Customer Targeting 101: Combining Generic and Personalized Content at Scale

Marketers increasingly want to deliver personalized messaging to customers. But personalizing CRM communications at scale is genuinely challenging. The key ingredient to successful personalization is ensuring that the strategic goal of each campaign aligns with the targeting method used.

By layering the three forms of targeting below, marketers can deliver relevant, tailored content at scale. This framework was originally developed for email marketing but applies to any form of below-the-line or direct marketing.

1. Rules-based targeting

2. Propensity-based targeting

3. Event-based targeting

1. Rules-Based Targeting

Rules-based targeting leverages traditional CRM segments to reach a large base of customers. It's ideal for standard, business-as-usual (BAU) campaigns — but even simple segmentation will consistently outperform "spray and pray" generalized sends.

Example: A fashion e-commerce business announces its Fall line to all customers, but tailors the creative and copy based on the categories each user has previously purchased.

• Type of segmentation: Business rule or clustered segments that exist within the CRM. Target 4–10 unique segments when using this method.

• How to operationalize: Create template creatives and copy (email, direct mail) that are used on a regular basis, tailoring content and/or offers for each target segment. Most email service providers (ESPs) allow segments to be built natively, or uploaded automatically from a database or FTP.

• Share of sends: This type of targeting should generally account for 40%–80% of sends in a given week. This will naturally vary based on the extent to which the other two methods are employed.

Tip: Rules-based targeting is the foundation. Get this right first — clean segments, clear definitions, tested creative templates — before investing in more sophisticated methods.

2. Propensity-Based Targeting

Propensity models are regression models that determine the likelihood that a particular event will occur — for example, the likelihood that a customer will make a purchase, become a high-value customer, or churn. This form of targeting traditionally falls within the domain of a data scientist, but with third-party tools or technically-savvy analysts, businesses can leverage propensity targeting without a dedicated data science team.

• Type of segmentation: Customer scoring — using regressions to predict the likelihood of a customer doing something (logistic regression) or the degree to which they do it (linear regression). Once models are built, data scientists use the model coefficients to score customers on a recurring basis.

• How to operationalize: Think of propensity targeting as situational targeting — triggered when a customer is on the verge of a lifecycle or behavioral change. Scores are uploaded to the ESP, and only users within a pre-specified score range receive the communication. These emails typically include a promotion or incentive, since the goal is to encourage or reverse a change in customer behavior.

• Share of sends: Targeting 10%–20% of the user base is a reasonable baseline. This form of targeting should have higher priority than rules-based targeting when a customer qualifies for both.

Tip: Propensity models are most powerful when rerun regularly (weekly or monthly). A model trained six months ago on stale data will underperform — keep scoring fresh.

3. Event-Based Targeting

Also known as real-time targeting, this tactic has become a staple in e-commerce and online businesses. It is triggered when a customer takes a specific action: makes a purchase, abandons a basket, registers an account. Highly effective — but it requires technical sophistication to execute well.

• Type of segmentation: This isn't traditional segmentation. Instead, you are tracking users' actions in real-time and targeting them based on those actions. Marketers must decide which actions to track and what communication will be triggered by each. Developers — or third-party platforms like Criteo — can flag these events and deliver communications in real-time.

• How to operationalize: This method combines the rule-based logic of segment targeting with the behavioral precision of propensity targeting. The technical execution is distinct: JavaScript tracking captures on-site behavior, and APIs communicate the triggering event to the ESP, which automatically initiates the send. In general, API-based execution is faster and more reliable than ingesting and processing behavioral data through a database.

• Share of sends: Because this targeting is largely automated (immediate communication is usually required), it's difficult to predict volume precisely. A rough range is 1%–15% of the user base, depending on the number of tracked events. Managing communication frequency — across all three targeting types — is critical.

Final Thoughts

Prioritization. These targeting methods will run concurrently, so prioritization matters. When a customer qualifies for all three simultaneously, event-based and propensity-based targeting should take precedence over rules-based. More granular targeting is expected to outperform generic content.

Testing. Implement a testing framework alongside each targeting method. Rigorous testing will reveal the optimal frequency, offer type, and creative approach for each — and compound returns over time.

Execution first. Modeling and data manipulation aren't the hardest parts of email execution. The harder challenge is building a streamlined executional framework that (a) works consistently, (b) delivers a good customer experience, and (c) can be measured and optimized. Always prioritize execution and UX over analytical sophistication.

The Data Strategist helps startups and scaleups become data-driven. Get a data scientist on-demand, or advice on building your analytical data stack.