The Made With Intent blog

You've done the work. You've split your lapsed customers from your actives, your high-AOV buyers from your one-time purchasers, your engaged openers from the dormant half of your list. You've built the flows and set the rules. And your recovery rates are still flat.

The problem is your email segmentation tells you who's on your list and what they've done. They don't tell you who's in a buying moment right now. That's not a gap in your execution, it's merely a limitation of how segmentation works. And until you see it clearly, you'll keep optimising the wrong thing.

Email segmentation is built on historical data. Purchase history, past engagement, demographic profile. It tells you who someone was. But, it doesn't tell you what they're about to do. The signal that tells you whether someone is actively considering a purchase right now is a different kind of data entirely and it lives somewhere most email strategies never look.

What email segmentation actually tells you

Standard email segmentation is genuinely useful. Don't let the argument of this article dispel this notion.

When you split your list by purchase history, you're identifying category affinity and buying patterns. When you score subscribers by recency, frequency, and monetary value, the classic RFM model, you're surfacing the customers most likely to respond to an offer at a population level. When you tag customers by product category or browsing history, you can send more relevant messages than a broadcast to everyone.

These are real improvements. A CRM manager who has moved from bulk email to properly structured behavioural segmentation has genuinely raised their ceiling. Open rates improve. Revenue per send goes up. Unsubscribe rates come down.

However, segmentation is a map of things that have already happened. It's also focused on past behaviours, i.e. when someone has already signed up for an email. The ability to act immediately simply isn't there.

Your post-purchase segment tells you someone bought. Your lapsed segment tells you they haven't bought recently. Your engagement segment tells you who opened your last campaign. None of it tells you which of those subscribers is actively considering a purchase right now, in this session, having visited your site three times this week to look at the same product.

That's the gap. And it's why flat recovery rates survive even well-built segments.

The gap: why past behaviour isn't the same as present intent

To understand why this matters practically, you need to separate two things that email marketing tends to conflate: who someone is on your list and where they are in their buying journey right now.

There are two dimensions to this: the in-session signal (what this person is doing right now) and the broader buying window (are they actively in the market at all?). Standard segments miss both.

The historical data problem

Think about RFM. A customer in your "champions" segment (high recency, high frequency, high value) is someone you'd rightly treat as a priority. But their segment tells you they've bought before and bought recently. It doesn't tell you whether they're actively considering a purchase today. They might be. They might also have bought what they needed last month and have no intention of buying again for six weeks.

RFM is excellent at predicting which customers are likely to respond to a campaign over time, at a population level. It isn't designed to tell you which specific customer is in an active buying moment right now. Those are different questions, and treating the first as a proxy for the second is where the gap opens.

The engagement data problem

The standard workaround is engagement-based segmentation: split your list into openers and non-openers, active and inactive, and weight your sends accordingly. The principle is sound. The inputs have a serious problem.

Since Apple introduced Mail Privacy Protection with iOS 15 in September 2021, a proportion of email opens have been pre-fetched by Apple's servers rather than triggered by a subscriber actually opening the email.

According to Litmus's email client market share data, over 50% of email opens now occur on devices with Apple's Mail Privacy Protection activated. For UK fashion, beauty, and home retailers, where iPhone dominates device usage, the proportion of affected opens on your list is likely significant.

If your engagement-based segments were built or last reviewed before late 2021, it's worth auditing what proportion of your "engaged subscriber" definition rests on open rate signals and whether click and conversion data could give you a more reliable proxy. However, this isn't us saying "abandon engagement segmentation". We're saying you've got a reason to check that your inputs still mean what you think they mean.

The timing problem

Even if your segment is perfectly accurate, it doesn't solve the timing problem.

Suppose you have a genuinely high-intent customer: they're in market, they've been browsing your site, they're ready to buy. If that customer is in your 60-day lapsed segment, they'll receive your win-back flow on whatever cadence that flow runs. If they're in your post-purchase segment, they'll receive your next post-purchase email at the point the flow schedules it.

