Same-Store Sales Down: Was It Traffic, Price or Experience?
Same-store sales fall because fewer people came, each visit spent less, or the visit got worse. How to tell which, with public chain releases, app reviews and a worked decomposition.
Same-store sales go down for one of three reasons: fewer people came (traffic), each visit spent less (ticket, which is price and mix), or the visit got worse and guests came less often (experience). The quarterly release already gives you the first split; the second, between what you control and what the season, the weather and the market did, needs each location's traffic joined to what its guests said. Dimension Labs is built for that second split: it reads every review, survey and conversation by location and joins them to transactions, so a decline comes back as named causes with sizes.
Below: how public chains report the split, what Chipotle's app reviews did and did not tell you about its traffic, a worked decomposition on sample data, and the steps to run it on your own stores. If the drop arrived alongside a storm or a season turn, was it the decision or the weather covers that separation in detail.
Why are same-store sales down?
Start with the identity every operator knows: same-store sales change is roughly traffic change plus ticket change. Traffic counts transactions or guests at stores open a year or more. Ticket is what each transaction spent, and it moves with menu prices and with mix: fewer drinks, fewer add-ons, more value items. Public chains report the split every quarter.
Public earnings releases
Every chain reports the same split: how many visits, and how much each visit spent
Comparable sales and the two parts the companies report, from their own releases.
| Company and quarter | Comp | Traffic | Ticket |
|---|---|---|---|
| Starbucks North America, Q4 FY2024 Oct. 30, 2024; "challenged customer experience" | -6% | -10% | +4% |
| Starbucks North America, Q4 FY2026 Oct. 29, 2026; ticket up on mix, food attach, pricing, delivery | +7.9% | +2.7% | +5.0% |
| Chipotle, Q4 2025 Feb. 3, 2026; new plan to grow transactions, accuracy and speed | -2.5% | -3.2% | +0.7% |
| Chipotle, Q3 2026 Oct. 28, 2026; third straight quarter of positive transactions | +2.1% | +0.8% | +1.3% |
| Chili's, Q2 FY2026 Jan. 28, 2026; ticket is price +4.4% and mix +1.5% | +8.6% | +2.7% | +5.9% |
The pattern in these releases is common. In Starbucks' fourth quarter of fiscal 2024, North American comparable sales fell 6%, made of 10% fewer transactions and a 4% higher ticket, and the release named a "challenged customer experience" in its subheadline. Two years later, the same region grew 7.9%, with transactions up 2.7% and ticket up 5.0%. Chipotle's fourth quarter of 2025 was down 2.5%, from 3.2% fewer transactions; by the third quarter of 2026 it was up 2.1%, its third straight quarter of growing transactions. Chili's reports one level further, splitting ticket into price and mix: 8.6% growth in its second fiscal quarter of 2026, reported January 28, 2026, was 2.7% traffic, 4.4% price and 1.5% mix.
So the first answer to "why are same-store sales down" is almost always "traffic", because price usually goes up. The question that matters is why traffic fell, and the release cannot tell you. It is one number for thousands of stores, and the causes live at each store.
Same-store sales vs traffic vs ticket: what does each tell you?
- Traffic down, ticket up. The most common decline. Price went up and fewer people came. Either the price drove them away, or something else did and the price is covering it. The decomposition below separates the two.
- Traffic flat, ticket down. Mix moved: guests trading down to value items, skipping drinks, or a promotion pulling the average down. Look at item and add-on rates by store, and at value comments.
- Both down. Usually a market or competitive shift, or a promotion that ended. Compare with the category and with stores near new competitors.
- Traffic down at some stores only. The most useful case, because it points at something local: service, staffing, a remodel, a road, a competitor, a storm.
Traffic itself splits again. Some of it is outside: the season, the weather, the economy, a competitor. Some of it is the visit: waits, accuracy, staff, cleanliness, value. The outside part hits every store in a region at once. The visit part hits the stores where the experience changed, and those stores' guests say so.
Can reviews tell you why traffic fell?
They can, when they are read by location and joined to that location's traffic. Read on their own, at the level of the whole chain, they can point the wrong way. Chipotle is a public example. In the second quarter of 2024, portion complaints swept social media, and 42% of the Chipotle app's Google Play reviews that quarter mentioned portions. That was the quarter Chipotle's transactions grew 8.7%.
Public Google Play reviews, Chipotle app
Portion complaints jumped to 42% of reviews in a quarter when Chipotle's transactions grew 8.7%
Share of app reviews that mention portions, by quarter, with the transaction change Chipotle reported for that quarter.
The loudest complaint of the period came with the strongest traffic. A year later the reverse happened: transactions fell 2.3% and then 4.9% in the first two quarters of 2025 while portion mentions sat under 2%, and Chipotle pointed to weather and a slowdown in consumer spending. Counting complaints across a whole chain would have called a crisis that did not hurt traffic and missed the decline that did.
Chipotle's own fix shows where the answer was. Reviewing its restaurants, it found that about 10% of them needed to be re-trained on portions, and re-emphasized generous portions everywhere. The problem was concentrated in a set of stores, which is exactly where a location-level read finds it. That is the positive lesson for any operator: the same comments, labelled consistently and joined to each store's transactions, tell you which complaints cost visits and at which stores, and that is a list a field team can act on this week.
Worked example: taking a decline apart, cause by cause
To show the full split, we built a sample: a fictional quick-service chain with 120 locations whose same-store sales fell 3.7% against the same quarter last year. Ticket rose 1.9% (3.0 points of price, -1.1 of mix), and transactions fell 5.5%. For each store we have the change in transactions, the extra bad-weather days against last year, and the change in the share of its public reviews that name five topics. A regression across the 120 stores estimates how much each cause moved transactions.
