients are not just solving today’s personalization challenges. They are stepping onto a path where AI-driven, orchestrated hyper-personalization becomes their competitive edge.