Parcel Inflation Peaked the Day We Published. Here's Why.
Data through September 30, 2026. Written October 15, 2026.
Given that, if we told you that carrier rates actually decreased quarter over quarter in Q3, would you believe us? It did, but here's the catch: your bill barely did. What operators actually paid per package sits within one percent of its June high and was edging up again as the quarter closed. The shipping cost deceleration appears to have been a temporary head fake as shipping costs are nearing back to their YTD peak.
The miss: parcel inflation peaked the day Issue 01 closed
Growth held its pace. Shipping costs cooled slower than everything else.
GMV growth closed the quarter at 11.7%, against 13.4% at the end of June, and order growth closed at 5.3%, against 8.9%. The June figures were inflated by an unusually strong May; the September figures didn't get that bump, leading to the drop. Read month by month, GMV growth has run near 10% and order growth near 3% to 5% since spring. Parcel inflation fell from 12.9% to 6.9% over the quarter, dropping every week, and growth in the prices operators charge slowed even faster, from 3.6% to 1.1%.
So prices are still rising, but slowly. Shipping costs are still rising too, 6.9% on a year ago. Selling prices are rising 1.1%. That gap is 6.1 times, wider than July's 3.6 times, even though both numbers came down, because selling price growth came down faster. Parcel rates follow fuel within weeks, not a quarter, which is the main driver behind this issue's shipping call.
Three moves this quarter
-
Cost
Shipping costs cooled during Q3. But get ready for an expensive peak season.
Carrier rates fell 3.3% on the quarter and what operators paid fell 0.6%, after 13 straight weeks of increases in Q2. By mid-October crude was back up about 40% from its July low. If rates still sit 3.3% below Q2, explore a new contracted rate on the longest term you can get. If fuel has pushed them up, trade for a cap on the fuel surcharge instead.
-
Demand
Plan Q4 on the pace you have been running
Operators closed the quarter with healthy growth: GMV at +11.7% and orders at +5.3%. There was no slowdown in Q3 to carry into Q4. Expect the Q4 growth rate to come in slightly below Q3, because Q4 2025 is a tough comparable (9.8% growth).
-
Inventory
Caught up to last year's, did yours?
Days on hand rose from 88 in mid-September to 93 at the close, level with last year for the first time since the destocking trend began. Weeks of cover, which weights velocity, rose about 9% this quarter and still sits roughly 6% below last year.
Key Commerce Signal Metrics
| Metric | Q3 Close | Vs Q2 close | Direction |
|---|---|---|---|
| GMV Growth | +11.7% | 13.4% | |
| Order Growth | +5.3% | 8.9% | |
| Parcel Inflation (market) | +6.9% | 12.9% | |
| Parcel Inflation (realized) | +6.9% | 5.8% | |
| Selling Price Inflation (platform) | +1.1% | 3.6% | |
| Inventory DOH | 93 days | 89 days | → Holding |
| Inventory Turns | 3.91x | 4.09x | → Holding |
Two notes on reading the table. Growth figures are 91-day reads at the quarter close; the Q2 close was lifted by a strong May, and month-by-month growth has held steady since spring. Realized parcel inflation rose while market inflation fell because the two compare against different bases a year ago; the quarter-over-quarter figures (market -3.3%, realized -0.6%) are in Market Pulse.
Shipping cooled. Demand didn't move.
The quarter closed with GMV growing 11.7% and orders 5.3%, down from 13.4% and 8.9% at the end of Q2. Looking month by month, almost all of that drop is one strong May leaving the comparison: GMV growth has run near 10% and order growth near 3% to 5% since spring, with a soft July in between. It was a strong quarter, with the gains split about evenly between more orders and bigger baskets (AOV +4.1%). Our read is that this is the pace, not a slowdown, and the calls are built on that assumption.
Parcel follows fuel within weeks, not a quarter.
Diesel drives your surcharge, and your surcharge resets every week. That's why your rate per package tracked crude so closely this year. Crude peaked April 7 at $113 a barrel and fell 39% to $69 by July 6. Your market rate peaked the week of May 8, bottomed the week of July 10, and dropped about 4% per package in the two weeks around July 1. It hasn't moved since: $10.35 in July, $10.42 in August, $10.36 in September. The chart below shows the weekly rate, which is why it turns two months before the quarterly index.
