What Is Recency in Marketing?
Recency measures how recently a customer completed a target action, most commonly a purchase. In direct marketing and CRM strategy, recency is the “R” in the RFM model (Recency, Frequency, Monetary value). It consistently ranks as the single strongest predictor of future purchase behavior. A customer who bought last week is far more likely to buy again than one who bought 18 months ago, regardless of how much they spent.
Why Recency Outperforms Other Predictors
Database marketing research going back to the 1960s confirms that recency carries more predictive weight than frequency or spend. Direct mail analyst Arthur Hughes’s work was foundational in documenting this pattern. The logic is behavioral: recent buyers have demonstrated current intent, their contact information is likely still valid, and the brand is still in their consideration set.
Amazon’s recommendation engine reflects this principle. When a customer purchases a product, Amazon’s follow-up emails arrive within 24 to 72 hours, targeting the window when purchase intent is still high. The company has publicly credited personalized, recency-triggered follow-ups as a contributor to its cross-sell revenue. That cross-sell revenue accounted for roughly 35% of total sales as of recent earnings reports.
Recency Scoring and the RFM Framework
Marketers quantify recency by calculating the number of days since a customer’s last action and assigning a score. Most implementations use a 1 to 5 scale, where 5 represents the most recent customers.
Standard Recency Scoring Formula
| Recency Score | Days Since Last Purchase | Segment Label |
|---|---|---|
| 5 | 0 to 30 days | Active |
| 4 | 31 to 60 days | Recent |
| 3 | 61 to 90 days | Lapsing |
| 2 | 91 to 180 days | At-Risk |
| 1 | 181+ days | Dormant |
The thresholds vary by industry. A grocery brand might define “active” as the past 7 days; a luxury furniture retailer might set that window at 90 days given its longer purchase cycle. Thresholds should reflect the brand’s median purchase interval rather than arbitrary calendar periods.
Calculating a Recency Score
The most common method is percentile ranking across the customer file:
Formula: Assign the top 20% most recent buyers a score of 5, the next 20% a score of 4, and so on down to the bottom 20% (longest inactive) receiving a score of 1. This creates equal-sized recency quintiles and adjusts automatically as the customer base grows.
Example: A brand with 100,000 active customers sorts them by days since last purchase. The 20,000 customers who purchased most recently receive an R score of 5. Brands then prioritize these customers for new product launches, loyalty offers, and upsell campaigns.
Recency in Email and SMS Marketing
Email deliverability platforms including Klaviyo and Mailchimp weight recency heavily in their engagement scoring algorithms. Sending to dormant addresses (R score 1 or 2) increases spam complaint rates and damages sender reputation. Both effects reduce deliverability for the entire list.
Sephora, the cosmetics retailer, segments its Beauty Insider email list by recency and suppresses customers inactive for more than six months from promotional sends. Instead, those customers receive a dedicated win-back sequence of two to three emails before being re-evaluated for suppression. This practice keeps Sephora’s open rates above the retail industry average of approximately 17%, according to Klaviyo benchmark data.
The same principle applies to customer lifetime value modeling: a dormant high-spender may have a lower predicted future value than an active moderate-spender, precisely because recency is decaying their probability of return.
Recency Decay
Recency is not a static snapshot. Its predictive value decreases as time passes, a concept called recency decay. Brands model this as an exponential decay function applied to engagement probability.
Simplified decay model:
P(return) = P₀ × e^(−λt)
- P₀ = baseline return probability at time of last purchase
- λ = decay rate (derived from historical churn data)
- t = days since last purchase
A customer with a 60% return probability at day 0 might decay to 35% at 90 days and 15% at 180 days. Subscription businesses like Dollar Shave Club use decay modeling to trigger win-back offers before a customer crosses the threshold where reacquisition cost exceeds projected revenue.
Recency vs. Frequency
Marketers sometimes confuse recency with frequency. Frequency measures how often a customer purchases over a defined period; recency measures when they last did. A customer who purchased 20 times last year but has been inactive for eight months scores high on frequency and low on recency. For predictive purposes, their recency score is the more reliable indicator of short-term conversion potential.
This distinction matters for customer segmentation. A “high frequency, low recency” customer is a lapsed loyalist worth targeting with reactivation messaging, rather than a standard promotional offer. Recognizing that distinction in campaign design typically improves win-back response rates.
Recency in Paid Media
Beyond CRM, recency drives retargeting logic across Google Ads and Meta. Advertisers create audience windows based on how recently a user visited a page, viewed a product, or abandoned a cart. Standard e-commerce retargeting windows run 7, 14, and 30 days, with bid multipliers that decrease as time since the visit increases.
A retailer might structure bids like this:
- Last 3 days: 150% of base CPM for users who visited a product page
- 4 to 14-day window: 100% of base CPM
- 15 to 30-day window: 60% of base CPM
This mirrors the same recency logic as RFM scoring, applied in real time to paid audiences. Connecting paid retargeting recency to conversion rate optimization allows brands to test whether shorter windows with higher bids outperform broader windows with lower bids on a cost-per-acquisition basis.
Applying Recency to Campaign Strategy
- Prioritize high-recency segments for product launches and time-sensitive promotions. Response rates for R5 customers can run three to five times higher than R1 segments.
- Set recency thresholds before suppression. Define the inactive cutoff based on category purchase cycle, not a generic 12-month default.
- Build win-back sequences that trigger at the inflection point where decay probability drops below your reacquisition cost threshold.
- Recalculate recency scores dynamically. A static RFM snapshot from last quarter will misclassify customers who purchased last week.
Recency is the starting point for meaningful behavioral segmentation. Combined with frequency and monetary data, it shifts marketing investment toward the customers most likely to generate near-term revenue and away from segments where spend yields diminishing returns.
Frequently Asked Questions About Recency in Marketing
What is recency in marketing?
Recency in marketing measures how recently a customer completed a target action, most commonly a purchase. It is the “R” in the RFM model (Recency, Frequency, Monetary value) and is consistently the strongest predictor of future purchase behavior among the three variables.
Why is recency the most important factor in RFM analysis?
Recency outperforms frequency and monetary value as a predictor because recent buyers have demonstrated current intent, their contact details are likely still valid, and the brand remains in their consideration set. Database marketing research dating to the 1960s, including work by direct mail analyst Arthur Hughes, has confirmed this predictive weight across industries.
What is recency decay?
Recency decay is the decrease in a customer’s predicted return probability as time passes since their last purchase. Brands model it as an exponential decay function: a customer with a 60% return probability at the time of purchase might drop to 35% at 90 days and 15% at 180 days.
How does recency affect email marketing deliverability?
Sending campaigns to dormant email addresses (low recency scores) increases spam complaint rates and damages sender reputation, reducing deliverability for the entire list. Most email platforms, including Klaviyo and Mailchimp, weight recency heavily in their engagement scoring algorithms.
How do you calculate a recency score?
The most common method is percentile ranking: sort your customer file by days since last purchase, assign the top 20% most recent buyers a score of 5, the next 20% a score of 4, and so on down to a score of 1 for the least recent 20%. The thresholds for what counts as “active” should reflect your category’s typical purchase cycle, not a generic calendar window.