At a glance
Share of Parcelforce consignments that hit any non-success scan event in the period: failed first attempt, customer-not-home, address issue, customs hold, damage, refused, “out for delivery, returned to depot”. Exception is broader than “late” because some exceptions ultimately deliver on-time after a re-attempt. This card is the operational-noise gauge for premium-tier UK shipping.
| What it counts | COUNT(DISTINCT shipments WHERE any_event IN ('FAILED_ATTEMPT','REFUSED','RTS','CUSTOMS_HOLD','DAMAGED','UNDELIVERABLE','REDIRECT')) / COUNT(shipments) over rolling 30 days. A consignment that triggered multiple exceptions counts once. |
| What counts as “exception” | Any non-success scan event in Parcelforce’s tracking webhook. Carded “Sorry we missed you” is an exception. Customer-not-home is an exception. “Address details required” is an exception. Successful re-attempt does not erase the exception flag, but the consignment may still mark on-time on par_otd_rate. |
| Customer-fault vs carrier-fault attribution | Pooled. Parcelforce’s failure_reason field tags each event but the headline rate does not split. Roughly 60 to 70% of premium-tier exceptions are customer-fault (recipient absent, address wrong, refused); 20 to 30% are carrier-fault (driver mis-route, depot delay); 5 to 10% are external (customs, weather). Pair with par_failed_delivery_count for the customer-fault-only view. |
| Service level scope | All services pooled (Express9, Express10, ExpressAM, Express24, Express48, Globaldirect, Saturday). |
| Money-back-on-late interaction | Customer-fault exceptions do not qualify for service-failure refund. Carrier-fault exceptions on time-definite tiers usually do. The exception count and the refund-eligible-late-count diverge in predictable ways; pair with par_open_claims. |
| Globaldirect customs | International customs holds register as exceptions and inflate the rate on heavy-EU merchants. Use par_shipments_by_destination to localise. |
| B2B vs B2C | Pooled. B2B addresses (corporate reception with named-recipient signature) generate fewer recipient-not-home exceptions; B2C addresses generate more. A merchant shifting B2C-heavy will see exception rate creep upward even with constant carrier performance. |
| Time window | 30D vsP (rolling 30 days, period-over-period vs prior 30 days) |
| Alert trigger | >3% critical, the gauge sentiment trips. Parcelforce premium tier typically runs 1.5 to 2.5%. |
| Roles | owner, operations |
Calculation
Calculated automatically from your Parcelforce Worldwide 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 UK gourmet-foods merchant: £75 average-order-value, perishable cohort (chilled ready-meals, fresh fish, artisan cheese) shipped on Parcelforce Express24 with mandatory next-day delivery. Reading taken at 09:00 GMT on 14 Mar 26 for the trailing 30 days (12 Feb 26 to 13 Mar 26).| Exception reason | Consignments | % of exceptions | % of total volume |
|---|---|---|---|
| Recipient absent (no safe place) | 84 | 47% | 1.4% |
| Address details required | 31 | 17% | 0.5% |
| Refused at door | 12 | 7% | 0.2% |
| Damaged in transit (chilled-leak) | 22 | 12% | 0.4% |
| Customs hold (Globaldirect EU) | 18 | 10% | 0.3% |
| Driver mis-route / depot delay | 13 | 7% | 0.2% |
| Total exceptions | 180 | 100% | 3.0% |
- Recipient-absent dominates at 47% of exceptions. Perishable cohort is recipient-not-home-sensitive: food spoils on the doorstep. The carded redelivery means the parcel sits at the depot overnight (chilled) or ruins (ambient). Customer-fault on Parcelforce’s side, but a merchant problem in customer experience and refund cost.
- Damaged-in-transit at 12%, almost certainly chilled-leak. Gourmet food in cardboard during March warming weather. Not a Parcelforce-fault per se; Parcelforce handle parcels per their published guidelines and a leaking chilled bag damages neighbour parcels too. Pair with packaging review (gel-pack vs dry-ice, double-walled cartons) before raising it as a carrier issue.
- Customs-hold 10% is structural for Globaldirect EU. Consider whether Parcelforce Globaldirect is the right service for chilled food into the EU; many gourmet merchants drop EU shipping post-Brexit because customs delays make the food spoil before clearance.