Neither of those flows asks: is this person ready to buy today? The segments tell you who to send to. They don't tell you when that person is receptive, or when they're actively in the middle of a buying decision that the right message could tip.

What "ready to buy" actually looks like

Buying intent doesn't show up in your CRM. It shows up in behaviour, but not in the way most email strategies assume.

The instinct is to look for observable signals: a customer who's visited the site twice this week, spent time on the outerwear category, or added something to their basket and removed it again.

These patterns feel meaningful, and they are. But they're proxies. They approximate intent rather than measure it. And proxies are very good at generalising.

Return visits to the same category could mean active consideration. It could also mean idle browsing, research with no near-term purchase plan, or a customer who's decided not to buy and is still processing why. Two customers can leave identical behavioural footprints with completely different intent trajectories.

The distinction that matters here is between an intent proxy and an intent prediction.

A proxy uses a single observable signal. So, page views, return visits, time on site, as a way to characterise specific behaviours.

A prediction uses hundreds of behavioural signals together. Things like scroll patterns, hesitation, comparison behaviour, click timing, revisit frequency, and models the probability that a specific visitor, right now, is likely to purchase.

That difference matters for email specifically because your flows are triggered on schedule, not on signal.

Consider a customer in your 90-day lapsed segment. Their segment says win-back flow, probably with a discount. But a real-time intent prediction across their session behaviour might tell a different story: they returned twice this week, built strong product affinity for outerwear without adding to cart — affinity that was visible well before any CTA click — and their intent has been rising, not falling, across the session.

That isn't a lapsed customer who needs persuading. Instead, it's someone ready to buy. They don't need a discount at all.

If you're offering discounts to people like that, all you're doing is eroding your margin. And let's just play with some numbers for a second. If you have a £120 average order value, a 20% discount on sale you would've made anyway costs you £24 pounds for that one sale. But it also costs for every single customer you've offered the same discount to.

Your email strategy can see who someone is and what they've done before. It can't see the modelled probability that they will purchase today. That requires a different layer of data entirely. (Spoiler: It's Made With Intent)

How to close the gap between your CRM and what's happening on-site

There are three ways to close it, and it's worth being honest about what each one costs.

1. Layer session behaviour onto send triggers — trigger sends based on a real-time on-site event rather than a profile segment on a schedule. If your ESP is Klaviyo, the ActiveOnSite flow trigger gets you closer. You're still dependent on session-entry events rather than continuous in-session behavioural signals, but it's a step in the right direction.

2. Prioritise timeliness over segment precision — tighten the timing of your event-based flows. Klaviyo's 2024 abandoned cart benchmark report, covering more than 143,000 flows, recommends sending the first recovery email within 2-4 hours of abandonment. Most brands set this window far wider. The window of peak intent closes faster than most email schedules assume.

3. Use intent-based scoring to qualify your existing segments — add a signal layer on top of your CRM segments that identifies which subscribers are currently showing on-site behaviour indicating an active buying moment. Platforms like Made with Intent analyse hundreds of behavioural signals in real time: return visit frequency, product page depth, comparison behaviour, session patterns. The result is a send informed by both who the customer is and where they are right now.

What this means for how you build segments

Audit your current flows against three questions:

• What is the trigger? Profile rule, onsite event or intent prediction?
• Does it rely on open rate engagement? If built before 2021, review whether click/conversion data could replace open rate as the proxy.‍
• Is it event-led? Tighten the send window to match the actual window of intent.

The question your segments can't answer

The most commercially important question: is this person ready to buy right now, isn't answered by a session behaviour alone. It's answered by a prediction built across hundreds of behavioural signals in that session, continuously updated as the customer moves.

That's not something a segment can produce. And it's not something a single observable proxy can substitute for.

Want to learn more about intent-based selling? Grab yourself a demo.

Social proof is everywhere. And that’s the problem.

