A side-by-side Van Westendorp comparison: 86 active Noah's customers vs. 119 California breakfast buyers. Where customer expectations align with the broader market, and where the gap is large enough to be a brand-loyalty signal.
Noah's customers expect to pay $1.24 to $1.44 less than the broader market at every classic SKU we tested. They are not a less-price-sensitive base. They are a more-loyal one with a sharper sense of what Noah's "should" charge.
57% of customers, when asked who they're comparing breakfast sandwich prices to, answered some version of "feels right" without naming a competitor. The market named McDonald's first (39%). When Noah's own customers think about pricing, they aren't anchoring on fast food, they're anchoring on Noah's.
The most-loyal customers expect the lowest classic-bagel price. Market OPP is $4.96; customer OPP is $3.52. Both samples answered the same question about a bagel-with-cream-cheese, and both shopped at California breakfast spots in the past three months. The gap is purely a brand-relationship effect.
Customers carry a memorised price tag for what a Noah's bagel should cost. Their "would skip" threshold lands around $4.50, just below the market's "good deal" anchor at $4.96. Pricing a classic bagel-with-schmear at the market OPP risks triggering the "starting to feel pricey" response in the brand's most engaged segment.
The implication is not "drop classic pricing." The implication is that premium-tier upgrades need to do most of the work in lifting average ticket, because customers won't absorb meaningful classic-SKU increases without resistance.
"Three dollars for a bagel and cream cheese feels right. I'd buy two."
When asked what they're comparing breakfast-sandwich prices to, 57% of customers said some version of "what feels right" without naming a competitor. Only 17% named McDonald's. The market sample's top answer was McDonald's at 39%, with no respondents giving a "feels right" answer. This is the single most important finding in the comparison.
The market's pricing ceiling on breakfast sandwiches is shaped by McDonald's. Noah's customers have opted out of that comparison. When customers say a price feels off, they're not benchmarking against the McMuffin, they're benchmarking against what they think a Noah's sandwich should cost.
The corollary: 49% of customers name breakfast sandwiches as the category where Noah's pricing feels most off (the highest-cited category in the customer survey). The gap between customer OPP ($5.54) and market OPP ($6.87) is real and is concentrated here.
"I'm not comparing to fast food. I'm comparing to what I think Noah's used to charge."
Customer OPP for a baker's dozen is $12.97, $1.24 below the market's $14.21. The gap is narrower than the bagel and sandwich gaps, but the dozen is the most-bought SKU among customers: 57% of Noah's customers report buying a baker's dozen (compared to 37% buying coffee). Price changes on the dozen reach a wider share of the customer base.
Customer dozen purchasing is anchored hard to Noah's Monday deal: respondents repeatedly referenced the Monday discount as their baseline reference price. The Monday-deal anchoring also explains why "lose Monday/discount deals" appeared in the would-stop-going analysis as a discrete concern, separate from "further price increases."
Practically: the dozen has the largest customer reach but the smallest price gap. The Monday-deal anchoring is a feature, not a bug. Removing or scaling back the Monday discount risks more attrition than the $1.24 OPP gap suggests on its own.
"Twelve bucks on Monday for a dozen is the whole reason I go on Mondays."
The clearest divergence outside the OPP comparison. The market study tested specialty bagels and a premium loaded sandwich, finding both clear at a 100% uplift over the classic ($3.50 → $7.00 bagel; $5.00 → $10.00 sandwich). The customer study asked about specialty cream cheese (the most-named real upgrade Noah's currently offers) and found a maximum acceptable price of $4.00, a 33% uplift from the $3.00 customer anchor.
The two samples were asked about different premium offerings (specialty bagels for market; specialty cream cheese for customers, since that's the upgrade they actually buy). But the structural read is consistent. The market is willing to engage with a substantial premium tier when the upgrade is visible and distinctive. Customers are tighter on what they'll pay for an in-category upgrade because their anchor is lower.
This is a real product-roadmap signal. The premium pricing room that exists in the market may not translate cleanly to Noah's most-loyal base. A specialty bagel priced at $7.00 (acceptable to market) is a 133% premium over the customer anchor of $3.00. That's likely outside the customer base's tolerance.
"I'll pay more for lox cream cheese, but not double. A dollar more, maybe."
Half of Noah's customers (49%) named breakfast sandwiches as the category where pricing feels most off, more than double the next category. Bagels were second at 24%. Coffee at 9%. The customer concern matches the OPP gap data exactly: sandwiches show the second-largest gap ($1.33) and are the most-cited friction point.
