Build-A-Bear Workshop, Inc. — BBW
IDENTITY
Name Build-A-Bear Workshop, Inc.
Ticker BBW (NYSE)
Form publicly traded C-corp
Filed in Delaware [A] "a Delaware corporation", 8-K ex-10.1, 2026-08-26
HQ 415 South 18th St, Suite 200, St. Louis, MO 63103 [A] same exhibit
Founded 1997 [A] company release 2026-08-27
Sector Consumer Cyclical / Specialty Retail
Employees — not yet looked up⚠ FISCAL CALENDAR — the first thing, and BBW is a trap
⚠ FY ENDS the SATURDAY NEAREST JANUARY 31. 52/53-week retail calendar.
fiscal 2026 = Feb 1, 2026 → Jan 30, 2027
★ NAMED FOR THE YEAR IT STARTS — the OPPOSITE of NVDA.The quarters of fiscal 2026, all [A] from the releases:
Q1 FY2026 Feb 1 – May 2, 2026 (13 wks) reported 2026-05-28
Q2 FY2026 May 3 – Aug 1, 2026 (13 wks) reported 2026-08-27
Q3 FY2026 Aug 2 – Oct 31, 2026 (13 wks) [M] derived
Q4 FY2026 Nov 1, 2026 – Jan 30, 2027 [M] derived⚠ Seeking Alpha calls this same quarter "Q2 2027." That is SA's house convention (name it for the end year), not the company's. Use the company's label to find the filing; use the months to think.
⚠ The prior-year comparable is 364 days back, not 365 — Q2 last year ended Aug 2, 2025. That 364 is the tell for a 52/53-week calendar.
Lineage — the label collision worth remembering
[A] NVDA's fiscal year also ends in late January (2026-01-25) and NVDA names it for the end year. So May–Jul 2026 is "Q2 FY2027" at NVDA and "Q2 fiscal 2026" at BBW. Same three months, labels one year apart. There is no sector convention; read the first sentence of the release every time.
WHAT THEY SELL
(Chapter I in progress — Joe fills this.)
Reported in three segments: Direct-to-Consumer · Commercial · International Franchising. Q2 FY2026 revenue mix [A]:
Net retail sales $106.5M 92.4% of total
Commercial revenue $8.1M 7.0%
International franchising $0.7M 0.6%
───────
Total revenues $115.3M 100.0%[A] 674 global locations at Aug 1, 2026 — 379 corporately-managed, 177 partner-operated, 118 franchise. ⚠ Only the 379 are company-owned stores; the other 295 are asset-light.
THE PRODUCT — how it actually works
(Not yet covered.)
THE MOAT — and its clock
(Not yet covered.) Why they win today, with the pricing-power evidence. Every moat has a clock. If I can't name the clock, I haven't found it.
WHAT KILLS IT
(Not yet covered — Chapter IV.) Munger's inversion: the ranked, dated ways this specific business stops working. Date each one, and write the honest counterweight to each.
THE NUMBERS
(Chapter II opened 2026-08-27 on the Q2 print. Revenue and EPS covered; margins, cash and the guidance cut still to teach.)
The two windows of the 2026-08-27 release
[A] Q2 FY2026 May 3 – Aug 1, 2026 revenue $115.3M −7.2%
[A] H1 FY2026 Feb 1 – Aug 1, 2026 revenue $240.6M −4.8%
[M] Q1 backed out: 240.6 − 115.3 = 125.3
[A] Q1 as separately reported 2026-05-28: $125.3M ✓ exact match⚠ The quarter is falling faster than the half (−7.2% vs −4.8%), because the half averages in a milder Q1. The quarter is the fresher signal. Both compare to the same window 52 weeks earlier.
Q2 FY2026 revenue lines, all [A]
Net retail sales $106.5M −7.1%
Consolidated e-commerce demand −15.6% ⚠ still falling
Commercial + international franchise $8.8M −9.0% ⚠ the "growth
engine" reversed
Total revenues $115.3M −7.2%The EPS trap
[F] Seeking Alpha headline: "Non-GAAP EPS of $0.70 beats by $0.02"
[A] the company: "Diluted EPS totaled $0.70, compared to $0.94"
0.94 − 0.70 = 0.24 ; 0.24 / 0.94 = 0.2553 → −25.5%★ BBW beat consensus and shrank 25.5% on the same morning. The wire's "Non-GAAP" label is also wrong — the company reported no Q2 adjustment; $0.70 is straight GAAP.
⚠ This is the mechanism that supported "results have been fantastic" all the way down. → MY POSITION, entry 2026-08-18 at $38.32.