Most brands run it sitewide, triggered on page-load to each and every visitors. It works brilliantly for some visitors, but damages conversion for others. Early urgency messages can cause anxiety and exits, especially for visitors who have yet to show any intent to purchase.

We covered the status quo of social proof already, but for a quick recap, if you only measure generic conversion rate uplift, you see the benefit for those it helps but miss the hidden downside for those it turns off. That’s why the first step is to rethink why you’re using it at all.

Remind yourself why you’re using social proof

Why are you really doing social proof? It may start off as a best practice, that is low-hanging fruit to “increase conversion rate” or “drive more revenue”, but that’s only half the story. Social Proof is ultimately about encouraging a certain behaviour from an individual by using the influence that the actions, choices or approvals of others have on them. And when you apply the message to everyone, all the time, you’re missing the real opportunity - and ultimately playing conversion roulette.

Urgency-style proof like “X people bought this today” might give high-intent visitors the final nudge, but it can spook a casual browser into leaving. Without understanding these nuances of an experience (by splitting results by audience and keeping a control group for each stage), you’ll never see the drop hiding inside your averages. Which brings us to the next question: who exactly are you showing it to?

Rethink who you’re delivering it to

Do all your visitors get the same message at the same time? They shouldn’t. High-intent visitors close to purchase are often persuaded by urgency. That same message can make a low-intent browser feel pushed and leave. In fact, urgency messaging can decrease conversion for low-intent visitors by 4–5%.

Instead, Bestseller messaging can help low-intent users refine their choices by pointing them to popular items. But it can also distract a high-intent shopper who’s already found what they want, just like showing unrelated recommendations in checkout can derail the final purchase. The key is knowing which messages work for which segments — and that’s where a more targeted, step-by-step approach comes in.

Step-by-step: How to get started with intent-based social proof

1: Analyse or test across intent stages

Start with what you’re running today, but split results by low, building, and high intent. Maintain a control group for each stage to compare “no message” against “message.” This is where the surprises appear. Urgency might lift high-intent conversion by 10% but drop low-intent by 4–5%.

2: Exclude the unsuited audiences

If urgency is scaring off browsers, remove it for those segments. Replace it with different messaging styles that fit their stage. Bestseller or top-rated messages can help low-intent visitors explore, while reassurance works better for those already on the brink of purchase. For segments where the original Social Proof didn’t work, consider testing alternative messages entirely.

3: Layer in real-time signals

Static triggers are blunt. Use signals like a drop in purchase confidence or increase in a visitor’s likelihood to abandon to time your messaging more precisely. For example, one jewellery brand predicted basket backtracking and swapped urgency for reassurance style Social Proof. This single change lifted conversion for that segment by 25%.

4: Bring in affinities and more

Make it personal. If someone’s deep into a specific brand or category, show them messaging that reflects it. “Custom rings designed this month” speaks to an engagement ring shopper. In a multi-brand store, tie the experience to brand loyalty, like “People who love [Brand X] also love this.”

5: Iterate and evolve

For segments without strong affinities, use it as a discovery tool rather than reassurance. Keep testing new messages for each stage, refining your targeting, and improving your timing. Social proof should get sharper over time, not sit as a static, set-and-forget feature.

A real-world example from a jewellery & watch retailer

A premium UK jewellery and watch retailer faced a familiar problem. Shoppers were reaching the basket, then backing out to browse again or revisit product pages. For higher-value, considered purchases, this hesitation was a sign of uncertainty. The team realised not every basket visitor needed urgency, some needed reassurance.

Using intent signals, they built a segment of visitors showing backtracking behaviour and delivered targeted in-basket reassurance, highlighting flexible delivery and secure payment options. This message appeared only to those who needed it, avoiding unnecessary noise for confident buyers. The result was a 25% uplift in conversion for that segment.

This example shows how the right message at the right moment can have a big impact, which leads to the bigger picture of what happens when you get intent-based social proof right. You can read the full customer story here.