- Refund-eligible exceptions are the 13 driver-mis-routes (7%) plus a portion of the damaged-in-transit (case-by-case). That is ~£140 of carriage-refund opportunity (13 × £8 Express24 charge + a few £20 chilled-food claims). Pair with
par_open_claims. - 3.0% exception rate against a 1.5 to 2.5% premium-tier benchmark suggests recipient-experience friction. The fix is delivery-instructions UX: prompt customer at checkout for “leave with neighbour” or “specify safe place” preferences, reducing the 84 recipient-absent exceptions by 30 to 50%. Premium gourmet customers are willing to be directive about delivery; the platform just needs to ask.
Sibling cards merchants should reference together
Exception rate is the broad operational-noise gauge. To triage, pair with these:| Card | Why pair it with Exception Rate | What the combination tells you |
|---|---|---|
| Failed Deliveries | Subset. All failed deliveries are exceptions; not all exceptions are failed deliveries (a customs hold that ultimately delivers is an exception only). | Subtract failed-deliveries from exceptions to see the customs / damage / mis-route bucket. |
| On-Time Delivery Rate | Some exceptions still deliver on-time after re-attempt. Comparing the two reveals the recoverable-vs-unrecoverable split. | High exception + high OTD = re-attempts working; high exception + low OTD = exceptions converting to misses. Two different fixes. |
| Open Claims | Refund-eligible carrier-fault exceptions feed open-claims. | If exception rate is 3% and claim filing rate is <30% of carrier-fault exceptions, money is being left on the table. |
| Returned to Sender | Terminal-exception subset. Some exceptions escalate to RTS after multiple failed attempts. | High exception + rising RTS = re-attempt logic is failing; consider AAR (alternate-address redirect) workflow. |
| Shipments by Destination | Geographic split. International / Globaldirect destinations carry higher customs-hold exception rates. | If exceptions are concentrated in EU postcodes, the problem is customs not carrier. |
| Express24 Service Day Promise | The premium tier’s exceptions are most expensive (refund value, customer-perception, B2B). | Express24 exception rate driving the headline = revisit your Express24 cohort first. |
Cross-connector: shopify.refund_rate | Downstream impact. Carded-and-spoiled, damaged-in-transit, refused-at-door all drive refund rate up at 3 to 7 day lag. | A 1-point exception-rate rise typically shows up as a 0.3 to 0.6 point refund-rate rise on Shopify a week later. |
Cross-connector: bigcommerce.refund_rate | Same downstream as Shopify. | Same caveat. |
Cross-connector: apc.apc_exception_rate | Peer UK premium shipper. | Useful for shop-around. APC and Parcelforce have similar operating profiles; Yodel and Hermes / Evri run higher on this metric structurally. |
Reconciling against the vendor’s own dashboard
Where to look in Parcelforce’s own dashboard: Parcelforce ParcelManager → Reports → Exceptions Summary. Filter by All services / All depots / Last 30 days. The closest like-for-like view is Tracking Events with Failure Reason. ParcelManager publishes a per-consignment audit at Track and Trace → Filter “Issues” with the reason code per event. Why our number may legitimately differ from Parcelforce’s portal:| Reason | Direction | Why |
|---|---|---|
| Distinct vs event-count | Ours lower | Parcelforce’s portal counts each exception event individually; a consignment carded twice then delivered shows as 2 events. The card de-duplicates at the consignment level (any exception flag = 1). The two reconcile when the merchant filters to “consignments with at least one exception”. |
| Customer-fault inclusion | Ours higher | Parcelforce’s account-team sometimes quote a “carrier-fault-only” rate that excludes recipient-absent / address-issue. The card includes all exceptions because all of them affect customer experience. |
| Time zone | Boundary | UK local time (GMT or BST) on both sides. DST-transition Sundays cause minor noise. |
| Globaldirect customs | Ours can be higher | Customs-hold counts as exception in the card. Some Parcelforce reports separate Globaldirect from domestic; merging them inflates the headline. |
| Card | Expected relationship | What causes legitimate divergence |
|---|---|---|
shopify.refund_rate | Downstream lag. Exceptions drive returns and refunds at 3 to 7 day lag. | App-install events, B2B / pre-order workflows. |
apc.apc_exception_rate | Peer UK premium. | Different consignments. |
