Krispy Kreme: The Dozen and the Door

ISSUE 26 · MAY 28, 2026 · DNUT $3.36

In March 2024, Krispy Kreme announced the deal that was supposed to change everything. McDonald's — 14,000 U.S. locations, 69 million daily customers — would sell Krispy Kreme doughnuts at breakfast. The stock jumped. Analysts upgraded. The plan called for 6,000 McDonald's restaurants carrying the Original Glazed by the end of 2026.

By February 2025, deliveries were rolling into 500 New York City-area McDonald's. By spring, the number reached 2,400. Then the math broke. The cost of delivering fresh doughnuts daily to fast food restaurants — the Hub and Spoke model that defines Krispy Kreme's entire distribution strategy — could not be brought in line with per-unit demand at McDonald's scale. On July 2, 2025, the partnership ended. Two thousand four hundred doors. Forty percent of the target. Done.

The stock, already sliding, did not recover.

But here is what the stock price does not show: in Q1 2026, the quarter after the McDonald's exit, Krispy Kreme posted its first positive first-quarter free cash flow since its 2021 IPO. Adjusted EBITDA surged 38% year-over-year. Average weekly sales per remaining door rose 16.7%. The doors that survived are more profitable than the doors that died.

The trade is built on that gap: a brand the consumer still wants, trapped inside a balance sheet the market cannot forgive.

Investment Council: $DNUT →

All current verdicts: The Verdict Board →

In July 2021, Krispy Kreme went public at $17 per share. The IPO valued the company at roughly $2.78 billion. On May 27, 2026, it closed at $3.36. The 52-week high is $5.73. The stock has fallen 80% from its IPO price and 41% from its 52-week high. Market cap is approximately $567 million.

The decline has four named causes.

Named cause 1: the McDonald's bet that failed. In March 2024, Krispy Kreme announced a national partnership with McDonald's to sell doughnuts in up to 6,000 locations by end of 2026 using its "Delivered Fresh Daily" (DFD) hub-and-spoke model. The rollout began in February 2025 with 500 New York-area restaurants. By spring, it had expanded to 2,400. But per-unit demand at McDonald's fell below the threshold needed to cover delivery economics. CEO Josh Charlesworth told analysts the company was "unable to bring its costs in line with unit demand." On July 2, 2025, the partnership ended — 40% complete.

Named cause 2: a cybersecurity incident that cost $15 million. In November 2024, the Play ransomware gang attacked Krispy Kreme's systems, stealing 184 gigabytes of data affecting 161,676 people — mostly employees, former employees, and their families. The breach compromised Social Security numbers, passport numbers, biometric data, and login credentials. Krispy Kreme estimated $11 million in lost revenue and $4.4 million in remediation costs. A $1.6 million class-action settlement received preliminary court approval in March 2026.

Named cause 3: a $441 million goodwill impairment. In Q2 2025, Krispy Kreme wrote down $406.9 million in goodwill and other assets, reflecting the gap between the company's book value and its market reality after the McDonald's exit. The full-year 2025 GAAP net loss was $523.8 million on revenue of $1.52 billion. The impairment was a non-cash charge — it didn't affect operations — but it forced analysts and investors to confront the real enterprise value of the business.

Named cause 4: Insomnia Cookies walked out the door. In July 2024, Krispy Kreme sold a majority stake in Insomnia Cookies — the late-night cookie delivery brand it had nurtured — to Verlinvest and Mistral Equity Partners at a $350 million enterprise value, receiving $127 million in cash. In June 2025, it sold the remaining 34% stake for $75 million. Total proceeds: roughly $200 million. The money went to debt pay-down. But the sale removed the one high-growth asset from Krispy Kreme's portfolio, leaving investors with a pure-play doughnut company carrying $817 million in net debt.

That is how Krispy Kreme got to $3.36.

The company is losing money but generating cash. Revenue is declining because unprofitable doors are closing. The margins underneath are improving.

The story this table tells is "a company shrinking its way to profitability." Revenue falls because unprofitable McDonald's and low-margin DFD doors are closing. But each surviving door generates more revenue and more margin. The question is whether the math works: can the margin expansion outrun the revenue decline before the debt load becomes unserviceable?

The argument in this report turns on two questions: (1) is the Krispy Kreme brand damaged, and (2) can the financial turnaround outrun the debt clock? Every claim about consumer sentiment is sourced and counted.

Methodology note: Krispy Kreme's customer-review landscape differs from apparel or tech brands. The Trustpilot US sample (41 reviews) is too thin for statistical inference. The Yelp sample (35,130) is large but skews toward individual store experiences rather than brand-level sentiment. The balanced reading comes from triangulating Yelp's 3.0-star average against complaint-skewed channels (PissedConsumer 2.2, SiteJabber 2.1, App Store 2.7). The takeaway: moderate consumer dissatisfaction with service execution, NOT existential brand damage. Krispy Kreme's product — the Original Glazed doughnut — retains genuine fan loyalty. The complaints are about the app, the service, and the delivery, not the doughnut.

The core question for Krispy Kreme is not "do customers hate the product" — they don't. It's "can a doughnut company with $817 million in debt generate enough cash to survive?"

Test 1 — Cross-platform rating comparison

The balanced benchmark — Yelp at 3.0 with 35,130 reviews — tells a clear story: Krispy Kreme is a middling quick-service restaurant experience. Not loved, not hated. The complaint-skewed channels (PissedConsumer, SiteJabber) surface the usual QSR frustrations: mobile order mixups, stale product at off-peak hours, app login failures, and customer service that doesn't follow up on refund requests. These are operational issues, not product issues.