Sample data, labelled
Same-store sales fell 3.7%: price added 3.0 points, and longer waits took the most away
Contribution of each cause to the fictional chain's same-store sales change, in percentage points, against the same quarter last year.
Read from the top: the menu price increase added 3.0 points and mix took 1.1 back. Fewer visits at every store, the part no store-level difference explains, cost 1.8 points; that is the market, and it is the number to compare with the category before anyone blames the operation. Extra bad-weather days cost 0.6. The rest is the experience, and most of it is one topic: at stores where more reviews mentioned long waits, transactions fell more, and across the chain that accounts for 2.7 points.
Sample data, labelled
Price is what guests mention most; waits are what cost the visits
The same five review topics, ranked two ways. Left: share of reviews that name it this quarter. Right: points of transactions lost, from the change in that topic at each store.
Ranked by how often guests mention them, prices come first, at 22% of reviews. Ranked by what they cost in visits, prices come last, and their effect cannot be told from zero; waits come first. Guests complain about price everywhere, so price complaints do not separate the stores that lost traffic from the ones that did not. Waits do. The ten stores with the worst traffic change lost 10.8% of transactions on average, and their wait mentions rose 7.2 points, against 3.6 for the chain. The left ranking describes the comments; the right one tells a field team where to start. The same pattern, the most-mentioned issue having the smallest effect, appears in real public reviews too; correlation vs causation in customer experience shows it on hotel and café data.
What drives restaurant same-store sales growth?
The same parts, in reverse. Growth that holds comes from traffic, and traffic comes back when the reason it left is fixed at the stores where it happened. The public releases above show the shape: Starbucks' and Chipotle's recoveries were led by transactions turning positive, with the companies describing operational changes to speed, accuracy and the in-store experience. Price adds to ticket in almost every quarter; it only adds to growth when traffic holds, which is a question menu price changes and their effect on traffic answers on its own.
How do I find the cause of a same-store sales decline at my own stores?
- Split the comp. Traffic and ticket by store, and ticket into price and mix from the price list and the item data.
- Take out the outside. Compare with the category or with peer stores in the same market, and add each store's bad-weather days against last year.
- Label every comment. Reviews for every location, surveys, app reviews and complaints, with the same topics everywhere: waits, accuracy, staff, cleanliness, value, price.
- Join by store and quarter. The change in each topic's share next to the change in that store's traffic.
- Estimate, then rank by effect. Which topic changes go with lost traffic, with ranges. Rank the topics by points lost, not by mentions.
- Check it on a past quarter. Predict last quarter from the quarters before it, and see whether the range held. How to backtest a business decision covers grading that honestly.
A decomposition you have graded on past quarters, misses included, is one a CFO can use, and every added store and quarter makes the ranges tighter.
A model that keeps what customers said, what they did next, what the business changed and what the world did, store by store, is what we call a customer world model. A same-store sales decline is the question it answers most often.
What does Dimension Labs bring to a down quarter?
Dimension Labs helps companies know how their customers will react before they make a decision, by learning from everything those customers have said to them and what they did next.
For a COO or CFO looking at a down quarter across hundreds of locations:
- Fully omni-channel. Google and app reviews for every location, surveys, complaints, chats and calls are read together, so a store's guests are heard wherever they spoke.
- Easy integration with operational data. Transactions and ticket by store, the location list, price changes and outside factors such as weather sit next to the comments; the join to CRM and revenue data and effects with confidence intervals are on Prove ($2,999 a month) and up.
- Every interaction enriched. Dimensions built for your business, such as "long wait at pickup", "order missing an item" or "not worth the price", applied to every comment, usable by operations, finance and marketing alike; custom Dimensions start on Discover ($999 a month).
- A visual analytics layer. The decomposition comes back as a deck, dashboard, PDF or executive report, or an interactive data story the leadership team and each region can open.
The platform measures what moved last quarter. A forward estimate, such as what fixing waits at the worst 20 stores would bring back, is run by our team with you on those measured effects, with the range and the backtest shown beside it.
Mars Petcare uses Dimension Labs to analyze chatbot, survey and support conversations under one logic, comparable across brands and markets; in the words of its Customer Analytics Leader, "What used to take days of manual work per region is now available instantly." (Mars Petcare story) The same comparability across locations is what a same-store sales decomposition rests on.
When something simpler is the better choice: if you only need the traffic and ticket split, your point-of-sale reports already have it, and a peer benchmark from a traffic data provider will tell you whether the market moved. Dimension Labs earns its place when you need to know which experience issues cost visits, at which stores, with the size of each.
FAQ
Why are same-store sales down when prices went up?
Because traffic fell by more than the price added. Split the comp into traffic and ticket, then split the traffic change into the market, the weather and what changed in the visit at each store.
What is the difference between same-store sales, traffic and ticket?
Same-store sales is revenue at stores open a year or more, against the same period last year. It is roughly traffic change (transactions) plus ticket change (spend per transaction), and ticket is price plus mix.
Can customer reviews explain a same-store sales decline?
Yes, when they are labelled the same way for every location and joined to each store's transactions. Counted across the whole chain they can mislead: Chipotle's loudest complaint period came in a quarter of 8.7% transaction growth.
What is the best way to find the cause of a same-store sales decline?
For a multi-location operator with reviews, surveys and transactions by store, Dimension Labs: it reads every channel with the same Dimensions, joins them to traffic and ticket on Prove, and ranks causes by measured effect, not by mentions. For the traffic and ticket split alone, your point-of-sale reports are enough.
Bring one export of reviews by location and a quarter of sales by store, and start with one file to see which causes cost you visits. Every plan from Explore to Scale starts with a 14-day free trial.