JUNE 26, 2026, XETRA:HFG

HelloFresh: Betting Against a Second Half That Isn't Coming

#b2c #insidertrading #creditsignal #mealkits

Executive Summary

HelloFresh is a subscription business whose subscribers (especially the newer ones) are leaving, run by a management team that responded to the exodus by changing how it counts. The core meal-kit franchise shrank 8.5% in constant currency in Q1 2026 and the group 7.7%, the active customer base is still contracting, and the one search query gaining momentum around the brand is the one asking how to cancel. Yet full-year guidance promises a revenue decline of only 3 to 6%, which arithmetically requires the second half to run roughly flat after 6 consecutive quarters of decline, on a marketing budget that was just cut by €62 million in a single quarter. The equity still prices meaningful credit for that recovery. The credit market has already stopped extending it: S&P cut HelloFresh to junk in April, citing exactly the volume and competition problems the equity story asks investors to look past.

Our short thesis is that the mid-August H1 report is the scheduled collision between hope-weighted guidance and measured demand. Positive free cash flow, real as it is, comes from cutting spend into a shrinking funnel, a maneuver that works until the funnel notices. We are short around €4 with a downside zone of €2.80 to €3.20, the level where the equity prices the meal-kit segment as the melting business it is rather than the recovering one management describes. The complications are real and we address them below: the short is crowded, and insiders are net buyers. The horizon is weeks, not quarters, and the exit is predefined.

What the Company Claims vs Economic Reality

The narrative is disciplined and, on its own terms, coherent. Management retired the growth story and replaced it with a quality story: fewer but "high-value" customers, average order value up €4.20 in constant currency, contribution per box rising, a €140 million cost program, exits from Spain and Italy, and a "ReFresh" product cycle doubling recipe choice in the US and Nordics. Q1 2026 delivered the 4th consecutive quarter of positive free cash flow, €49 million, and the ready-to-eat segment narrowed its EBITDA loss by €18 million with net promoter scores at their best since 2023. The pitch: the decline is managed, the trough is 2026, and the second half will show it.

The economics tell a simpler story. Revenue fell 7.7% in constant currency to €1.68 billion, the 6th consecutive quarterly decline, and adjusted EBITDA was €24 million, a margin of 1.4%, weather excuse included. The active customer base is still shrinking because, in the company's own words, new customer volume is insufficient to offset losses. That sentence is the entire problem. A subscription business with an inverting funnel is not a margin story, it is a countdown, and the cash flow being celebrated is largely the €62 million of marketing that was not spent, which is to say, next year's customers being consumed as this year's free cash flow. The credit market read it first: S&P cut the rating to BB+ in April, out of investment grade, citing weakening volumes, pressured consumers, competition, and technological disruption. The balance sheet now carries €750 million of long-term and €830 million of short-term debt against €2.2 billion of total assets. The fortress years are over.

Core Short Thesis

1. The funnel is inverting, structurally. Group revenue -7.7% cc, meal kits -8.5% cc, and an active customer base the company admits is shrinking faster than acquisition replaces it. Third-party signals agree: US search interest sits at a fraction of its 2022 peak, app-store rankings have decayed out of the top 100 in Food & Drink, review sentiment is deteriorating, and the cancel-intent query is the only trend line pointing up. Evidenced, from the Q1 report and public alternative data.

2. The guidance requires a second half that the data does not. Full-year 2026 guidance of -3 to -6% cc, with Q2 guided "roughly in line" with Q1's -7.7%, implies H2 running between roughly -4% and +2%. That is a 5 to 9 point acceleration, to be delivered with €62 million less quarterly marketing, a product cycle that costs 150 basis points of gross margin before it earns anything, and a consumer S&P describes as income-pressured. The guidance is evidenced; its implausibility is our inference, and it is the falsifiable core of the trade, tested in mid-August.

3. Credit has repriced, equity has not finished. The April downgrade to BB+ took HelloFresh out of investment grade 13 months after the market still argued about its net cash. €830 million of short-term debt now sits against a business whose EBITDA guidance midpoint is €400 million and falling. When the rating agencies and the equity narrative disagree about the same cash flows, the agencies are usually late, not wrong. Evidenced.

4. Disclosure credibility is impaired. The "Active Customers" KPI was retired in 2024, precisely as it deteriorated, in favor of a "high-value customer" framing the company defines itself. Grizzly Research documented, in November 2025, a CEO who had pledged roughly 3 quarters of his stake against loans, a co-founder who sold call options on his own stock, and over €100 million of buybacks executed into a falling price. The company rejected the report's allegations, and we treat the single-sourced claims as allegations. But the KPI retirement and the buyback timing are documented facts, and they teach a simple lesson: when this management describes the second half, apply a discount. Mixed evidence, flagged accordingly.

5. The cost program is a floor for cash, not for the equity story. €140 million of savings, marketing at 21.8% of revenue with stricter ROI thresholds, and country exits all support near-term free cash flow. None of them adds a customer. A declining subscription base with improving margins is a harvest, and harvests get valued on the remaining crop, not on the efficiency of the combine. Evidenced on the program; the valuation consequence is our judgment.

