Website Personalisation Basics

Website personalisation — showing different visitors different experiences based on who they are or what they've done — is CRO's most-hyped and most-misapplied frontier: promised as the future where every visitor sees their perfect page, delivered too often as complexity that costs more than it converts. The honest reality is that personalisation works in specific, high-value cases and is over-engineering everywhere else. Here's personalisation grounded: what it is, where it genuinely pays, and where the simpler path wins.

What personalisation actually is

Personalisation shows tailored experiences based on visitor signals: who they are (location, device, traffic source, new-vs-returning, logged-in data), what they've done (past visits, past purchases, cart contents, browsing behaviour), or where they came from (the message-match writ dynamic — matching the page to the ad/campaign/query that brought them). It ranges from simple (returning visitors see "welcome back," location shows local info) to complex (full behavioural targeting, dynamic content per segment, predictive recommendations). The promise is relevance-drives-conversion (a more relevant experience converts better); the reality is that relevance does help but personalisation's cost and complexity often exceed the gain — so the question isn't "does personalisation work?" (relevance helps) but "does this personalisation's gain exceed its cost?" (often no).

Where it genuinely pays

The high-value cases where personalisation earns its complexity: the obvious, low-cost tailoring (location-based content where location matters, device-appropriate experiences per mobile, language/market matching — the personalisation that's really just doing the basics right, cheap and clearly beneficial); returning-visitor and customer recognition (showing returning visitors/customers relevant continuations — the "welcome back, here's where you left off," the account-based experience — genuine value at reasonable cost for logged-in/known users); behavioural relevance at scale (for high-traffic ecommerce with real data — product recommendations, cart-based offers, the personalised commerce that genuinely lifts revenue when the data and volume support it); and campaign message-match (matching landing experiences to traffic source — the dynamic message-match that reduces the mismatch-bounce). These share a pattern: clear relevance value, reasonable implementation cost, sufficient data/volume to matter.

Where the simpler path wins

The over-engineering to avoid: complex personalisation without the data/volume to support it (behavioural targeting on low traffic — not enough data to segment meaningfully, complexity for no gain); personalisation over a broken fundamental (tailoring a page whose value proposition is unclear or whose checkout is broken — fix the fundamental for everyone first; personalising a bad experience just makes bad-experiences-tailored); the maintenance and complexity cost (every personalised variant is more to build, test, maintain and debug — the complexity that can exceed the conversion gain, especially for smaller operations); and the privacy considerations (personalisation uses visitor data, with the privacy and consent obligations that entails). The honest guidance: do the simple, clearly-valuable personalisation (the basics-done-right), and get the fundamentals right for everyone before adding complexity — a well-optimised single experience beats a poorly-executed personalised one, and personalisation is the advanced move for after the fundamentals are solid and the data/volume justify it, not the first resort. On traffic content and authority earned.

Frequently asked questions

Is website personalisation worth it?

In specific high-value cases (clear relevance, reasonable cost, sufficient data/volume — the location/device basics, returning-customer recognition, high-traffic behavioural commerce, campaign message-match), yes; as complex targeting on low traffic or over a broken fundamental, usually not (complexity exceeding gain). The question is whether this personalisation's gain beats its cost — often the simpler well-optimised single experience wins.

Do I need personalisation to compete on conversion?

No — a well-optimised single experience (strong value prop, smooth checkout, good mobile, real trust) outconverts a poorly-executed personalised one. Get the fundamentals right for everyone first; personalisation is the advanced layer for after, where data and volume justify it — not a prerequisite for good conversion.

What's the simplest personalisation worth doing?

The basics-done-right — location-appropriate content where location matters, device-optimised experiences, language/market matching, returning-visitor recognition — the low-cost, clearly-beneficial tailoring that's really just serving visitors well. Start there (cheap, clear value); add complex personalisation only when the fundamentals are solid and the data/volume support it — on traffic content and authority deliver (our half).