At a glance
Percentage of registered customers in their “should have ordered by now” zone, weighted by historical lifetime value. The card the marketing and retention teams use for proactive winback, the customers in the at-risk band are slipping away and need intervention now.
| What it counts | For each registered customer with at least 2 historical orders: compute average inter-order gap. Customers whose days-since-last-order > 1.5× their avg gap are “at risk”; > 2× are “lapsed”; > 3× are “churned”. Headline is the % of registered customer base in at-risk + lapsed bands. |
| VAT / tax treatment | n/a (rate metric). |
| Shipping | n/a. |
| Discounts | n/a. |
| Refunds | Refunded orders count for inter-order gap (the customer placed the order). |
| Cancelled / voided orders | Excluded from gap calculation. |
| Currency | n/a. |
| Channels / sources | All channels contribute. POS-only, web-only, B2B-only customers have very different reorder patterns; consider per-channel views for cleaner segments. |
| Risk bands | At-risk (1.5-2× normal gap), Lapsed (2-3× normal gap), Churned (>3× normal gap). At-risk gets highest-priority winback; lapsed gets one final attempt; churned removed from active marketing. |
| Two-order minimum | Customers with only 1 historical order are excluded; cross-reference Customer Count for the new-customer cohort. |
| B2B Edition note | B2B has wider, more predictable gaps. Configure B2B-specific thresholds (e.g. 1.2× for B2B to catch slippage earlier than retail’s 1.5×). |
| Time window | 90D (evaluation window; uses lifetime history for baselines) |
| Alert trigger | >25% of base inactive 90d |
| Roles | owner, marketing |
Calculation
Calculated automatically from your BigCommerce data. See the At a glance summary above for what the metric tracks and the worked example below for a typical reading.Worked example
A US homewares brand on BigCommerce Pro, 90-day evaluation 14 Feb 26 to 14 May 26.| Risk band | Customers | % of base | Avg historical LTV | Total at-risk LTV | Suggested action |
|---|---|---|---|---|---|
| Active (within normal gap) | 6,420 | 64% | $214 | $1.37M | None, healthy |
| At-risk (1.5-2× gap) | 1,840 | 18% | $186 | $342k | Winback campaign now |
| Lapsed (2-3× gap) | 980 | 10% | $158 | $155k | Final-attempt winback |
| Churned (>3× gap) | 760 | 8% | $124 | $94k | Remove from active marketing |
| At-risk + Lapsed total | 2,820 | 28% | $497k LTV at stake | Headline action zone |
- **28% of customers (~342k LTV) is where winback campaigns get the highest ROI; lapsed customers are harder to recover.
- At-risk customers have higher avg LTV (158) because at-risk includes more recent, higher-engagement customers; lapsed has been declining for longer. Winback ROI is higher on the at-risk band.
- Churned at 8% / $94k LTV is “accept the loss”. Customers >3× their normal gap have moved on (changed brand, life event, lost interest). Continued marketing wastes spend without recovery; remove from active campaigns.
- A 28% rate is on the high side; healthy stores see 18-25%. Possible causes: (1) recent acquisition surge of low-quality customers (high one-time-buyer rate); (2) product-experience issues driving customers away; (3) competitive pressure (a major competitor entered or won SEO ranking); (4) email-marketing degradation (open rates falling, fewer touches).
- Per-segment winback economics. A typical at-risk winback email achieves 5-12% conversion to a re-order. At 1,840 at-risk × 8% conversion × 13,200 of recovered revenue** per winback campaign. Email cost: <$50. ROI: 200×.
- Launch at-risk winback campaign this week. Personalised email: “We miss you, here’s 15% off [product they bought before]”. Target: 1,840 customers.
- Lapsed final-attempt campaign in 2 weeks. More aggressive incentive (20-25%), final outreach, then de-list.
- Investigate root cause of 28% rate if recurring quarter-over-quarter, structural retention issue (product, experience, or pricing). One-off may be a recent acquisition wave.
- Configure ongoing winback automation Klaviyo flow triggered when customer crosses 1.5× their avg gap; recovers customers without manual campaigns.
- Quarterly: re-baseline gap thresholds, customer behaviour drifts; review if 1.5× / 2× / 3× still represent real risk bands or need adjustment.