And yet, crude came back up. It averaged $94 in September, touched $102 on September 10, dipped to $88 in early October and was back at $97 on October 15, about 40% above the July low. The September spike was too short to move a weekly surcharge, so your rate didn't move. If crude holds in the mid $90s through October, expect it in your surcharge by November. If it drops back under $80, the rate you had at the close is the rate you take into peak. Going into October the trade press was calling this the most expensive peak on record. That's true for the level, but overall, rates have been trending down since June.
Parcel cost follows crude within weeks, not a quarter
Steady, not slowing
Each bar on the chart below is that month's growth, each read at the end of the month over its final 30 days. The quarterly figures in the Key Metrics table use 91-day windows, which is why the Q2 close reads higher than any single month in it.
Rates fell. Bills barely moved.
Carrier rates dropped 3.3% per package from Q2 to Q3. What operators actually paid dropped 0.6%. The difference isn't your contract. It's what you shipped.
So let's check what happened. The chart is looking at parcel rates normalized to 100 for the first day of each quarter. You can easily see the story of each quarter, declining against Q2's growth. To further explain what you're seeing, the market line is looking at carrier rates with the mix of packages held fixed, so it changes only when rates move. The realized line is what actually shipped, at what it really cost. In Q2 the two climbed together, 5.0% and 4.3%. In Q3 rates gave back 3.3% and bills gave back 0.6%. So what's driving the 2.7% difference at the end of September? A mix of things: heavier packages, faster services, more residential deliveries, and regional carriers raising fees to track the nationals.
The TD Cowen/AFS Freight Index, on a different set of shippers, found ground parcel cost per package down 2.6% from Q2 and still 5.2% above a year earlier. Ours: down 0.6% on the quarter, up 6.9% on the year.
Rates rose through Q2 and gave most of it back in Q3. Operators' bills did not.
The consumer picture.
The University of Michigan sentiment index closed September at 48.1, down 3.6 points from August and within four points of May's record low, on fuel prices and trade tensions. We have seen this pairing before: in May, at the record low, orders were running 5% above the prior year and GMV almost 10%.
Sentiment measures how people feel about prices; the platform measures what they do about them…and they kept buying. This September they are still buying, a little less often than in the spring and in slightly bigger baskets. Sentiment is not the constraint.
Consumers say they feel worse. They keep ordering.
Brands rebuilt. 3PLs stayed lean.
Days on hand caught up with last year in September
Brands carry about three more days than 3PLs
Median days on hand closed the quarter at 93, up from 88 in mid-September and level with the same week of 2025 for the first time since the destocking began. The 18-month destocking trend ended in December; days on hand then held in a band eight to twelve percent below last year through August, and the gap closed in the final three weeks of September, with just over half of operators adding days.
Weeks of cover, which weights stock by velocity, rose about 9% during the quarter and closed roughly 6% below last year, from about 20% below at the start of July. That reconciles the import data: NRF's Global Port Tracker shows June containers up 13.2% and calls retailers “well stocked”, and the August Logistics Managers' Index shows inventory contracting upstream and expanding downstream.
Well stocked upstream, level at the floor at the quarter close, still thin on cover. The build came late because operators held back all year on an uncertain outlook and started from the lowest level in the series, about 87 days in August; the catch-up took three weeks once it began. The comparison worth holding is 2022, the last year with record diesel, when the network entered peak on a post-COVID inventory glut. This year it enters roughly 12% below its April 2025 high.
94 days on hand in the last week of September, about one above the network median, after running roughly five days leaner than last year through the quarter. Heavier inventory is the exposure, and the financing got more expensive in September with the 10-year Treasury at 5% for the first time since 2023. That is the pressure pushing brands to let 3PLs carry the stock.
91 days on hand in the last week of September, three below brands and level with a year ago. 3PLs are carrying less stock into a peak that arrives at nearly twice the September run rate, which raises the stakes on 3PL capacity in the two weeks that matter.
Four calls for peak 2026. On record, scored in January.