The impact of using intent in social proof

When intent guides your targeting and timing, you keep the uplift without the hidden drop-offs. Better timing can amplify gains, and removing harmful triggers boosts coverage. Over time, social proof becomes a natural part of the journey, helping with discovery, reassurance, and evaluation, instead of a blunt, one-size-fits-all tactic.

Social proof works when it works for the right people. Intent makes that possible. Want to see what that could look like for your brand? Book a demo.

Social proof should be one of the most powerful tools in ecommerce. At its core, it’s the influence that the actions, choices or approvals of others have on an individual’s behaviour.

People look to others when they’re uncertain about what to choose, who to trust, or whether to act. In ecommerce, that influence can appear anywhere in the journey. As reassurance that a brand is worth buying from, or as urgency to act before missing out.

It comes in many formats: scarcity messages (“only 3 left”), activity indicators (add to baskets, recent views, recent purchases), reviews and ratings, and trending or bestseller labels. Used well, these cues can reassure, create urgency, and help people find what’s popular or trusted.

The problem is, social proof has become one of the most overused and underthought tactics in the game. It’s often deployed as a blanket message to everyone, with little thought about whether it fits their mindset or the brand experience.

Retailers love it because it’s quick to turn on and almost always delivers an aggregate uplift. But those uplifts are often driven by a smaller group, and the negative effects on others are hidden in the averages.

The status quo of social proof

Most ecommerce teams apply it generically, showing the same messages to everyone – often on every product page. The most common use is as a conversion-driving technique late in the journey, but there’s a growing trend to apply it earlier in discovery (e.g., “bestseller” on PLPs).

Its popularity comes from being considered “best practice,” easy vendor implementation, and the reliable ROI it shows on aggregate. But those aggregate numbers are disproportionately influenced by high-intent visitors, which hides the harm it can cause to others.

What works for one mindset can actively put another off. As part of our research for The Intent Gap Report, we found:

• “Trending” overlays on PLPs positively impact low-intent browsers.
• “X sold last week” overlays on checkout pages deliver an average +5% conversion lift for high-intent visitors but cause a -1% drop for low-intent visitors.

Luxury and exclusivity-driven brands often avoid generic social proof entirely. In high-consideration categories, it can feel out of place – an engagement ring buyer doesn’t want to hear that “20 others bought this today,” and a £3000 jacket doesn’t need a flashing urgency tag over carefully curated imagery. In these cases, overlays can jar with the brand and undermine the premium feel.

When social proof is everywhere, it stops providing reassurance or focus. The message becomes noise, prompting the question: why stick with this approach?

Because most retailers rely on page-type triggers (e.g., PDP = ready to buy). But many PDP visitors are still browsing. Without behavioural context, tactics are based on where someone is, not how they’re behaving. That one-size-fits-all approach ignores timing and mindset. And that’s exactly why it needs a rethink.

Social proof with intent

Social proof can reassure early in the journey or create urgency later, but timing and fit are critical. Softer cues like “bestseller” or “trending” help those still discovering products. Urgency or scarcity works best when someone has decided what they want and just needs a final nudge. Use it too soon, and it risks creating anxiety or distraction.

Think of walking into a DIY store paint aisle: if you’re browsing, you don’t want someone saying, “Only three tins left – buy now!” before you’ve chosen a colour. But if you’re holding the exact tin you want, that message might spur you to buy. The same logic applies online.

Or picture a luxury sales assistant with a £3000 jacket. They wouldn’t start with “20 people bought this today.” They’d focus on its quality, heritage, or popular combinations, tailoring the message to the moment.

Real-time intent signals mean you can:

• Show discovery-style social proof to those exploring
• Reserve urgency and scarcity for visitors with strong product interest or signs of hesitation
• Avoid showing it altogether to those it might deter

When you match the message to the moment, social proof stops being background noise and starts driving action.

The path to better social proof

While we’ll cover how to move from generic application to something more intent-based in a follow up, the core steps are:

1. Analyse performance by visitor mindset, not just aggregate.
2. Exclude audiences where a message harms conversion.
3. Adapt style and timing to fit both brand tone and visitor context.

