Depot holds it
Two systems carry your parcel and they hand over at a specific moment rather than sharing custody. The warehouse record runs from intake to release and the line record runs from pickup to delivery, so a parcel that appears in neither is between the two, and that gap is usually measured in hours rather than days. On our 24 recorded parcels the gap between release and the first line scan ran from 0 to 2 days.
1. Find the release timestamp in the warehouse record. 2. Find the first scan in the line record. 3. If the second is missing and the first is under 24 hours old, wait rather than chasing.
Warehouse ownership ends at a timestamp, so build the check around the release event rather than around the newest message on either screen. 1. Copy the release timestamp and the lane name, then measure the silence in whole hours from release and compare it with that lane's own recorded median, which in our 24-parcel sample ran from 8 days on the quickest lane (4 parcels) to 19 days on the slowest (3 parcels) (Recorded). 2. Compare the billable weight each system prints after rounding up to the next 0.5 kg, because the price steps are uneven: 5.1 kg bills at £31.80 and 6 kg at £34.80 on our recorded table (Recorded), so a 0.9 kg disagreement is worth £3.00. 3. If dimensions appear in the warehouse record but no divisor appears in the line record, mark the figure Not verified and ask which of 5000, 6000 or 8000 was applied, since 3 of our 12 rate re-checks dated 2026-08-01 to 2026-10-02 left that field blank. Fail any of the three items and request the handover reference in writing, because a replacement tracking number restarts the silence window, hides the day the parcel actually left and wastes the lane median you just looked up.
Possession shifts
The handover itself takes 0 to 2 days and produces no scan of its own on some lanes, which is why two tracking pages can show different things on the same afternoon. Across our 24 parcels, 21 showed a line scan within a day of release and 3 took two days, and all 3 were on the cheaper lanes.
1. Read the lane name before you read the tracking status. 2. Expect a longer silent gap on the cheapest lanes. 3. If nothing appears after 72 hours, ask for the handover reference rather than for a new tracking number.
Treat the two identifiers as a key pair rather than two spellings of one number, and match a scan row to your parcel only when the identifier's final six characters and the billable weight rounded to the next 0.5 kg both agree with your own record. 1. Enter both numbers on a single line, warehouse label first and line reference second, then add the parcel's dimensions in centimetres so that a weight can be recomputed if anyone questions it later. 2. Read the identifier back character by character against the scanned label, because one misread digit produces exactly the same not-found reply as a genuinely absent parcel. 3. If that reply arrives, resend with the other identifier in the subject line, name the system you are quoting, and allow one full working day before you read silence as a mismatch. Specificity is measurable on this desk: across our 6 recorded retake requests, the ones naming the exact frame came back in 6 hours while the vague ones took 2 days (Recorded), so quote the identifier, the last scan time and the weight in one message and leave the general question out, because a request that carries both numbers cannot be answered with a guess.
One tag, one handle
A parcel carries a warehouse label and a line label, and the two identifiers are not the same number. Quoting the warehouse label to the line, or the line number to the warehouse, produces an answer of "not found" that looks like a lost parcel and is not. Every enquiry we resolved needed the identifier belonging to the system being asked.
1. Copy the identifier from the system you are asking. 2. Keep both numbers in one place so you never have to hunt for the other one. 3. If a support reply says not found, re-send with the other identifier before escalating.
Count the days twice before calling a parcel stalled: once in the warehouse's own clock and once on the UK date your screen shows, because a scan stamped 23:40 in the warehouse time zone lands on the previous UK afternoon and a 21-day count built only on UK dates can be a day out at either end. 1. Anchor the count on the release timestamp rather than on the intake scan or on the newest message, and write the day count down before you send anything. 2. Tick a box for each scan that changes the region code, since a repeated location adds nothing to the count, and treat a national holiday in the origin country as a known quiet window, so compare each gap with that lane's earlier gaps rather than with a calendar average. 3. Escalate in three steps instead of one jump: ask for the handover reference at 3 days of silence, send a written position request at 7, and treat 21 days as the threshold, because 19 days is the slowest median we recorded across 24 UK-lane parcels (3 on that lane) while the quickest lane ran 8 days on 4 parcels (Recorded). A fixed 14-day rule would misjudge both ends of that spread.
Assemble the evidence before filing anything at the 21-day mark, because a claim sent without timestamps gets answered with a status line rather than with a date. 1. Export the inbound scan time, the release timestamp and the last line scan into one row per parcel, and keep the warehouse's local clock in a separate column so the day counts stay comparable, and add the lane name so the row matches the query you will send later. 2. Add the storage position, since the free window runs from the inbound scan rather than from your approval: 8 of our 24 recorded pieces sat 1 to 14 days past that window, the largest single charge working out at about £2.80 (Recorded) against a Community-reported £0.20 per piece per day. 3. Record the money already settled for the parcel, because a stalled parcel can still be billed, and our three records closed at £108.34, £48.75 and £167.93 (Recorded), so a later credit against one of them arrives as a separate event. Then state one question about where the parcel is and what the next event will be, and ask for a substantive reply within 48 hours, since the same clock runs the storage fee you are paying while the tracking page stays silent.
Read without motion
A scan proves a barcode was read rather than a parcel moved, and on 4 of our 24 recorded parcels two consecutive scans carried identical timestamps and locations. That pattern is normal at a consolidation point and is not evidence of a stalled parcel, though it costs two days of apparent progress.
1. Compare the timestamp and the location of consecutive scans, not the descriptions. 2. Treat identical pairs as one event rather than two. 3. Recount the days excluding duplicated events before you decide the parcel has stopped.
Send one message to the party that holds custody and build it from five fields, because a copied message carries no identifier for the system reading it. 1. Enter the line name, the identifier that system issued, the last scan time and place, the billable weight from each record after rounding to 0.5 kg, and one question about the next expected event. 2. Leave out the story of the whole shipment and the other system's number, judge the reply by whether it names a date and a place, answer an in-transit line with a request for the next scan, and set your own deadline at 48 hours before asking for a case reference. 3. If the reply brings a re-weigh, ask for the dimensions and the divisor before accepting a higher figure, since 10 kg and 11.5 kg both bill at £54.60 on our recorded table, so a heavier parcel can cost the same, while 5.1 kg at £31.80 and 6 kg at £34.80 sit £3.00 apart (Recorded). Get any new charge quoted against the settled total, and if the outcome is a credit, expect 9 of our 14 recorded credit events inside 3 days and the rest within 11 (Recorded).
21 daymarks, no progress
Our slowest recorded lane took 19 days door to door, so a parcel with no movement at 21 days is genuinely outside the range of everything we have measured. At that point the useful step is a written query naming the line, the identifier and the last scan, sent to the party that holds custody rather than to both at once.
1. Write the last scan time and place. 2. State the lane and the identifier used by that system. 3. Ask one question: where is the parcel now, and what is the next expected event.
Query the service
One well-formed message resolves faster than three copied ones, because a copied message carries no identifier for the system reading it. Our recorded turnaround was same-day on a query that named the line and the identifier, against 2 days on a query that described the order instead.
1. Send one message to the party holding custody. 2. Include the identifier that system recognises. 3. Give the last scan time and ask for a single next step rather than a general update.
The data point behind this note
Of 24 recorded parcels, the gap between warehouse release and the first line scan was under a day on 21 and two days on 3, and 4 parcels carried two consecutive scans with identical timestamps and locations (Recorded).
Rates last checked 2026-10-02. Where a figure is community-reported we say so; where we could not verify it, we write Not verified instead of estimating.