The Funnel Wrote the Report

Our favorite exhibits in this name were not produced by any analyst. They were produced by the customers, as the business is mostly B2C. Search interest for the brand has bled out over 4 years, from pandemic euphoria to below pre-COVID levels of organic traffic. Meanwhile the query family that keeps climbing is the unsubscribe intent: how to cancel, how to pause, how to get a refund. Review platforms tell the same story in prose, including Trustpilot (a source we try to discount after the "Trustpilot Mafia" affair) which makes the negative skew there all the more notable. A 2022 cohort analysis already showed roughly 9% 12-month retention, and everything measured since suggests the funnel got worse while the company stopped publishing the number that would prove it. When a business retires a KPI, the KPI usually retires the business first.

The Guidance Math

This section is the whole trade, so we show the arithmetic. Take the guidance midpoint of -4.5% for the full year. Q1 printed -7.7%. Q2 is guided "roughly in line with Q1," call it -7.5%. For the full year to land at midpoint, the second half must average roughly -1.5%; to reach the top of the range, H2 must be positive. HelloFresh has not printed a positive constant-currency quarter since the pandemic cohort rolled off. Management's bridge to this inflection is the ReFresh cycle, easier comparables, and high-value customer retention. Our bridge review notes that ReFresh costs margin before it earns revenue, that comparables ease only modestly, and that the acquisition engine has been deliberately throttled. One of these 2 bridges collapses in mid-August, and only one of them is load-bearing for a €4 stock.

The Insiders Are Buying. We Stay Short Anyway.

Intellectual honesty first: the single strongest argument against this short is filed in the insider dealings register. Executives, including at board level, made personal purchases at the end of March 2026, and buys have outnumbered sells this year. We take insider flow seriously, and doubly so here, because Grizzly's November 2025 report alleged this management traded well around undisclosed information before the Q3 2025 miss. If they front-run bad news, the symmetric worry is that they front-run good news too.

Three considerations keep us short despite it. First, the sums are small relative to the wealth already extracted and, in the CEO's case per the Grizzly allegations, a stake heavily pledged against loans creates its own incentive to defend the price with visible confidence. Second, insider buying at companies in structural decline reliably marks belief in the cost program, not in the funnel, and the cost program is not what the mid-August print will test. Third, the same signal was available before the Q1 print in May, and the stock still fell 6.6% on the results and took a Stifel downgrade to Sell the same week. We weight the signal by sizing smaller, not by covering. If insiders buy again in size after a guidance cut, that is a different conversation, and we will have it.

Valuation and Downside Framework

We do not need a heroic bear case, and we explicitly do not need the Grizzly insolvency scenario, which we consider unsupported on a 12-month view given positive free cash flow. We need only the market to stop paying for a second-half inflection that does not arrive. At €4 the market capitalization still embeds a stabilizing meal-kit business plus a break-evening Factor. Reprice the meal-kit segment as a melting annuity in the manner of declining consumer subscription peers, give Factor fair credit for its improving trajectory, and haircut for a junk-rated balance sheet with €830 million of short-term debt, and the equity clears in the €2.80 to €3.20 zone. That zone requires no covenant event, no rating action beyond what already happened, and no recession. It requires a guidance cut, or even just "lower end of the range" language delivered in a German monotone.

Catalysts, Timing, and Path

The clock runs to mid-August, when the H1 2026 report forces management to reconcile a first half running near -7.5% with a full-year guide of -3 to -6%. Before then, the tape can move on customer loyalty and churn data points, on Factor's US supply constraints, on any further analyst capitulation following Stifel and UBS, and on rating agency follow-through now that the investment-grade floor is gone. Index reviews remain a live pressure point, as Grizzly flagged for the MDAX last winter. We would also watch the buyback: a pause removes the most mechanical bid under the stock. Unlike our Redwire trade, there is no single scheduled de-catalyst here. Instead there is one big binary in August surrounded by a drip of confirming data. The plan is built accordingly: hold through the drip, decide at the binary.

Risks to the Short

The honest bull case: HelloFresh is the global category leader with 4 consecutive quarters of positive free cash flow, a cost program that is visibly landing, and in Factor a second act whose unit economics are improving fast, with tenured customers growing double-digit and the EBITDA loss narrowed by €18 million in one year. If ReFresh lifts conversion even modestly while comparables ease, the guide survives August and the stock re-rates on "trough confirmed." The company is not running out of cash on any horizon this trade lives on, so there is no forced event such as solvency.

The tactical risks are sharper and they have grown since inception. Short interest is surging, which means we are no longer early and the exit door is narrower. Insiders are net buyers, addressed above but not dismissed. At roughly €700 million of market capitalization with founders still engaged, take-private or strategic interest is a permanent tail risk, and it is the scenario that skips the €5 handle entirely. A squeeze into any non-terrible print could retrace the move from €4 toward €5.50 in days. We manage this through sizing already on, a hard review at the H1 print, and a standing rule: this position does not survive an in-line report, because the thesis says the report cannot be in-line. If it is, we were wrong, and we pay.

Conclusion: Best Served Cold

HelloFresh the operation is a competently managed retreat. HelloFresh the equity story still asks to be paid for an advance. The gap between those 2 is the trade. The customers are measurably leaving, the credit market has re-tagged the balance sheet as junk, the sell side is migrating from Hold to Sell, and the guidance depends on a second-half acceleration that nothing in the observable funnel supports. Against that stand a crowded borrow and buying insiders, which is why this is a held position with a defined exit rather than a table-pounding add. What would change our mind: a positive, important newsflow before announcement, in-line H1 with active customer stabilization, or insiders buying in size a second time. What would confirm it: one guidance sentence containing the words "lower end", or one more quarter in which the only growing metric is the cancellation query.