The benefits? Higher incremental gains, reduced brand risk, and interactions that build trust.

Social proof works – but not for everyone, not everywhere, and not all the time. The more you align it with intent, the more it delivers.

Ready to deliver social proof that meets the moment? Discover how Feature Delivery with Intent works.

Abandoned cart emails became a key part of the CRM playbook for a reason. They’re easy to set up, look great in reports and are seen as a no-brainer for driving conversions. But let’s be honest. Most of them are blunt. They ignore why shoppers abandoned in the first place and often end up adding noise instead of value for the visitor.

In a recent piece, Rethinking abandonment emails with intent, we explored why this tactic so often falls short. The reach is limited to visitors you can actually email. The timing often misses the moment. And blanket discounts don’t just erode margin, they train shoppers to delay purchases.

If you haven’t read that yet, it’s worth a look. But this article is about moving forward. Here’s how CRM teams can use use intent with abandonment emails to make them smarter, more targeted and more effective.

Review why you are sending abandonment emails (and who to)

It’s easy to assume the goal of abandonment emails is simple: recover a lost sale. But was it a lost sale to begin with? Just having items in a cart isn’t always a signal of high purchase intent. Shoppers use them to shortlist products, compare options or as a save-for-later tool.

If the goal is to recover real opportunities, this tactic needs refining.

Not every abandoner should get an email. Without context, CRM teams risk:

• Sending to shoppers who are still browsing and not yet ready to buy.
• Triggering emails too soon or too late.
• Flooding inboxes with irrelevant reminders.

The consequences? Unsubscribes. Inbox fatigue. And lost trust. Excluding shoppers who aren’t ready to buy isn’t only a better experience. It also leaves room for emails that actually work, allowing you to send them with impact.

If someone’s adding to cart to compare or wishlist items, hitting them with a salesy reminder could risk turning them off completely.

So how do you make abandoned cart emails smarter, more targeted and more effective? Start here.

Optimising abandonment emails with intent

This isn’t about rebuilding from scratch. It’s about fixing the foundations first, reducing downside and then optimising for growth.

Step 1: Analyse

Start by reviewing your current campaigns with intent data. Segment visitors by mindset and product affinity. Understand which groups engage and which ones churn. Look beyond standard metrics like open rates. Ask who clicked, who converted and, crucially, who unsubscribed.

Step 2: Exclude

Stop sending to low-intent visitors. Protect your list health by cutting out disengaged shoppers who are unlikely to convert. Focus efforts where they’ll actually move the needle.

Step 3: Improve

Optimise the emails you do send to high-intent visitors. This isn’t just about tweaking subject lines. Think about mindset. If a shopper is in discovery mode, avoid hard-sell copy. Instead, highlight educational content, social proof or unique selling points to build confidence. For those showing strong purchase intent, timely nudges and delivery reassurance might be all they need to convert.

One area many retailers get wrong is discounting. Blanket incentives erode margin and train shoppers to wait. Instead, reserve offers for visitors showing clear signs of hesitation, with an unlikelihood to return to site.

Grow with better emails and beyond the inbox

Once you’ve reduced downside and optimised for impact, you’re ready to grow further.

In email, you can tailor creative based on intent stage and affinities. Think beyond discounts or nurture flows. You can even explore dynamic recommendations for basket builders and cross-sell opportunities. This isn’t about flooding them with options but about making the right product feel obvious.

One of our customers, a leading UK jeweller, faced this challenge head on. Their CRM team realised their “one-size-fits-all” abandonment email was limiting relevance and risking engagement. They started with our exclusion and optimisation steps, then moved onto more context-driven creative.

They created three visitor groups: low, building and high intent. Each group received tailored emails. High-intent abandoners got a timely, persuasive message tied to browsed products. Lower-intent visitors were sent softer campaigns focused on brand USPs. Some received no email at all to protect list health.

