Proxy contest filings and AI analysis
| Ticker | Form Type | Company Name | Description | Filing Link | Filed At |
|---|---|---|---|---|---|
| LULU | DFAN14A | lululemon athletica inc. | Form DFAN14A - Additional definitive proxy soliciting materials filed by non-management and Rule 14(a)(12) material | View Filing | 3/31/2026 |
| LULU | DFAN14A | lululemon athletica inc. | Form DFAN14A - Additional definitive proxy soliciting materials filed by non-management and Rule 14(a)(12) material | View Filing | 3/27/2026 |
| LULU | PREC14A | lululemon athletica inc. | Form PREC14A - Preliminary proxy statements, contested solicitations | View Filing | 3/27/2026 |
| LULU | DFAN14A | lululemon athletica inc. | Form DFAN14A - Additional definitive proxy soliciting materials filed by non-management and Rule 14(a)(12) material | View Filing | 3/20/2026 |
| LULU | DFAN14A | lululemon athletica inc. | Form DFAN14A - Additional definitive proxy soliciting materials filed by non-management and Rule 14(a)(12) material | View Filing | 3/19/2026 |
| LULU | DFAN14A | lululemon athletica inc. | Form DFAN14A - Additional definitive proxy soliciting materials filed by non-management and Rule 14(a)(12) material | View Filing | 3/18/2026 |
| LULU | DFAN14A | lululemon athletica inc. | Form DFAN14A - Additional definitive proxy soliciting materials filed by non-management and Rule 14(a)(12) material | View Filing | 3/17/2026 |
| LULU | DEFA14A | lululemon athletica inc. | Form DEFA14A - Additional definitive proxy soliciting materials and Rule 14(a)(12) material | View Filing | 3/17/2026 |
| LULU | DFAN14A | lululemon athletica inc. | Form DFAN14A - Additional definitive proxy soliciting materials filed by non-management and Rule 14(a)(12) material | View Filing | 3/13/2026 |
| LULU | DFAN14A | lululemon athletica inc. | Form DFAN14A - Additional definitive proxy soliciting materials filed by non-management and Rule 14(a)(12) material | View Filing | 3/12/2026 |
| LULU | DFAN14A | lululemon athletica inc. | Form DFAN14A - Additional definitive proxy soliciting materials filed by non-management and Rule 14(a)(12) material | View Filing | 3/10/2026 |
| LULU | DFAN14A | lululemon athletica inc. | Form DFAN14A - Additional definitive proxy soliciting materials filed by non-management and Rule 14(a)(12) material | View Filing | 3/6/2026 |
| LULU | DFAN14A | lululemon athletica inc. | Form DFAN14A - Additional definitive proxy soliciting materials filed by non-management and Rule 14(a)(12) material | View Filing | 3/5/2026 |
| LULU | DFAN14A | lululemon athletica inc. | Form DFAN14A - Additional definitive proxy soliciting materials filed by non-management and Rule 14(a)(12) material | View Filing | 3/5/2026 |
| LULU | DFAN14A | lululemon athletica inc. | Form DFAN14A - Additional definitive proxy soliciting materials filed by non-management and Rule 14(a)(12) material | View Filing | 3/2/2026 |
| LULU | DFAN14A | lululemon athletica inc. | Form DFAN14A - Additional definitive proxy soliciting materials filed by non-management and Rule 14(a)(12) material | View Filing | 3/2/2026 |
| LULU | DFAN14A | lululemon athletica inc. | Form DFAN14A - Additional definitive proxy soliciting materials filed by non-management and Rule 14(a)(12) material | View Filing | 2/27/2026 |
| LULU | DEFA14A | lululemon athletica inc. | Form DEFA14A - Additional definitive proxy soliciting materials and Rule 14(a)(12) material | View Filing | 2/27/2026 |
| LULU | DFAN14A | lululemon athletica inc. | Form DFAN14A - Additional definitive proxy soliciting materials filed by non-management and Rule 14(a)(12) material | View Filing | 12/30/2025 |
| LULU | DEFA14A | lululemon athletica inc. | Form DEFA14A - Additional definitive proxy soliciting materials and Rule 14(a)(12) material | View Filing | 12/30/2025 |
| LULU | DFAN14A | lululemon athletica inc. | Form DFAN14A - Additional definitive proxy soliciting materials filed by non-management and Rule 14(a)(12) material | View Filing | 12/29/2025 |
| LULU | DEFA14A | lululemon athletica inc. | Form DEFA14A - Additional definitive proxy soliciting materials and Rule 14(a)(12) material | View Filing | 12/29/2025 |
The proxy materials were submitted for AI analysis to four major models, and Claude was asked to generate a "Consensus" view that compares the responses. This is pure analysis, not a recommendation for your voting by Proxyanalyst.