royal_mail.rm_exception_rate | Value-tier sister carrier. | Different network, different scan vocabulary. |
Known limitations / merchant FAQs
Why is exception rate broader than late rate? A consignment can be on-time and exception-flagged (driver carded, customer popped out, redelivered same day on time). It can also be late without an exception (sat in the depot waiting for a connection scan, no attempt failed, just slow). The two cards measure different operational realities: exceptions track “things going wrong”; lateness tracks “the customer’s promise was broken”. Use both. How do we reduce the recipient-absent share? Three checkout-UX changes typically cut it 30 to 50%. (1) Prompt for delivery preferences explicitly: “leave with neighbour”, “safe place”, “specific entry instructions”. (2) Send a pre-delivery SMS or email morning-of with a 1-hour predicted window (Parcelforce’s API exposes the predicted-window data). (3) Offer an optional pickup-point delivery for Express48 cohort (Parcelforce works with the Post Office network). Parcelforce’s premium-tier customers are generally directive when prompted. Damaged-in-transit is rising; is it the carrier or our packaging? Parcelforce handle parcels per published ISTA-3A guidelines (drop, vibration, edge-crush). If your damaged-in-transit rate is above 0.5% of total volume, the issue is almost always packaging, not carrier handling. Compare against APC, DPD on the same product cohort if you ship via multi-carrier; if all three rates rise together, it is packaging. Pair withpar_open_claims for the financial recovery view.
Can we filter customer-fault exceptions out of the headline?
Today no; the card surfaces all exceptions. Most ops teams want the full picture because customer-fault exceptions still drive customer-service load and re-shipment cost. If you need the carrier-fault-only view, use par_failed_delivery_count plus par_open_claims together; the eligible-for-claim set is the carrier-fault set. The full exception count is the customer-experience set.
Carded “Sorry we missed you” parcels are rising; what is the action?
Three actions, in order of leverage. (1) Checkout delivery preferences (above). (2) Predictive SMS / email the morning of delivery with a 1-hour window. (3) Negotiate access uplift with Parcelforce on B2B and apartment-block addresses (concierge handover, key-card access, etc.). The card will dip 30 to 50% on these; the residual is structural (genuine recipient-absent cases) and not worth chasing further.
Refused-at-door exceptions: are these a quality issue?
Often yes. Refused-at-door correlates with two things: (1) packaging that arrived visibly damaged so the customer refused (look at par_open_claims for damage claim filings), or (2) the customer changed their mind / identifies the parcel as a duplicate / a Klarna-buyer-regret moment. Cross-reference Shopify cancellation patterns; refused-at-door often follows a customer-initiated cancellation that did not reach the warehouse in time.
Globaldirect customs holds inflate our headline; should we drop EU shipping?
Cost-benefit. Customs-hold exceptions on Globaldirect EU run 5 to 12% of EU volume (vs <1% on domestic). For most merchants the EU revenue is worth the noise, but the operating profile is structurally different and should be tracked separately on par_shipments_by_destination. If EU is <5% of volume and customer-service tickets are dominated by EU customs queries, simplify to UK-only and let DHL / FedEx handle premium-EU. The card cannot make that decision; it surfaces the cost.
B2B addresses generate fewer exceptions; should we tier our pricing on that?
Many merchants do. B2B is structurally cheaper to fulfil because exception rate is half the consumer-DTC equivalent (no carded redeliveries, named-recipient sign-off, predictable hours). Parcelforce account managers will negotiate a B2B-specific rate band on volume. Pair this card with parcelforce_xc_otd_by_channel for the channel split that supports the conversation.
During Q4 peak, exception rate spikes; is this signal or noise?
Both. Q4 typically lifts exception rate by 1 to 2 percentage points across UK premium carriers (volume saturation, weather, more first-time-buyer customers with imperfect addresses). The signal-vs-noise call is whether Q4 exception rate is elevated relative to last Q4; comparing to a flat 30-day baseline is unfair. Adjust your weekly-review expectations during November / December and reset the comparison to the YoY equivalent week.