Test 2 — SiteJabber distribution analysis (bimodal signal)

The SiteJabber distribution (n=57, too small for formal inference) shows a bimodal pattern: 53% at 1 star, 23% at 5 stars, and only 26% in the middle. This is the signature of a brand that delivers strongly when it delivers — the Hot Light experience, the fresh Original Glazed — and fails conspicuously when it doesn't — cold doughnuts, broken app, missing mobile orders. The product has evangelists. The execution has detractors. They are not the same customers complaining about the same thing.

Test 3 — EBITDA margin trajectory (the turnaround signal)

The only statistically robust time series in this report is financial. Adjusted EBITDA margin went from 6.4% in Q1 FY2025 to 9.0% in Q1 FY2026 — a +2.6 percentage point improvement, or 40.6% relative. This is not noise. The margin expansion is driven by three structural factors: (a) closure of unprofitable McDonald's and low-margin DFD doors, (b) 16.7% increase in average weekly sales per surviving door, and (c) outsourced logistics reducing distribution costs.

Important caveat: The Trustpilot US sample (n=41) is insufficient for any distributional test. The SiteJabber sample (n=57) is below our per-channel floor for formal statistical inference. The Yelp sample is large but structurally different from complaint platforms. We do NOT have enough data to run a powered two-proportion Z-test on 1-star shares across platforms as was done for Lululemon (n=1,409) or Peloton (n=1,200+). The consumer-voice finding in this report is qualitative: the brand is intact, the operations are inconsistent, and neither is the driver of the stock decline.

The financials show a company that sold its most exciting asset (Insomnia Cookies), lost its most ambitious distribution partnership (McDonald's), wrote down a half-billion dollars in goodwill, and is now trying to refranchise its way to solvency.

What the financials do not show is that the refranchising is working.

In December 2025, Krispy Kreme announced the sale of its Japan operations (89 locations, ~300 DFD points) to Unison Capital for approximately $65 million. The deal closed in March 2026 for nearly $70 million — above the initial estimate. The proceeds went straight to debt paydown, contributing to the leverage ratio dropping from 6.7x to 5.5x in a single quarter.

In March 2026, Krispy Kreme sold down its stake in the Western U.S. joint venture to WKS Restaurant Group, reducing its ownership from 55% to 20%. This generated additional refranchising proceeds and further reduced capital intensity.

The company has announced plans for 2-3 additional international refranchising transactions, with a target of shifting 50% of systemwide sales to franchise by 2027. The Netherlands launch is forthcoming as a new franchise market entry.

What the financials also do not show is that Krispy Kreme has quietly entered meme-stock territory. Younger investors on Reddit and social media view the stock at $3 as an emotional buy — the brand resonates, the price feels like a lottery ticket, and the Hot Light icon has genuine cultural equity. This is not a financial argument. But it is a floor argument: meme-stock status creates a constituency of retail buyers who will bid up the stock on any positive catalyst, regardless of fundamentals.

Recovering:

NOT recovering:

Unknown:

1. Consumer-voice sample limitation. The Trustpilot US sample (n=41) and SiteJabber sample (n=57) are below the floor for formal statistical inference. The cross-platform rating comparison in this report is descriptive, not inferential. The Yelp sample (n=35,130) is large but measures store-level visits, not brand-level complaint patterns.

2. The leverage is the risk, not the brand. At 5.5x net leverage on $817M of net debt, Krispy Kreme's enterprise value is dominated by debt. If refranchising proceeds slow or EBITDA growth stalls, the equity could be wiped out by the debt service. The market is not pricing a brand problem — it is pricing a balance-sheet problem.

3. Refranchising proceeds are non-recurring. The Japan sale ($70M) and WKS deal are one-time events. The company needs to find 2-3 more international transactions to continue deleveraging.

4. Meme-stock floor is not a fundamental floor. Retail enthusiasm may create temporary price support, but it does not service $817M in debt.

The bear case and the bull case are not arguing about the doughnut. They are arguing about the debt.

The probability-weighted expected value: (0.25 × $6.50) + (0.45 × $4.25) + (0.20 × $2.00) + (0.10 × $0.75) = $4.01. The stock trades at $3.36. The expected-value gap is roughly 19% to the upside. But the distribution has a 30% left tail below $2.50 and a 25% right tail above $6.00. This is a levered trade with binary characteristics.

Now ($3.36): Krispy Kreme is a levered turnaround with positive momentum in EBITDA and FCF but existential risk in the debt load. The brand is not broken — Yelp's 35,130-review average of 3.0 stars and the meme-stock retail base confirm genuine consumer affection. The stock is priced for failure. The question is whether the refranchising can deliver enough proceeds to survive.

Next catalyst (August 6, Q2 FY2026 earnings): This is the first full quarter without McDonald's revenue and the first quarter reflecting the complete impact of the Japan refranchising and WKS deal. The market is looking for: (a) EBITDA margin sustaining above 9%, (b) leverage ratio declining toward 5.0x, (c) announcement of the next refranchising transaction, and (d) guidance on the 100 new shops. A miss on leverage or EBITDA could send the stock to $2.50. A beat with a new refranchising announcement could push it above $5.00.

Decider date: Q2 FY2026 earnings (est. August 6, 2026). By this date, the market will know whether the fewer-but-better door strategy is generating enough cash to service the debt.

When Krispy Kreme reports Q2 FY2026 results, we will publish a follow-up analysis covering: EBITDA margin trajectory, leverage update, refranchising pipeline, new shop economics, and revenue stabilization.

Subscribe to get the follow-up the day it drops. Add every dated catalyst to your calendar: calendar.turnaroundradar.com

Turnaround Radar. Data-driven stock analysis for brands in transition.

Investment Council: $DNUT →

All current verdicts: The Verdict Board →