All four models independently concluded that the activist case is materially stronger than management's defense, though they differ on the precise scope of support warranted. The convergence across models is striking: each independently identified the same core failures — catastrophic financial underperformance (-67.9% stock decline, ~$17B in value destruction), systemic operational missteps (Mirror write-off, product recalls, CEO succession failures), and objective governance deficiencies (classified board, Advent entrenchment, long tenures) — as decisive factors favoring board change. Management's defense was uniformly characterized as thin, reactive, and unresponsive to the substance of Wilson's critique.
Three of four models (Claude, Grok, OpenAI) recommend full support for all three of Wilson's nominees and the declassification proposal. One model (Gemini) recommends a split ballot — supporting Maurer and declassification while withholding on Gentile and Hirshberg — primarily citing concern about concentrated Wilson influence. This divergence is the contest's central point of analytical disagreement.
The weight of evidence firmly supports shareholder-driven board refreshment. The status quo is indefensible by any standard fiduciary framework.
| Model | Recommendation | Confidence |
|---|---|---|
| Claude | Support Activist (all 3 nominees + declassification; withhold on Advent-affiliated incumbents) | 8/10 |
| Grok | Support Activist (all 3 nominees + declassification) | 9/10 |
| OpenAI | Support Activist (all 3 nominees + declassification) | 8/10 |
| Gemini | Split Ballot (Maurer + declassification only; withhold on Gentile and Hirshberg) | 7/10 |
All four models converge on the following findings with high consistency:
1. Financial Underperformance is Disqualifying in Severity
Every model independently flagged the -67.9% stock decline vs. S&P 500's +44.7% return, the ~110 percentage point spread, and the three-year TSR lagging peer median by 63.6 percentage points as extraordinary and not adequately explained by cyclical or sector-specific factors alone. The scale of value destruction — approximately $17 billion over five years — is treated as a decisive accountability trigger by all four analysts.
2. Operational Failures Are Systematic, Not Isolated
All models identified the same constellation of failures: the Mirror acquisition ($452.6M written off), the Breezethrough and Get Low product failures (with documented single-day stock drops), three consecutive CEO departures without a named successor, and eight straight quarters of Americas comparable sales decline. Every model characterized these as evidence of structural board oversight failure rather than management bad luck.
3. Governance Structure is Objectively Substandard
Universal agreement that the classified board (shared by only ~10% of S&P 500 companies), extended director tenures averaging 8+ years with four directors exceeding 10 years, and persistent Advent International affiliations six years post-exit constitute material governance deficiencies. All models noted that long-tenured directors receiving ≤82% support even while running unopposed is an extraordinary and ignored shareholder signal.
4. CEO Succession Failure is a First-Order Board Dysfunction
All four models treated the three consecutive CEO departures without a succession plan as among the most damning evidence against the incumbent board. This was independently identified as the single most important board responsibility and its repeated failure as systemic rather than circumstantial.
5. Management's Defense is Substantively Inadequate
Every model characterized the company's response as procedural deflection — emphasizing engagement disputes and process claims — while offering no substantive rebuttal to the financial data, operational failures, or governance critiques. The "business as usual" framing during a period of historic underperformance was uniformly described as tone-deaf.
6. Declassification Proposal Should Be Supported
All four models recommend supporting the declassification proposal without reservation, treating it as a foundational governance improvement that stands on its merits independent of the director contest outcome.
7. Marc Maurer is the Strongest Nominee
All models singled out the former On Holding Co-CEO as the most directly credible nominee, given On Holding's revenue growth trajectory during his tenure and the direct comparability to lululemon's premium athletic positioning. Notably, Maurer is also the one nominee the board actually met, lending additional procedural legitimacy.
8. Wilson's Track Record is Genuine
All models acknowledged Wilson's 368% TSR during his directorship tenure vs. S&P 500's 63%, and Amer Sports' 184.9% TSR vs. -62.3% for LULU during a comparable period, as credible evidence of operational value creation capability rather than mere founder nostalgia.
1. Full Slate vs. Split Ballot (Primary Divergence)
This is the contest's key analytical fault line:
Claude, Grok, and OpenAI recommend supporting all three nominees (Gentile, Hirshberg, Maurer), arguing that the severity of underperformance justifies comprehensive board refreshment, that all three nominees bring relevant credentials, and that splitting the ballot risks perpetuating incumbent board control while signaling only partial accountability.
Gemini recommends a split ballot — supporting only Maurer while withholding on Gentile and Hirshberg — citing concern that installing all three nominees risks giving Wilson disproportionate influence over the company's strategic direction, potentially substituting one form of concentrated control for another. Gemini weights the "founder control risk" more heavily than the other models.
Assessment of Divergence: The split-ballot rationale is intellectually coherent but arguably insufficient given the depth of failure. Three directors on a ten-person board does not constitute control, and the board's classified structure means only three seats are even contestable in this cycle. The concern about Wilson overcorrection is legitimate but does not reach the threshold that would justify leaving qualified nominees off the board during a period of documented multi-year operational and governance failure. The three-model consensus for the full slate is the more defensible institutional position.