Two threads here. First, the customer-perceived problem area is sandwiches, the same category where the OPP gap is most consequential. The quantitative gap and the qualitative perception agree.
Second, 57% of customers define value as "what you get for the price" (quality, portions, experience), not as the menu price itself. Pricing-led messaging speaks to roughly 27% of customers; for the rest, the merchandising lever is portion, ingredient quality, and experience, not the headline number.
"The sandwiches got smaller and the price went up. That's the part that bothers me, not the price by itself."
56% of the customer sample is in the Bay Area; 17% in the Sacramento region; 16% in Southern California. The age distribution skews older, with 72% of respondents aged 45+ (and only 1% in the 18-24 bracket). Both findings are unweighted: this is the active customer sample as it was recruited from Noah's owned channels, and the demographic skew may differ from Noah's full transactional base.
The customer sample does not represent Noah's full transactional base. It is an engaged, opt-in, gift-card-incentivized subset reached through brand-owned channels (email, app push notification). The Bay Area concentration likely reflects both Noah's California store footprint and the survey's distribution mechanics. The older skew is consistent with a brand-loyal segment.
The implication for the OPP-gap finding: customer expectations are anchored by an audience whose price memory may run further back. The $3 bagel and $5 sandwich are not arbitrary numbers; they're price points from a few years ago that this base remembers and treats as baseline.
Two parallel surveys fielded between April 22 and April 28, 2026. A 119-respondent market study screened California breakfast and coffee buyers (Van Westendorp pricing for four classic SKUs and four premium upgrades). An 86-respondent customer study recruited active Noah's customers through brand-owned channels with a $5 gift card incentive. The same Van Westendorp questions were asked of both samples for the three SKUs Noah's customers buy regularly.
Market sample. 119 verified-complete California-resident respondents recruited through a third-party panel provider. All passed the same three-question screener (see below) plus a California residency qualifier. Conversational survey method.
Customer sample. 86 verified-complete active Noah's customers recruited through Noah's owned channels (email, app push notification) with a $5 gift card incentive on completion. No external panel involvement. Customers self-identified as having visited Noah's in the past 30 days at the time of recruitment.
All market-side respondents passed the same three-question screener before qualifying to take the survey:
California residency qualifier. The market study added a California residency qualifier on top of the three screeners above. Respondents outside California were screened out.
Customer sample qualification. Customer-side respondents were not asked the three-question screener (they were already known to be Noah's customers from the recruitment channel). They self-confirmed having visited Noah's in the past 30 days at the start of the survey.
| Product | Market n | Customer n |
|---|---|---|
| Classic bagel + cream cheese | 90 | 78 |
| Bacon, egg & cheese sandwich | 92 | 78 |
| Baker's dozen bagels | 84 | 47 |
Per-product n variation reflects price-validity exclusions only (respondents whose answers were non-numeric, missing, or outside plausible ranges were dropped from that product's curves, not from the broader sample). All percentages reported in the body are calculated against the full sample (n=86 customer, n=119 market) unless otherwise stated.
Validation was IP-address-based for both samples. Duplicate IP addresses, IP addresses flagged by the panel provider as low-quality, and IP addresses associated with internally-inconsistent Van Westendorp answers were removed before analysis. On the market side, 18 of 137 sessions (13.1%) were excluded. The customer sample had no similar exclusion rate, as it was recruited through authenticated brand channels.
For each price on the analysis grid, three cumulative percentages were computed: the share of respondents whose "good deal" threshold was at or above that price (descending), whose "starts to feel pricey" threshold was at or below that price (ascending), and whose "would skip" threshold was at or below that price (ascending). OPP, IPP, and PME are derived as in the standalone market report.
Methodology variant note. The market study used a 3-point Van Westendorp variant (good deal / starts pricey / would skip; no PMC). The customer study used a 4-point variant (the standard), which adds a "too cheap to be good" question. For the side-by-side curves in this report, only the three points common to both studies are charted. The 4-point customer PMC is not displayed in the comparison.
Free-text responses (sandwich anchor comparison, value definition, pricing-off categories) were classified using a regex-based bucket assignment. The same classifier was applied to both samples for any question asked in both surveys. Multi-coded buckets are flagged as such in the relevant charts; single-select questions report unique-respondent percentages. All classifications can be re-run from the source CSVs.