DBX
DBX 5Price and target zones价格与目标区间
Near the buy zone接近买入区间
What the company does公司做什么
No company description is on file. Run python .mcp/profiles.py to get one.
Investment thesis投资逻辑 from the watchlist来自关注列表
👀 ANALYZED 8/11. WATCH — a public LBO that works, already priced as though it does. Marginal fit for this sleeve and worth naming: there is no re-rating catalyst here, only arithmetic. Bogle: 0% growth + ~10%/yr share retirement = ~10%/yr compounding from a business that grows not at all — 363M diluted shares in FY22 to ~226M today, with $1.385B of authorization remaining = 18.6% of the market cap, re-authorized June 2026.
What's working
- The bear premise weakened: paying users +96K QoQ to 18.19M, the best net add in ~3 years and the third straight up quarter after five down
- LBO chatter live; Half Moon Capital pushing to collapse the dual class; Houston's vote is the blocker
Red flags
- Dash was demoted from product to feature. New CEO Ashraf Alkarmi: "We're embedding Dash intelligence directly into Dropbox itself." Revenue contribution has never been disclosed — no seats, no ARR, only ">150,000 users" via ChatGPT/Claude integrations. A year ago it was "standalone and embedded." A product that were winning would have a seat number attached.
- Drew Houston is no longer CEO (5/26) — executive chairman, keeps supermajority Class B, explicitly leaving to build in AI elsewhere; continuous 10b5-1 selling
- Stockholders' deficit $(2.19)B and widening by construction against $3.37B of debt. Interest is $50.0M/quarter and rising — it cut net income 24% YoY while operating margin EXPANDED. Net leverage ~2.8x vs a 5.0x covenant; $693.3M of 2028 converts to refinance. Levering a no-growth business is standard practice; levering a shrinking one is a countdown, and nobody can currently tell which this is.
- The FY25 margin jump (op income $357M→$688M) is R&D cut 22% plus two layoff rounds — headcount 2,113, smallest since 2017. Real and durable, and a one-time harvest with no second round available.
Watch
- But management credits onboarding friction removal (12 steps → 4) — self-help, non-repeatable — and ARPU is guided DOWN
- Use levered FCF, not the quoted 8.3x: unlevered guide ≥$1.070B less ~$200M interest = $3.85/sh → 8.9x, EV/FCF 11.9x.
- Graham is N/A — book value is negative, which is the finding: no asset floor exists anywhere. 🔍 Reverse-DCF: EV $10.38B on ~$870M prices FLAT FCF in perpetuity — precisely the base case, so you are paid nothing for the terminal risk. ⚖️ Sell-side median $29, below spot; 1 Buy / 7 Hold / 3 Sell; BofA double-downgraded to Underperform; JPM $25; JMP $37. Short interest ~12.9% of float, +59% since Jul 2025 — float is only ~42% of shares out, so the buyback tightens the borrow mechanically
- The portfolio already owns this exact pattern in DOCU — flat growth, high FCF, buyback-driven, bundling threat. Adding DBX doubles it. FV $27-40 (central $33). Entry $26-29. Trim 11x levered FCF.
- Sleeve rule applies.
Would upgrade on
- Upgrade: two more quarters of positive net adds ARPU stabilising a credible LBO bid.
