Analytics
Measure what matters
RFM analysis
RFM buckets customers by Recency, Frequency, and Monetary value. Each dimension is scored 1–5 and combined into a single bucket.
The 11 buckets
| Bucket | Who they are |
|---|---|
| Champions | Recent, frequent, high-spend. Your most valuable. |
| Loyal | Frequent but not always recent. Reward to keep them. |
| Potential loyalists | Recent and frequent but low spend. Encourage upgrades. |
| New customers | Recent, infrequent. Onboard aggressively. |
| Promising | Recent, low frequency, low spend. Nurture. |
| Need attention | Average on all dimensions. At risk of slipping. |
| About to sleep | Recency starting to slip. |
| At risk | Previously strong customers falling off. |
| Can't lose them | High value, long since last seen. Win-back priority. |
| Hibernating | Infrequent, low-value, long-inactive. |
| Lost | Lapsed; unlikely to return without strong incentive. |
How scores are computed
For each customer, Pulse computes:
- R — days since last qualifying event (default:
purchase_completed). - F — count of qualifying events over the analysis window.
- M — sum of
totalproperty on qualifying events.
Each dimension is quintile-bucketed against the workspace population.
Using RFM
- Filter any segment by RFM bucket.
- Trigger win-back journeys on At risk → Hibernating transitions.
- Reward Champions and Loyal with early access or referrals.
Refresh schedule
RFM is recomputed nightly. You can trigger a recompute manually from the RFM page.