The cyclospora scare took about 5 points off the quarter

Sweetgreen ran +4.8% ■ in the first two fiscal weeks. It then fell to −12.3% ■ from Jul 13, the week the CDC alert went out ◆, while restaurants overall kept growing.

Against restaurants, Sweetgreen's gap widened by 9.8pp ⊘ over the seven outbreak weeks. That is about 5pp of the quarter's −9.7% ■; management put the hit to July comps at about 600bp ▲.

Sweetgreen does not use the implicated iceberg lettuce ▲. Its decline was the same in outbreak states and elsewhere ■, so this was fear of fresh produce, not exposure.

The lost sales were visits, not smaller orders

In Q3 to Sep 18, card transactions fell −8.7% ■ while ticket slipped only −1.5% ■. In Q2 it was the reverse: transactions +0.4% ■, ticket −4.6% ■.

Sweetgreen has taken no price increase in over a year ▲, so ticket moves are mix. The Q2 split matches the disclosed 4.2% mix drag ▲; the Q3 damage is guests staying away.

Salad chains fell together, but Sweetgreen was already lagging

Every fresh-produce chain slowed: CAVA to +2.5%, Just Salad to −9.9% and Taco Bell to −13.1% ■. Fast casual as a whole ran −0.9% ■.

Sweetgreen trailed fast casual by 7.4pp in Q2 ■ and 9.2pp in Q3 ■. The outbreak widened a gap that was already there. CAVA is the chain its customers most often also use ■.

Foot traffic missed the dip; card spend tracks revenue far better

Visits per tracked store were flat in July ■ while card spend fell 9.9% ■, so the two datasets disagree. The store roster has not grown since February ■, which damps the visit read.

Scored against six reported quarters, card spend correlates with revenue at 0.98 ■ ⊘ and visits per store at 0.39 ■ ⊘. Sweetgreen also cut digital advertising sharply from July ■.

What could move this number

The correction. Since Q4 FY24 reported revenue has run about 7pp above card spend, for reasons we cannot name. If the gap fell back to its 18-quarter average of 3.0pp, the estimate would be about −6.6% ⊘.

The rule we chose. Other ways of applying past gaps give −1.6% to −2.7% ⊘. Dropping the largest quarterly gap gives −2.9% ⊘.

The worst week. Removing the single worst fiscal week (Jul 13 to 19) lifts the estimate to −1.4% ⊘. The last three days of the quarter are unobserved, about 3% of it.

Strongest evidence the other way. Foot traffic shows no July dip, and an even split of the full-year guide implies about +3.5% for Q3 ⊘.

This preview reads revenue only. It says nothing about margin, EPS or the stock.

Every alternative we scored

Same card series, different ways of applying past gaps. Each was scored on the quarters before it.

Spread across rules: 1.1pp, inside the 3.9pp average miss, so the choice of rule does not change the read.

How to hold this number

It is a revenue growth rate for the quarter, from card spend on 97% of its days. The 3.9pp average miss covers ordinary quarters; it does not cover the gap snapping back to its older level, which would move the read toward −6.6%. Without the worst week it is −1.4%.

What card spend can and cannot see

Sweetgreen is US-only and every restaurant is company-operated, so the card merchant is the reporting company.

The ticker and the brand return identical spend, so there is one brand to read. Card spend was about 0.75% of revenue in June, in line with CAVA and Chipotle at about 1.1%.

The Q4 FY24 shift is unexplained

Through Q3 FY24 the gap swung between −4pp and +8pp; since Q4 FY24 card spend has run 6 to 11pp below revenue. It holds for the same cardholders ■ and not for Chipotle ■ ▲. With no named cause the series stays in use, corrected on post-shift quarters only.

Per-quarter backtest

Every reported figure links to its SEC filing. Q4 figures are the 10-K year less the three 10-Q quarters.

What this preview will and will not claim

Candidate events, searched before the data was pulled

Dates from primary sources, confirmed before becoming chart markers.

Searched and found nothing material: Chipotle and CAVA releases, macro shocks inside the quarter. A single menu launch is not detectable at this panel's weekly swing.

What this means

Absent the outbreak, Sweetgreen's quarter would have looked like Q2 in card spend, about −4%. The open question for Q4 is whether visits return now the headlines have faded; one week of data leans that way but cannot yet be called.

Q2 against Q3, with unit growth

Card spend YoY, panel weeks (Q2: 13 weeks to Jun 26; Q3: 12 weeks to Sep 18). Unit growth from each company's latest filing.

Sweetgreen's year-ago count is derived: 287 at Jun 28, 2026 less net openings of 2, 4, 15 and 6 in the four quarters before. Chopt was tested and is too thin in the panel to read.

What this means

The quarter's shortfall is mostly the outbreak, which Sweetgreen shared with every fresh-produce peer. The part that is Sweetgreen's own is the gap to fast casual that was already there in Q2 and widened in Q3.

Data limit

The full ranking of every brand Sweetgreen's customers use is too large to return here, so this view covers a 25-company set chosen before the pull. Rank movement against last year was not run.

The agreement table

Direction against card spend over the same window, Sweetgreen with CAVA as the peer.

Why the two disagree

The foot traffic roster has held Sweetgreen at 280 stores since February (287 disclosed ▲) and CAVA at 422 (476 disclosed ▲), a confirmed data limit. Its season and weather adjustment also damps sudden shifts. Card spend is the stronger read.

What this means

Less advertising during the outbreak may have deepened the traffic loss, but its effect cannot be separated from the scare itself. Ask whether marketing has resumed; it bears on how fast visits return.

Checks on the data

How the panel was corrected, and what the alternative gives

What could move this number

Where the data comes from

Out of reach: orders placed through delivery apps, catering, office pickup points, gift-card redemptions and cash.

What we looked for before we looked

Break claims in this report: two. The Q4 FY24 gap shift is unexplained and the series stays in use; the frozen foot traffic roster is confirmed against filed store counts. The next-quarter read was not run: Q4 FY26 starts Sep 28 and has no settled weeks.