👀 8/11 已分析。 观察（WATCH）— 一场行得通的公开市场 LBO，而市场已经按「它行得通」定好了价。 收入是平的，产品也没有在赢；这两点都不是投资逻辑。 逻辑是每年约 $870M 的杠杆后自由现金流注销掉约 10% 的股本 — 摊薄股数从 FY22 的 3.63 亿降到今天的约 2.26 亿，授权额度还剩 $1.385B，等于市值的 18.6%，并在 2026 年 6 月获得重新授权。 ✅ 看空的前提这个季度被削弱了：付费用户环比增加 96,000 至 1,819 万，是近三年最好的净增，也是连续第三个上升季度（此前连跌五个季度）。 ⚠️ 但管理层把功劳归给注册流程从 12 步简化到 4 步 — 这是自助式的、不可重复的 — 而且每用户平均收入（ARPU）指引是向下的。 🚩 Dash 从一个产品被降级成一个功能。 新任 CEO Ashraf Alkarmi：「我们正把 Dash 的智能直接嵌入 Dropbox 本身。 」它的收入贡献从未被披露 — 没有席位数，没有 ARR，只有通过 ChatGPT/Claude 集成的「超过 150,000 名用户」。 一年前的说法还是「既独立销售也嵌入」。 一个真在赢的产品，会带着席位数字出场。 🚩 Drew Houston 已经不是 CEO 了（5/26 宣布）— 转任执行董事长，保留 B 类股的绝对投票权，并且明确表示他要去别处做 AI。 10b5-1 计划下的持续减持。 🚩 股东权益为 $(21.9) 亿的赤字，并且按其运作机制在持续扩大，对上 $33.7 亿的债务。 利息支出是每季度 $5,000 万且在上升 — 它已经让净利润同比下降 24%，而同期营业利润率是扩张的。 净杠杆约 2.8x，对上 5.0x 的契约上限；还有 $6.933 亿的 2028 年可转债需要再融资。 给一家零增长的公司加杠杆是常规操作；给一家在萎缩的公司加杠杆是一个倒计时，而目前没人分得清这是哪一种。 🚩 FY25 的利润率跃升（营业利润 $3.57 亿→$6.88 亿）来自 研发费用砍掉 22% 加上两轮裁员 — 员工 2,113 人，是 2017 年以来最少。 真实且持久，但那是一次性的收割，没有第二轮可用。 ⚠️ 请用杠杆后自由现金流，而不是外界引用的 8.3x：未计利息的指引 ≥$10.70 亿，减去约 $2 亿利息 = 每股 $3.85 → 8.9x，企业价值/自由现金流 11.9x。 ❌ 格雷厄姆模型不适用 — 账面价值是负的，而这本身就是结论：完全不存在资产底。 🔍 反向 DCF：企业价值 $103.8 亿对上约 $8.70 亿，定价的是永续持平的自由现金流 — 恰好就是基准情形，所以你为终值风险拿不到任何折扣。 ⚖️ 卖方中位目标价 $29，低于现价；1 家买入 / 7 家持有 / 3 家卖出；美银直接双级下调到「跑输大盘」；摩根大通 $25；JMP $37。 空头占流通股约 12.9%，自 2025 年 7 月以来增加 59% — 注意流通股只占总股本约 42%，所以回购会机械性地收紧融券。 ✅ 私有化传闻是活的；Half Moon Capital 正在推动取消双层股权结构；Houston 的投票权是拦路石。 ⚠️ 投资组合里已经持有完全相同的模式：DOCU — 零增长、高自由现金流、靠回购驱动、面临捆绑威胁。 加进 DBX 是把它翻倍。 合理价值 $27-40（中枢 $33）。 买入区间 $26-29。 减持 11x 杠杆后自由现金流。 ⚠️ *板块规则适用。 *
Competitive advantages竞争优势 from the thesis and the reports来自投资逻辑和报告
- But be honest about the cause: management attributes it to removing onboarding friction (12 steps → 4) and retention work.
- ROIC on invested capital of $1.04B against NOPAT ~$540M is nominally 50%, which is an artefact of a book value hollowed out by buybacks rather than evidence of a moat.
- Read Dash as a retention patch, not a growth restart — a product that were winning would have a seat count attached to it.
- Bull $40 — 10–11×; user stabilisation proves durable, Dash-embedded retention lifts ARPU, leverage falls, or an LBO clears at a premium.
- Moat replies that the machine is bolted to an asset being bundled away for free by two of the largest companies on earth, and that the one product meant to fix it was just demoted from a SKU to a feature.
Related companies相关公司 🔴 AI-displacement value cluster (new 8/11)
Four names analysed the same day, all screening cheap, all cheap for the SAME reason: software and services being repriced for AI displacement.
Size the THEME once, not the tickers. Take one name — G — and treat INFY, DBX and AKAM as passed, not pending. Adding a second is doubling a single bet while feeling diversified.