The result? A 12% uplift in click-through rates and a strategy that felt more like a conversation than a conversion ploy. Read the full customer story here.

But your response doesn’t have to stop at the inbox.

Onsite, once you detect exit signals in real time, you can trigger supportive nudges before visitors abandon, such as delivery reassurance or save-for-later prompts. You can also surface email capture prompts for unknown visitors at the right moment to grow your contactable base.

When onsite and email journeys are connected, you’re no longer chasing abandoners after they’ve gone. You’re helping them complete the journey in the moment.

The impact when you get this right

When you rethink abandonment emails with intent, shoppers feel understood instead of pestered. CRM teams send fewer, smarter emails that actually drive revenue.

Metrics improve across the board too:

• Unsubscribe rates drop dramatically due to less inappropriate emails
• Click-through rates (CTR) climb as relevance improves
• Higher return visits and positive movement on intent to return metrics.
• On average, Made With Intent users see a 65 percent increase in campaign impact overall

This isn’t just about improving KPIs. It’s about changing how shoppers feel when they hear from you. Intent-based emails create relevance, reduce noise and rebuild trust. They don’t just recover sales. They set the stage for long-term growth.

Want to see how intent-first abandonment emails work in practice? Get a demo to learn how Made With Intent helps teams recover more revenue without damaging their shoppers’ experience.

Abandoned cart emails are often seen as the gold standard for CRM success. They’re easy to set up, look great in reports, and are widely viewed as a no-brainer for driving conversions. But here’s the uncomfortable truth: they’re not the silver bullet we’ve been treating them as.

Most retailers rely on them as a core tactic. Yet the reality is they only reach a small fraction of abandoners. These are the people you’ve identified and secured permission to email. Even when these emails land in a shopper’s inbox, the moment has often passed. It’s like walking out of a store and having the assistant chase you down the high street an hour later. That window to influence the decision has already closed.

To make matters worse, customers have learned how to game the system. Many now abandon carts deliberately to trigger a discount code. According to our Intent Gap research, 83% of online shoppers have used a discount code even when they were ready to pay full price. That’s margin erosion, but also proof that current approaches are blunt, and shoppers know how to exploit them.

It’s time to rethink how we handle abandonment. And it starts with the emails themselves.

The status quo: Abandonment emails as the default fix

Abandoned cart emails feel like an easy win: a shopper adds something to their cart, leaves, and a templated flow comes to the rescue. Subject lines like “Forgot something?” or “Your cart misses you” flood inboxes, often paired with a discount to lure the customer back.

On the surface, these campaigns perform well. High open rates. Strong click-throughs. Solid ROI. But let’s not kid ourselves: those metrics don’t tell the whole story.

• Limited reach: Only a fraction of abandoners are identifiable and contactable.
• Delayed timing: By the time the email lands, the shopper’s attention has moved on. Or worse, they’ve bought from a competitor.
• Added friction: Unless you’ve captured an email address and marketing consent, most visitors are already out of reach.
• Predictable patterns: Shoppers now anticipate these emails and wait for discounts.
• Generic messaging: Emails rarely account for why someone abandoned in the first place.

If we’re honest, these emails are less of a personalised recovery tactic and more of a reactive safety net. And safety nets don’t work for everyone.

The Problem: Why they fall short

There’s no denying abandoned cart emails deliver some results. But they’re flawed:

• Low impact at scale: Most shoppers won’t even see one. No email means no campaign.
• Lack of context: “You left something behind” doesn’t consider intent. Were they comparing prices? Still browsing? Waiting for payday?
• Delay kills momentum: The longer you wait, the colder the lead gets. What felt relevant in the moment quickly becomes noise.
• Margin drain: Blanket discounts train customers to delay purchases and wait for incentives.

These emails aren’t inherently bad. But in their current form, they’re blunt and reactive. They’re also increasingly easy for shoppers to tune out or exploit.

The Reframe: Fix the email, then think bigger

We don’t need to throw out abandoned cart emails. But we do need to evolve them.