Sibling cards merchants should reference together
| Card | Why pair it with Churn Risk |
|---|---|
| BC Top Customers | At-risk top customers are the highest-value recovery; cross-reference. |
| Repeat Rate | Repeat-customer health correlates inversely with churn risk. |
| Customer Count | The denominator base. |
| Order Frequency | Per-customer order frequency informs the “normal gap” baseline. |
| Customer Trend | New-customer acquisition trend; balance recoveries against new customers. |
| BC Customer Segments | New / returning / VIP segments; churn risk concentrates in some segments. |
klaviyo.kl_at_risk_segment | Klaviyo’s predictive at-risk segment; should align with this card’s at-risk band. |
shopify.churn_risk | Cross-platform reference. |
Reconciling against the vendor’s own dashboard
Where to look in BigCommerce Control Panel: BC does not provide a native customer-churn-risk view. Customers → View sorted by last-order-date ascending shows customers who haven’t ordered recently, but the per-customer “normal gap” baseline isn’t computed natively. This card is one of the highest-value adds. Why our number may legitimately differ from any BC native view:| Reason | Direction |
|---|---|
| Inter-order gap calculation. We compute per-customer baselines; BC has no equivalent. | Vortex IQ provides what BC doesn’t |
| Risk-band thresholds. Default 1.5× / 2× / 3×. Configurable per merchant tolerance. | Configuration-dependent |
| Two-order minimum. Customers with one order excluded; some merchants want them included as “first-order at risk”. | Configurable |
| Channel filtering. POS / web / B2B have very different reorder cadences; default is all-channels. | Different patterns per channel |
| Time-zone. UTC default; gap calculations affected at midnight boundaries. | Minor |
| Card | Expected relationship | What causes legitimate divergence |
|---|---|---|
klaviyo.kl_at_risk_segment | Klaviyo’s predictive at-risk customers should overlap with this card’s at-risk band. | Klaviyo uses email-engagement signals; this card uses purchase-cadence signals. ~70-80% overlap typical. |
hubspot.hs_lifecycle_stage | HubSpot’s “lapsed customer” stage. | HubSpot’s stage is manually-set or rule-based; expect partial alignment. |
Known limitations / merchant FAQs
My churn risk is 28%, is that bad? On the high side. Healthy stores see 18-25%. Above 25% triggers the alert. Investigate root causes: (1) recent low-quality acquisition wave; (2) product/experience issue; (3) competitive pressure; (4) email-marketing degradation. The number itself is symptomatic; the root cause is what to fix. Why does my B2B account show as at-risk when they’re on a quarterly cycle? Default thresholds (1.5×) are too tight for B2B. A B2B account with a 90-day normal gap flagged at 135 days (1.5× = “at risk”) may just be a week late. Configure B2B-specific thresholds (1.2× and 1.5× and 2.0×) to better match B2B cadence. Should I send winback to lapsed customers or focus on at-risk only? At-risk first (highest ROI). Lapsed second (final attempt). Don’t ignore lapsed entirely; one final email recovers 3-5% of them, which is still positive ROI given low send cost. Don’t market to churned (>3× gap), spend wasted. A customer is in at-risk band but they just ordered last week, why? Either: (1) their normal gap is short (e.g. weekly subscription customer skipped a week), so the 1.5× threshold is hit even with recent activity; (2) data sync lag (their recent order hasn’t indexed yet); (3) the recent order was on a different channel and the per-customer aggregation is missing it. Refresh the card; if persists, investigate channel mapping. Can I exclude one-time-buyer customers? The card already does (two-order minimum). One-time buyers go in their own segment for “first-order winback” campaigns; that’s a different problem with different mechanics. My churn risk dropped after a winback campaign, did I really save them? Possibly. The win is whether the recovered customer’s next order happens within their normal gap. If they re-engaged from the winback then disappeared again, the recovery was temporary. Track 90-day re-engagement rates for true ROI measurement. Why is my marketplace customer cohort missing from this view? Marketplace customers (Amazon, eBay) often have anonymised contact info; BC doesn’t get the customer email orcustomer_id. Each marketplace order looks like a guest order. Marketplace customers cannot be churn-evaluated in this card; they’re managed via marketplace-side tools.
Should I retire customers from active marketing once churned?
Yes. Email lists with high “dead address” rates degrade overall deliverability (Gmail, Yahoo penalise senders with high bounce / non-engage rates). Removing churned customers protects your active list’s deliverability.
My retail VIPs aren’t showing up as at-risk, are they really safe?
Maybe. VIPs have long histories and irregular cadences (a VIP buying twice a year for $5k each isn’t at risk if their “normal” is twice a year). The baselining accommodates this. Manually monitor top-50 VIPs separately, the LTV concentration is too high to rely on aggregate metrics for them.
Multi-currency: do customers cross currencies in this analysis?
Customer ID is currency-agnostic; the same customer can transact in multiple currencies. The inter-order gap calculation uses all their orders regardless of currency. The LTV column blends currencies (use a single-currency conversion view for clean LTV).
Can I see churn risk by acquisition source?
Yes via filter. Customers acquired via paid ads typically have higher churn than organic / referral. Filter the card by acquisition channel for source-specific retention insights; the result often informs ad-channel ROI calculation.
What’s the difference between churn risk and lapsed customers?
Churn risk is the predictive view: customers in their “should have ordered” zone but haven’t yet. Lapsed customers are a sub-band of churn risk (2-3× normal gap). Churned customers (>3× gap) are considered lost. The card surfaces all three for differentiated marketing actions.