Not yet worked in class — parked raw
[A] Q2 pre-tax income $11.6M = 10.1% of revenue (vs $15.3M = 12.3%)
−220bp: gross margin −340bp (occupancy deleverage
+ promotions), SG&A −80bp (lower incentive comp)
[A] Q2 EBITDA $15.2M = 13.2% (vs $18.8M = 15.1%)
[A] H1 GAAP EPS $2.16 · adjusted EPS $1.73 — strips a ~$7M IEEPA tariff
refund for PRIOR-year costs ← non-GAAP that cuts
AGAINST the company. Direction not yet discussed.
[A] cash $14.0M, down 64.2% from $39.1M a year ago
[A] inventory $81.1M, −0.8%
[A] capex H1 $15.4M vs $6.3M last year (2.4x)
[A] buyback H1 $17.1M / 403,236 sh · $43.2M left of $100M auth
[G] FY2026 GUIDANCE CUT revenue $500–525M (was $530–550M)
pre-tax $60–68M (was $72–78M)
adjusted pre-tax $53–61M
[A] 8-K same day: Chief Growth Officer David Henderson terminated
WITHOUT CAUSE effective 2026-08-26 — the day before
the print. Not mentioned in the release.WATCH EVERY QUARTER
- E-commerce demand — −26.1% (Q1) → −15.6% (Q2). Decelerating
decline, still a decline. Is there a floor?
- Gross margin — it broke first here (−340bp). Margin is the line
that breaks first when demand softens.
- Commercial + franchise — the asset-light engine, now −9.0% in Q2
after +11.6% for the half. It turned inside the year.
CUSTOMERS & DEMAND
(Not yet covered.)
MY POSITION
FLAT. Two completed round trips, two accounts. Never blend them (archive-is-scoped-to-ira-660-only).
Trip 1 — taxable 8526, WIN
buy 2025-10-16 12 sh @ 59.31 = 711.72
buy 2025-10-17 105 sh @ 56.22 = 5,903.10
───────────────────────────
117 sh, cost 6,614.82 avg 56.5369
sell 2026-01-05 117 sh @ 61.40 = 7,183.80
7,183.80 − 6,614.82 = +$568.98
568.98 / 6,614.82 = 0.08601 → +8.60% held 81 daysTrip 2 — IRA 660, small loss, ★ CORRECT EXIT
buy 2026-08-18 09:55 CDT 300 sh @ 38.32 = 11,496.00
sell 2026-08-18 14:29 CDT 300 sh @ 37.7067 = 11,312.01
11,312.01 − 11,496.00 = −$183.99
183.99 / 11,496.00 = 0.01600 → −1.60% held 4h 34mLedger note at the time: "Exit per the 2026-08-18 review: no dated catalyst, value-trap risk Joe himself flagged; sold before the 8/27 print by plan, not on the tape."
⚠ SCORE THE EXIT — what holding would have done
[A] 2026-08-26 close 39.10 ← it ran ABOVE his exit first
[A] 2026-08-27 09:50 CDT, post-print 30.18 = −22.82% on the day
hold to now 300 × 30.18 = 9,054.00 − 11,496.00 = −$2,442.00
what he did −$ 183.99
────────────
the exit was worth +$2,258.01★ This is the counterexample to the standing worry that he liquidates quickly under pressure. He did not flinch out on the tape — the stock went UP $1.39 from his exit over the following six sessions and he did not chase it back. He left on a dated reason (an unpriced earnings report he had no edge on) and the date proved him right. Recorded as a win of PROCESS, not of forecasting: he never predicted the print.
⚠ The honest counterweight: the same discipline would have shown a −$183.99 loss and looked foolish had the print been good. Grade the rule, not the outcome.
THESIS & EXIT
No live thesis. Four BBW theses (#2, #3, #4, #12) were closed 2026-08-23 when the thesis layer was retired wholesale in favour of /weekend and /calendar. #4 is the substantive one and its verdict still stands as of the last reading: the original "great business, unfair selloff" premise was refuted leg by leg — tariffs turned into a tailwind, the problem is demand, and the deterioration is in the P&L. What survived was a different, colder deep-value argument.
⚠ The Q2 print of 2026-08-27 is the first hard test of that surviving argument. Not yet worked in class.
SESSION LOG
2026-08-27 — course opened on the Q2 print. Q1–Q3 asked; the fiscal calendar took three passes and was genuinely frustrating (there is no convention to learn — that IS the lesson). Landed: the two-window structure of a release, and backing Q1 out of the half by subtraction. ⚠ Pace note: the session went too fast in the middle and had to reset. Smaller steps on this course.
A worksheet is my study notes on one company, not advice. I am not a financial adviser and hold no licenses; nothing here is a recommendation to buy, sell, or hold anything. Every figure is as of the day I wrote it down and none of them are maintained — a number true last month may be wrong now, and the tier tags say which are the company’s and which are mine. Any position described is what I held when I wrote it; the record is the only current answer, and the full legal terms are on the front page.