Sailthru Personalization: The Data-Driven Email Strategy for Retail

Sailthru was built around a specific thesis: that the most effective email marketing is not about sending better campaigns to large segments — it’s about building an accurate model of each individual’s interests and serving content and products that match that model precisely.

For retail and media brands with the data and traffic to feed this model, Sailthru’s personalisation technology produces measurably better results than segment-based approaches. This guide explains how the personalisation technology works, what it requires to function at its best, and how to implement it effectively.

Sailthru’s Horizon Personalisation Technology

Sailthru’s core personalisation engine is called Horizon. It builds individual interest profiles for every user by observing their behaviour across channels — email, web, mobile, and app. Over time, these profiles become increasingly accurate models of what each individual is interested in.

Horizon operates continuously. It does not take a daily snapshot; it updates in real time as users interact with content and products. Every email open, click, website visit, product view, and purchase refines the individual’s interest profile.

The profile data Horizon builds includes:

• Interest scores by category or topic: How interested is this individual in each content category, product category, or brand? Scores are computed from engagement signals weighted by recency, frequency, and depth of engagement.
• Engagement timing: When is this individual most likely to engage with an email? Sailthru uses this to power send-time optimisation, called Sailthru Smart Send.
• Content affinity: For media companies, Sailthru tracks which types of content (editorial topics, content formats, authors) each user engages with most.
• Product affinity: For retail brands, which product types, price ranges, brands, and styles does this individual engage with?

This data is used to power personalised email content — specifically, the product recommendation and content recommendation blocks that are Sailthru’s signature feature.

Interest Scoring: How It Works in Practice

Interest scoring is the mechanism by which Sailthru translates user behaviour into ranked preferences. For a retail brand, every product category, brand, and style in the catalogue receives a score for each user based on their interactions.

The score is influenced by:

• Recency: Recent interactions carry more weight than older ones. A category someone browsed yesterday is scored higher than one they browsed three months ago.
• Depth: A purchase carries far more weight than a page view. Sailthru’s default weighting assigns the highest score lift to purchases, then cart additions, then product page views, then category page views.
• Frequency: Repeated interactions compound. Someone who has browsed the same category five times signals stronger interest than someone who browsed it once.

For email personalisation, interest scores determine which products or content appear in a user’s personalised recommendation block. The product shown first is the one Sailthru predicts is most likely to drive a click from that specific individual.

Building Individual Profile Data

Sailthru’s personalisation is only as good as the data flowing into individual profiles. Profile quality depends on two things: data breadth (enough signal types to build an accurate model) and data volume (enough events to move beyond noise into signal).

Web Tracking

Sailthru’s JavaScript tag should be installed across your entire website — not just the email click landing pages, but product pages, category pages, search results pages, and checkout. Every page view and interaction feeds the profile.

For authenticated users (logged-in account holders), Sailthru can link web behaviour to the email profile directly. For anonymous visitors, Sailthru uses a cookie-based approach that links to an email profile when the user opens or clicks an email in the same browser session.

Purchase Event Integration

Purchase data should flow to Sailthru as quickly as possible after a transaction. This typically happens via Sailthru’s API at the point of purchase confirmation. Each purchase event should include:

• SKU, product name, product category, and brand
• Price paid
• Sailthru user ID or email address of the purchaser

Purchase events are the highest-value signal for interest scoring. Brands that don’t pass purchase data to Sailthru miss the most powerful input to the personalisation model.

Email Engagement Data

Email opens and clicks are automatically captured by Sailthru as part of its normal tracking. These events feed back into the interest profile — a click on a specific product category in an email increases that category’s interest score.

Personalised Product Recommendations in Email

Once individual profiles are built, the most visible application is personalised product recommendation blocks in email. These blocks pull content dynamically from the product catalogue, filtered and ranked by the individual’s interest profile.

In Sailthru’s email editor (or via API-driven sends), a recommendation block is configured with:

• Recommendation type: What algorithm drives the selection (interest-based, similar items, trending in category, editorial picks)
• Filtering rules: Product availability filters, category constraints, price range limits, and exclusion rules (don’t recommend items the user has already purchased)
• Layout and design: How many items to show, image/text format, button style

When the email is sent, Sailthru renders each recipient’s recommendation block with the products most relevant to them individually. Recipient A sees the hiking boots and outdoor gear that match their browse history. Recipient B sees the workwear and accessories their profile predicts they’re interested in. Same email template, completely different product selection.

Open-Time Rendering

Sailthru supports open-time email rendering for recommendation blocks, meaning the recommendations are generated fresh when the email is opened rather than when it’s sent. This is valuable because it means:

• Product availability and pricing are current at the moment of opening
• If a user has made a purchase between when the email was sent and when they open it, that purchase is factored into the recommendations
• Long campaigns (sent over a multi-day window) don’t serve stale recommendations to late openers

Content Personalisation for Media Companies

Sailthru has a particularly strong install base in media and publishing, where content personalisation is the core use case rather than product recommendations.

For media companies, Sailthru’s Horizon technology builds interest profiles based on the content each user reads — topics, authors, content formats, and publication sections. These profiles power personalised daily or weekly email digests where each subscriber receives a curated selection of articles matched to their demonstrated interests.

The typical implementation:

1. Every article published is tagged with metadata (topic, author, section, content type) and pushed to Sailthru’s content library via API
2. Sailthru tracks which articles each subscriber has engaged with (via email click or web page view)
3. The personalised digest email pulls from the content library, ranking articles by the match between article metadata and the individual subscriber’s interest profile
4. Each subscriber’s digest looks different — different articles, potentially different sections highlighted, ordered by predicted interest

Media companies implementing Sailthru content personalisation typically report significant lifts in click-through rate and content consumption compared to editorially-curated, one-size-fits-all newsletters.

Data Requirements for Personalisation at Scale

Sailthru’s personalisation performs best when individual profiles have sufficient data to model accurately. New subscribers — or low-engagement users — have sparse profiles, and the recommendations for these users are less accurate than for established, active users.

Strategies for handling sparse profiles:

Category-level fallbacks. Configure recommendation blocks to fall back to trending items within a user’s highest-interest category when individual-level personalisation confidence is low.

Explicit preference collection. A preference centre or onboarding survey that asks new subscribers to indicate their interests populates the profile with explicit signals before implicit behavioural data accumulates. These explicit preferences can be stored in Sailthru and used to seed the Horizon profile.

Engagement-triggered profile enrichment. As new subscribers engage with emails and the website, their profiles build. An automated welcome sequence designed to surface a range of category content helps Sailthru identify interests faster.

---

Sailthru’s personalisation technology is among the most sophisticated available to retail and media brands. When properly implemented — with web tracking, purchase data integration, and well-structured product or content metadata — it produces email experiences that are genuinely individualised and measurably more effective than segment-based personalisation.

At Excelohunt, we help brands get the most from Sailthru’s personalisation capabilities — from data integration architecture to recommendation block implementation and performance analysis. If you want to understand what Sailthru personalisation could deliver for your programme, we can show you.

---

Related Excelohunt Services

Looking to implement these strategies with expert support?

• Sailthru — learn how we implement this for clients Get a free audit →