2. Wilson's Prepaid Forward Contracts as a Risk Factor
Claude explicitly flagged Wilson's prepaid variable share forward transactions (floor price $336.40 on ~328K shares maturing July 2026, deeply underwater at current ~$153 prices) as creating complex incentive structures that could generate pressure for short-term stock action and warrant post-contest monitoring.
Grok and OpenAI did not specifically address these instruments.
Gemini touched on Wilson's general control-seeking behavior but did not specifically analyze the derivative positions.
Assessment of Divergence: Claude's identification of this issue is the most analytically complete treatment. These instruments create a genuine conflict of interest that institutional investors should flag, though they do not reverse the directional recommendation. Governance-focused institutional investors should seek clarification from Wilson nominees on strategic timeline expectations post-election.
3. Weight Given to Chip Bergh's Appointment
Claude treated the Bergh appointment as procedurally tainted but substantively potentially valuable, recommending investors use judgment on whether to withhold as an accountability signal or support as an acknowledgment of credential.
Grok characterized it as a "tactical move" insufficient to overcome the weight of negative evidence without a broader strategic context.
Gemini treated it as a "proactive step" and a genuine positive in the company's favor, weighting it more favorably.
OpenAI noted the appointment raises questions about transparency but did not directly assess its strategic merit.
Assessment of Divergence: The mid-contest, non-consultative nature of the Bergh appointment is a legitimate procedural concern. However, the substantive apparel experience he brings is real. The models' varying weights on this reflect a genuine ambiguity that is difficult to resolve without more information about Bergh's actual independence from the incumbent board faction.
4. Confidence Calibration
These differences are calibration-level rather than directional — all models express meaningful confidence in the activist case.
Support Activist
Strength: Strong
| Proposal | Consensus Vote | Model Agreement |
|---|---|---|
| Laura Gentile (Wilson Nominee, Class I) | FOR | 3/4 models (Claude, Grok, OpenAI); Gemini withholds |
| Eric Hirshberg (Wilson Nominee, Class I) | FOR | 3/4 models (Claude, Grok, OpenAI); Gemini withholds |
| Marc Maurer (Wilson Nominee, Class I) | FOR | 4/4 models — unanimous |
| Declassification Proposal | FOR | 4/4 models — unanimous |
| Long-tenured Advent-affiliated incumbents | WITHHOLD | 3/4 models explicitly; implied by 4th |
| Chip Bergh (Company nominee, Class I) | WITHHOLD or exercise judgment | Mixed; procedural concerns dominate |
The four-model consensus rests on five pillars that are individually significant and collectively overwhelming:
Accountability for proven financial devastation. Nearly $17 billion in shareholder value destroyed over five years, with underperformance of 110+ percentage points vs. the S&P 500, cannot be attributed to factors outside board control. At some point, the board owns the outcomes.
Systemic rather than episodic governance and operational failure. The convergence of Mirror, product recalls, three CEO succession failures, and eight straight quarters of Americas comp decline is a pattern, not a run of bad luck. Patterns require structural remediation.
Governance structure is objectively unjustifiable. A classified board retained by fewer than 10% of S&P 500 peers, combined with 8+ year average director tenure and persistent Advent affiliations, is not a defensible governance architecture for a company of this size experiencing this level of underperformance.
Wilson's nominees are specifically qualified, not merely credible. Maurer's On Holding experience, Hirshberg's brand-building track record, and Gentile's sports marketing expertise are directly relevant to lululemon's stated strategic needs — this is not a placeholder slate.
Management has offered no credible alternative. The company's defense is procedural rather than substantive. No strategic plan for recovery has been articulated, no accountability for documented failures has been accepted, and no credible rebuttal to the financial record has been offered.
Institutional investors who vote with the activist should establish engagement expectations for any newly elected directors regarding: (a) independence from Wilson's personal financial positions (particularly the July 2026 forward contract maturity); (b) realistic strategic timelines that avoid short-termism; and (c) collaboration with management rather than displacement of executive function. The risk of overcorrection via founder influence is real, even if it does not change the directional vote recommendation.
Confidence: 8/10
Rationale: The directional conclusion — that the activist case is materially stronger and warrants support for board change — carries near-consensus conviction across all four models. Confidence is calibrated at 8/10 rather than higher for the following reasons:
(+) Factors supporting high confidence: The financial underperformance data is objective and unambiguous; governance deficiencies are measurable and industry-comparable; operational failures are documented with specific dollar and stock price impacts; management's response is demonstrably thin on substance; all four models independently reached the same directional conclusion.
(−) Factors preventing higher confidence: Wilson's complex derivative positions create post-contest incentive uncertainty; limited visibility into board deliberations and whether Bergh represents genuine course-correction; proxy contests are dynamic and additional filings may alter the record; lululemon retains genuine brand equity that could recover under various leadership configurations; and the split-ballot question (full slate vs. Maurer only) introduces legitimate uncertainty about optimal board composition even if the directional vote is clear.
The 8/10 represents high directional confidence with appropriate humility about execution risk — the appropriate posture for a fiduciarily responsible institutional investor.