Each call names the metric, the direction, a number we can be wrong about and the rule we will use to score it. High confidence requires a full-quarter trend, outside corroboration and, after last quarter's miss, the daily series and fuel both moving in the called direction. Quarter-over-quarter calls are scored on shipment-weighted quarter averages. Full reasoning, external checks and validation notes are on the methodology page. Read the methodology
| # | Call | Direction | Scoring rule |
|---|---|---|---|
| 01 | GMVfewer new orders, slightly bigger ones | ↓ FallingModerate | GMV below 11.7% and at least 2.0x order growth |
| 02 | Parcelpeak costs more per package than Q3 | ↑ RisingModerate | Q4 average market cost per shipment above Q3's; market inflation 4% to 8% at the close |
| 03 | Inventorypeak draws stock down below where it started | ↓ FallingModerate | DOH closes December below 90 and below its November peak |
| 04 | Carrier mixCarrier mix fragments; the savings narrow at peak | ↑ Rising shareModerate | Alternative-carrier share of Nov 23 to Dec 27 shipments above 2025; like-for-like savings positive and below their Q3 level |
fewer new orders, slightly bigger ones
GMV growth closes Q4 below 11.7% and at least twice order growth, which holds in a 3% to 6% band. The Q4 2025 base averaged 9.8% against 3% to 9% in Q3, so November's growth rate comes in lower on the comparison alone.
GMV below 11.7% and at least 2.0x order growth
Scored at the Dec 31, 2026 close
peak costs more per package than Q3
Q4's average market rate per package comes in above Q3's, and market inflation closes December at 4% to 8% year over year. Peak surcharges from October 25 and crude at $97 on October 15, about 40% above its July low, push it up. Crude back under $80 would leave Q4 level with Q3.
Q4 average market cost per shipment above Q3's; market inflation 4% to 8% at the close
Scored at the Dec 31, 2026 close
peak draws stock down below where it started
Median days on hand builds into November, then peak sells it down faster than receipts replace it, closing December below 90 days, under the 93 it entered at. Last year it ran from 95 to 103 to 88. This year's build came three weeks later and from a lower base, so the close is a call, not a certainty.
DOH closes December below 90 and below its November peak
Scored at the Dec 31, 2026 close
Carrier mix fragments; the savings narrow at peak
Alternative and regional carriers, including Amazon's shipping service in its first peak open to all businesses, carry a materially larger share of peak shipments. Matched like for like on service, weight and zone, the saving is a few percent, and it narrows at peak.
Alternative-carrier share of Nov 23 to Dec 27 shipments above 2025; like-for-like savings positive and below their Q3 level
Scored at the Dec 31, 2026 closeWhat This Means for You
If you set your Q4 plan in June, you planned for 13% growth and 13% shipping inflation.
You will get about 11.7% and 6.9%. You do not need to redo the budget. You need to move two numbers inside it: what you ask the carrier for, and when your last inbound receipts land. On the carrier: rates closed the quarter 3.3% below their Q2 average and sat there through mid-October. Before peak surcharges start October 25, the ask is the base rate and a cap on the surcharge. After October 25 the cap is off the table and the ask becomes January's base rate, so track what peak actually costs you per package and bring that number to the renewal. On receipts: anything scheduled after mid-October lands in the ramp, so if you are reading this inside the ramp, the question is no longer when they land but which SKUs you expedite.
If your receipts land after mid-October, they land in the ramp to Peak.
Peak runs the weeks ending November 28 and December 5, at nearly twice the September order rate, and the ramp begins the week of November 9. Days on hand caught up with last year in the final three weeks of September; weeks of cover has not, so receipts scheduled after mid-October land in the ramp.
Peak week runs at nearly twice the September order rate
Every call goes on record. Here is how Issue 01 scored.
- Parcel InflationMiss
- GMV GrowthHit
- Inventory DOHHit
- Inventory TurnsHit
Three of four. The miss was our highest-confidence call, and it is the highlight in this issue. We called parcel inflation rising because it had risen thirteen straight weeks and surcharge announcements were imminent. Both of those things were true, and the call still missed.
Here's why. Crude oil peaked April 7 and fell 39% by July 6. Parcel rates follow fuel within weeks, and on the daily data they peaked the week of May 8 and were already stepping down when Issue 01 closed on June 30. The index we were watching is a 91-day average, so it kept rising for weeks after the rates underneath it had turned. We were looking at the right market through a window that showed us April.
Two corrections follow. Inside a quarter we now read the daily series, and the smoothed index is for the year-over-year trend only. High confidence on a shipping call now requires the daily series and fuel to be moving in the called direction.
Holding means within 15% of the value at publication. The four calls in this issue are scored at the Dec 31, 2026 close and reported in the next issue.
