Price creep
Produce price swings vs quiet creep: telling them apart
· by the Candor team — operators auditing real restaurant invoices
In produce, a price jump on your invoice can genuinely be the market and not your distributor. USDA's numbers for June 2026 make the point loudly: farm-level vegetable prices were 59.2% higher than a year earlier, while retail fresh-vegetable prices were up 9.9% (USDA ERS Food Price Outlook, July 24, 2026 release — archived copy; the live page updates monthly). So the useful question on a produce invoice isn't "did my price go up?" — it's "did my price move with the market, or on its own?" Swings and creep look identical on a single invoice. They come apart the moment you add a date axis and a reference point. Here's how to tell them apart from your own paper.
Produce really does swing — the official numbers
The USDA Economic Research Service publishes a monthly Food Price Outlook; the July 24, 2026 release (reflecting June data) forecasts, for 2026:
- Fresh vegetables at retail: +6.8% (prediction interval 4.3% to 9.5%)
- Fresh fruits at retail: +2.0% (interval 0.2% to 3.9%)
- Farm-level vegetables: +26.7% (interval 12.9% to 44.8%)
- Farm-level fruits: −3.8% (interval −9.3% to 2.3%)
(All four from the same archived ERS release.) Two things in that table matter for reading your invoices. First, the farm-level numbers — closer to what your distributor pays — move far harder than the retail numbers: vegetables +26.7% at the farm level against +6.8% at retail. Whatever cushioning happens between the field and a supermarket shelf, your distributor sits nearer the volatile end. Second, fruit and vegetables aren't even moving the same direction at the farm level this year. "Produce is up" is never one fact; it's a different fact per commodity.
What a real swing looks like
A market swing has a cause you can point to, hits buyers across the market, and shows up in public price series. Tomatoes in early 2026 are a clean example. The U.S. Commerce Department terminated the tomato Suspension Agreement with Mexico effective July 14, 2025, putting an antidumping duty on fresh tomatoes from Mexico — 17.09% for most exporters (Federal Register). By spring, consumer tomato prices rose 18% from February to March 2026 and stood 24% above a year earlier, with New York and Philadelphia terminal markets showing strong increases (Rutgers Plant & Pest Advisory, April 2026).
If your tomato line jumped that spring, the market is the first suspect — but "first suspect" is still a thing to verify, not a conclusion: whether the duty touched your SKU depends on where your product was sourced, and a market-wide jump is also a moment when an extra margin can ride along with the headline. The check runs the same direction as before: does your increase roughly track a public series that matches your region and product, and does your price step back down when the series does? A rep can't waive a tariff — but they can tell you which part of your increase was the tariff.
What quiet creep looks like
Creep is the opposite profile: no headline, no public-series move behind it — the same SKU just costs a little more than it did a few invoices ago, and a little more again after that. The June 2026 ERS data shows why the two are easy to confuse if you only glance at totals: retail fresh-vegetable prices actually fell 1.2% from May to June 2026 while sitting 9.9% above June 2025 (same archived release). Month-to-month wiggle and year-over-year drift are different signals — and creep lives in the drift. Why distributor prices creep covers the mechanics on the distributor side; the produce-specific problem is that the category's genuine volatility gives creep excellent camouflage.
Telling them apart from your own paper
Three steps, all doable with your own invoices:
- Track the same SKU at per-unit prices. Same SKU, same pack size, unit price over time — that's the creep detector. If the pack changed, redo the per-unit math first (how to read the line). A price that "bounces" is normal in produce; a floor that quietly rises across months is the thing to catch.
- Get a market reference point. USDA Market News publishes terminal-market wholesale produce prices — the Atlanta terminal market vegetable report is one example. Distributors publish their own read of the market too: Sysco's "Market Corner" produce updates have appeared on a near-weekly cadence in early 2026 (example issue), and produce-management firms like Produce Alliance put out weekly market reports. A market-update genre that runs weekly is itself a tell for how fast this category moves.
- Ask the direction question. A swing moves both ways: if the public series eased and your invoice didn't follow within a delivery cycle or two, that's the conversation — "The market report shows this coming down; my last two invoices don't. Walk me through it." Creep lacks that alibi — but a flat reference series is a signal to investigate, not proof by itself: no public series is a perfect match for your region, grade, pack, or freight, and any of those can move independently. What the divergence does is move the burden of explanation into the conversation — your rep may have a good answer, and now they need to give it.
The honest caveat
None of this tells you whether your level is right — a price can track the market's every move and still be high for your volume and region. That takes a reference point beyond both your paper and the public series: what other operators actually pay. That part is what we sell: email your last 3 invoices and the free top-3 benchmark comes back within 48 hours, your per-unit prices set against prices other operators have paid or been quoted. The full audit also tracks unit prices across your invoices for every SKU your invoices list, and flags the ones that climbed, with the prices and dates — every finding tied to the specific invoices it came from — invoice number and date, plus the SKU on item findings where your invoice lists one. No citation, no claim. Prefer DIY? The self-audit method is free and takes about an hour.