Start by making them smarter:

• Segment for context: A high-intent abandoner may only need reassurance. A low-intent visitor might require education or a compelling USP.
• Time with care: Not every shopper needs a follow-up within an hour. Some need space.
• Rethink the content: Shift from discount-first to value-first messaging. Highlight free returns, flexible payments, or social proof instead.

This isn’t theoretical. One UK high-street jeweller used intent data to personalise abandonment emails, tailoring content and timing to match each visitor’s mindset. The result? A 12% uplift in click-through rates and a strategy that felt more like a conversation than a conversion ploy. Read the full story here.

But even the smartest emails have their limits. If we know when and why a shopper is about to abandon, why wait until they’ve left to act?

Every abandonment email is, by definition, too late. The shopper has already gone. That’s why leading retailers are complementing smarter emails with in-session interventions.

With real-time intent data, you can:

• Detect when a shopper is hesitating in the cart.
• Surface supportive messaging before they leave (e.g., save-for-later prompts or delivery reassurance).
• Reserve discounts for visitors showing exit signals, rather than everyone.

This approach doesn’t just recover abandoners; it prevents abandonment in the first place. And because interventions happen in the moment, they feel like help rather than a hard sell.

Future Vision: Abandonment reimagined

Abandoned cart emails still have their place. But they’re no longer enough on their own.

The smarter play combines:

• Smarter emails: Contextual, well-timed, and less reliant on discounts.
• In-session interventions: Adaptive experiences that engage all abandoners, not just the small percentage you can email.

It’s a shift from generic flows to contextual journeys. From chasing abandoners to understanding them. From reactive tactics to proactive engagement.

And when you get this right, abandonment isn’t just reduced. It’s transformed.

Ready to rethink your abandonment tactics? Learn how intent makes abandonment emails more impactful or read our article on getting started with intent-based abandonment.

Ask most ecommerce teams how they know what a visitor cares about, and you'll hear the same answers. Last product viewed. Most time spent. Recent purchases.

These are proxies. Not signals. Not intent. Not interest.

And yet, this is how most of the industry claims to "know" what their customers are interested in.

The reality? Ecommerce has spent a decade optimising for what people click, not what they care about. The assumption that engagement equals interest is the core flaw. Affinities should be a core capability in ecommerce but they’re largely missing, and worse, often faked with inaccurate signals.

This article unpacks what affinities really are, why current methods fall short, and how Made With Intent’s approach reframes what personalisation should actually mean.

The illusion: Mistaking engagement for affinity

Most common drivers for determining product affinity strategies:

• Last viewed
• Most viewed
• Longest viewed
• Previously purchased

This leads to wildy inconsistent outcomes and is essentially guesswork. For example, see the following two sessions:

‍

[Session 1]
───────────────────────────────────────────────────────────────────────
Product A → Product B → Product C → Product D → Product E  
              ↑                      ↑          
        (Longest Viewed)     (Previously Purchased)    

‍

[Session 2]
───────────────────────────────────────────────────────────────────────
Product F → Product G → Product B → Product H → Product I
                            ↑                     ↑
                       (Most Viewed)        (Last Viewed)


‍

‍

The problems with these interpretations:

1. Last viewed ≠ Highest intent: Product I was viewed last, but only once and briefly. It’s a poor indicator of interest or conversion potential.‍
2. Most viewed = Curiosity, not commitment: Product B’s repeated views may reflect uncertainty, not preference. It could also be a comparison reference or an accidental revisit.‍
3. Longest viewed can be misleading: Product C was dwelled on, but that could reflect confusion, poor UX, or open-tab idling, not genuine interest.‍
4. PreviouspPurchase ≠ future intent: Just because Product D was purchased before doesn’t mean the user wants it again. Relevance might now be low.

Ultimately, engagement is a misleading approach as it works in both directions and remains open to interpretation.

