Personalisation in E-Commerce: Relevant Without Being Creepy

There is a fine line, and almost everyone has felt it. On one side is the shop that seems to just get you — it remembers what you were looking at, suggests the thing you actually wanted, and quietly makes life easier. On the other side is the shop that knows a little too much, that follows you around the internet with the exact pair of shoes you glanced at once, that makes you feel watched rather than understood. Same technology. Wildly different feeling.

That line is what this guide is about. Personalisation — the practice of tailoring what a shopper sees based on who they are and what they have done — is one of the most powerful tools in online retail. Done with care, it lifts sales and genuinely improves the experience. Done carelessly, it unsettles people and chases them away. The goal is to land firmly on the helpful side of that line, and to understand exactly where it sits.

What personalisation really means

At its simplest, personalisation is showing different shoppers different things based on what is likely to be useful to them. A returning customer sees “welcome back” and the items they were eyeing. A first-time visitor from a particular interest sees products that match it. Someone who just bought a camera sees suggestions for a case and a memory card rather than another camera. It is the digital version of a good shopkeeper who remembers your last visit.

None of this requires knowing someone's secrets. The most effective personalisation is built on simple, obvious signals — what someone clicked, what they bought, what is in their cart — used to make the shop more relevant. The trouble starts when stores reach for data that feels invasive, or use ordinary data in ways that feel like surveillance. The data is rarely the problem; the way it is used is. This kind of relevance is part of what makes a high-converting product page feel tailor-made.

Where personalisation genuinely helps

The best personalisation feels like a service, not a trick. It removes friction, surfaces relevance, and saves the shopper effort. A handful of uses deliver real value without ever crossing into discomfort, and they are the sensible place to start.

“You might also like” done well

Product recommendations are personalisation's friendliest face. Showing items that genuinely complement what someone is viewing — a lens for a camera, socks for the boots — helps the shopper and naturally raises order value. The key word is genuinely. Relevant suggestions feel like good advice; random or pushy ones feel like clutter. Handled well, this overlaps neatly with thoughtful upselling and cross-selling that adds value rather than annoyance.

Remembering, so the shopper doesn't have to

Bringing back a half-filled cart, saving a wishlist, remembering a size preference, picking up where someone left off — these are small kindnesses that shoppers appreciate. They reduce effort and signal that the store respects the person's time. This kind of gentle continuity also quietly helps recover lost sales that would otherwise drift away.

Relevant search and discovery

When a returning shopper's searches and browsing reflect their evident interests, finding things gets easier. A store that learns someone shops mostly for one category can surface that category's new arrivals first. This is personalisation working hand in hand with strong site search to make discovery effortless.

Where it tips into creepy

The discomfort almost always comes from one of three things: data the shopper did not realise you had, relevance so sharp it feels like surveillance, or pressure dressed up as personalisation. Knowing these tripwires helps you stay on the right side of the line.

The first is the unexpected. When a shop seems to know something a shopper never told it — or never expected it to remember — the reaction is unease, not delight. The second is intensity. A single helpful reminder is welcome; the same item chasing a person across every site and inbox for a fortnight feels like being followed. The third is manipulation. When personalisation is used to apply pressure, hide information, or make someone feel they are getting a worse deal than others, it stops being a service and becomes a reason to distrust the brand entirely. Trust, once dented, is slow to rebuild — which is why personalisation should reinforce, never undermine, the work you do on reviews and social proof and overall credibility.

The principles that keep you on the right side

You do not need a rulebook for every situation. A few guiding principles will keep almost any personalisation feeling helpful rather than unsettling.

Be transparent

People are remarkably forgiving when they understand what is happening. “Recommended because you viewed similar items” turns a mysterious suggestion into an obvious, reasonable one. Quiet, hidden personalisation breeds suspicion; openly explained personalisation builds trust. When in doubt, show your working.

Give people control

The ability to clear a browsing history, adjust preferences, opt out of tracking, or say “not interested” transforms the relationship. Control turns personalisation from something done to the shopper into something done with them. People rarely object to relevance they can steer; they object to relevance they cannot escape.

Use the lightest data that does the job

You usually do not need much. What someone is looking at right now, what is in their cart, what they bought last time — these simple signals power most of the value without venturing anywhere intrusive. Resist the temptation to hoard data just because you can. The more sensitive the data, the higher the risk and the lower the reward. Respecting privacy is not just ethical; in a world of tightening rules and rising awareness, it is good business.

Personalisation across the journey

Done thoughtfully, personalisation does not live in one place — it threads quietly through the whole experience. A relevant homepage greets a returning visitor. Helpful recommendations sit on product pages. A gentle, well-timed reminder follows an abandoned cart by email. After the sale, suggestions reflect what the customer actually bought, strengthening the mobile experience where so much of this now plays out.

The same relevance increasingly extends into conversation. As shopping moves toward chat-based experiences and conversational commerce, a system that remembers a customer's preferences can make a chat feel like talking to someone who knows you — in the good way. The broader direction of this, where intelligent systems tailor the whole journey, is explored in our piece on agentic AI in e-commerce.

Start small, measure, and stay honest

You do not have to personalise everything at once, and you certainly should not. Start with the safe, obvious wins — relevant recommendations, saved carts, a warm welcome back — and watch what happens. Measure not just sales but sentiment. Are customers engaging more, or are opt-outs and complaints rising? The numbers will tell you whether your personalisation feels like a service or a stalker.

Above all, keep asking the simple human question: would this delight me or unsettle me if a shop did it to me? That instinct, applied honestly, will steer you right more reliably than any clever algorithm. The shops that win long-term loyalty are the ones that use what they know to make life genuinely easier, and never to make their customers feel watched. Get that balance right and personalisation becomes exactly what it should be — the digital echo of a shopkeeper who remembers your name, in the warm way, and nothing more. If you would like a hand designing it sensibly for your store, you can always reach us through the contact page.

Frequently asked questions

References

1. McKinsey & Company. “The Value of Getting Personalisation Right.” mckinsey.com.
2. Nielsen Norman Group. “Personalisation and User Trust.” nngroup.com.
3. Forrester. “Privacy, Trust and Customer Experience.” forrester.com.