The concept is sound. The execution is where it collapses.
A persona built from real behavioural data, wired to journey stages and connected to what the platform actually does, is an engineering input. It shapes rules. It drives decisions. It gives your personalisation logic something to act on. Without that connection, it is just a slide.
The capability exists. Optimizely's Opal handles AI-driven content and layout adaptation based on behavioural signals, without heavy dev dependency. Real-time adjustment, smarter experimentation, recommendations that reflect what users are doing. Yet Opal is one of the most underused parts of the Optimizely suite, and the reasons are usually the same: content not built to flex, data not connected, no clear owner for the output.
The first thing to look at is data. Personalisation logic needs a signal to act on. If the behavioural data being collected is incomplete, inconsistently tracked, or not connected to the platform in a usable form, the rules have nothing to work with. The practical question is whether the data your platform has access to actually reflects how users behave on your highest-value journeys: join, apply, renew, enquire, buy. If the honest answer is partial, that is where the work starts.
The second element to look at is content. Adaptive personalisation requires content built to flex. A single page with a fixed headline and a generic CTA behaves the same way for every user regardless of how sophisticated the underlying logic is. Teams that have made personalisation work have usually built content variants before they built the rules, treating content architecture as part of the platform configuration rather than a separate workstream.
The third is governance. Experimentation and personalisation stall when decision-making is slower than the cadence they require. Three rounds of sign-off to change a content variant, or no clear owner for experiment results, means the programme learns slowly and loses internal confidence. The cadence of decisions needs to match the cadence of the programme.
For organisations in regulated or high-consideration sectors - pensions, membership bodies, financial services, healthtech - the commercial stakes on key journeys are high enough that getting personalisation right is worth the effort. Renewal journeys that recognise and respond to a ten-year member differently to a prospect. Acquisition flows that adapt based on the channel a user arrived from. Self-service experiences that surface the right answer at the right point and absorb demand that would otherwise hit a call centre.
These outcomes require data, content, platform configuration and ways of working to move together. When they do, each experiment informs the next, audience understanding compounds, and the platform earns its licence fee.
Personalisation performs when strategy, insight, operating model and performance are running in parallel, each informing the others. Optimizely makes adaptive personalisation operable at scale. A clearer operating model and a platform configured to deliver it are where the commercial difference shows up.
Mando Group's Optimise service is built around those conditions. Digital strategy, insights, operating model and performance running as a joined-up programme, designed to turn a live platform into a system for continuous improvement. For digital leaders where performance directly impacts growth, retention or trust, it is the difference between personalisation as an aspiration and personalisation as a working capability.