Yet this is how almost every ecommerce platform infers "what a customer cares about." Here’s why I think this fails:

• Recency bias fools the system. Someone can hate-scroll a product page and look "interested" when they aren't.
• No context of intent. Clicking or viewing does not equal liking. Hovering does not equal wanting.
• Secondary behaviour pollutes the data. A shopper adding toothpaste after buying a fragrance does not mean they love toothpaste.
• Teams aren't even aligned. CRM, paid media, and onsite teams all use different definitions of "interest," based on whichever proxy suits their tool or process.

And that’s the core of the issue. What most ecommerce teams call 'affinity' is nothing more than an engagement proxy. It’s recency. It’s frequency. It’s volume. But it’s not interest. And it’s definitely not intent.

To be fair, it’s not like the industry ever had this easy. Affinity has never really been an out-of-the-box capability for ecommerce teams.

You could try to cobble it together by blending last viewed, most viewed, time spent, but it meant building custom rules, manually interpreting engagement and hoping it told the right story. Most teams never had the tools to move beyond that.

And even when teams do try to build affinity models themselves, it rarely scales. Every time you want to understand affinity for a new attribute, whether it’s price, brand, category or anything, you’re forced to define rules, retrain models, or manually stitch data together.

The result? A fragile process that breaks the moment something changes. That’s why most teams default back to blunt proxies like recency. They’re simple and work ‘well enough', even if they’re wrong.

The low ceiling of engagement proxies

Let’s be clear. This stuff does work. Kind of.

Last viewed is better than nothing. Most viewed does something. This is why the industry keeps doing it.

But it's a ceiling, not a scalable solution.

It's effective, but not to the same degree. You're essentially marking your own homework.

The real opportunity isn't about fixing something broken. It's about lifting the ceiling entirely.

Less noise and cleaner signals. More precise targeting without over-discounting or over-messaging. Alignment across teams instead of different, conflicting definitions of “interest".

That’s why we define an affinity not just on what a visitor looked at, but on what contributed to their intent.

If a visitor browses three pairs of shoes at different price points, the traditional model might recommend the one they spent the most time on. Our model identifies which of those shoes actually built purchase intent. Because time spent isn't the same as value contributed.

Imagine visiting a health and beauty store. You spend five minutes looking at shampoo, toothpaste, and a razor. But the real reason you came in was for a fragrance, you checked that out first and decided quickly. Then you browsed around for other products.

The typical ecommerce system thinks you're deeply passionate about toothpaste. Ours knows the fragrance mattered most.

Why actual affinity data matters to online retailers

The real power here is prioritisation. When you use affinity based on contribution to intent, you stop drowning in noisy data. You can weight engagement by what actually mattered. What contributed. What moved someone forward. Not just what they clicked.

This isn’t just about more data. It’s about clarity. About knowing which signals matter—and which are just noise.

Getting this right isn't just about better product recommendations. It’s about:

• Cleaner data on what your visitors actually care about.
• More appropriate personalisation. Less irrelevant spam.
• Consistent messaging across CRM, onsite, and paid.

It’s also about unlocking higher-margin tactics. Look at how Seasalt Cornwall applied affinity data to drive an 89 percent conversion uplift with affinity-based discounts. Or how they increased conversion by 8 percent with homepage personalisation.

Both use cases speak to one truth: when you understand what people care about, you sell better. And you sell smarter.

‍

I believe in the (near) future, intent-based affinities will be the new baseline. A foundation for any business serious about personalising at scale.

When paired with real-time intent, it unlocks a fundamentally more appropriate, more effective way of serving visitors.

In five years, retailers will look back at recency-driven personalisation the way we now look at irrelevant banner ads or spammy pop-ups. Crude. Inappropriate. Obsolete. Affinity without intent will feel as outdated as demographic targeting does today.

If your tools can’t tell you what a visitor truly cares about, not just what they clicked, then you're flying blind.

If you're not using real-time affinity and intent signals, you're not personalising. You're approximating.

This is the next evolution of ecommerce. And it's already happening. Take a look at Made With Intent if you don’t believe me.