Financebotresearch desk研究台

FISV › analyze

FISV · Analyze

HOLD Fintech

Baseline: Output/Stocks/Fintech/FISV/analyze-2026-07-09.md (2026-07-09, HOLD [5.5]) · Prior-prior: 2026-06-04 [5.5] Now: $49.45 · MktCap $26.3B · P/B 0.98 · P/E (ttm) 9.6 · Fwd P/E on new guide midpoint $7.30 = 6.78x (Yahoo's 6.12 is stale, keyed to withdrawn $8.15 guide — pitfall-vendor-forward-eps-is-stale-on-the-day-of-a-guidance-cut)

1. What this updates

Ten weeks since the baseline. The single-event that the baseline's whole verdict hinged on — the July 29 Q2 print that management "explicitly called the trough" — happened, and management were wrong. Q2 came in worse than Q1 on the revenue line, margins compressed further, and full-year guidance was cut a second time.

Event list since 2026-07-09

A portfolio-specific passage was removed from the public build.

The muted -3.7% net price move is deceptive: the Aug 6 print took the stock to $47.90 in the pre-market and it clawed back. The market absorbed a full guide-cut and closed roughly flat over the period, which is itself an information event — see §3.

2. The delta ledger

16 baseline claims tested. 1 retracted, 3 superseded, 4 refreshed, 5 carried, 2 drifted, 1 untested. Judgment rows: verdict CARRIED, conviction DOWNGRADED [5.5 → 4.5] per baseline's own explicit trigger.

🔄 SUPERSEDED — the trough thesis broke

#1 · Judgment · "Q2 2026 is the expected trough for organic revenue decline; the July 29 print will validate or break the trough thesis." → Q2 organic came in -5%, worse than Q1's -4%. FY guide was cut, not affirmed. Management no longer calls Q2 the trough. Every downstream row that inherited "we're at the bottom" needs re-adjudicating without that assumption. This is the single most important status change of the pass, and it fires the baseline's own explicit downgrade rule ("Q2 organic revenue -4% or worse for a second consecutive quarter → downgrade to [4.5]").

#2 · Trend · "Full-year 2026 guidance maintained at organic +1% to +3% / adj EPS $8.00-$8.30 — new CEO isn't resetting expectations." → New guide: organic -1% to flat / adj EPS $7.20-$7.40. The new CEO is resetting expectations — just three weeks after the baseline said he wasn't. The 2026-07-09 read of "at least the new CEO isn't still resetting" was accurate as of that date and became false on 2026-08-06. Cause: Argentina headwinds worse than expected, client-implementation delays (specifically an e-commerce client M&A push-out), stepped-up technology spend for Project Elevate.

#3 · Price · "Fair value base $95–$114 (10-12x FY27E adj EPS ~$9.50 recovery); bear $52–$65 (8x FY26 guided $8.15)." → On the cut guide of $7.30 midpoint, apply the same 10-12x recovery frame: base $73–$88. Bear (8x on $7.30) = $58. Bull (14-15x on FY28E $9.00) = $126-$135. The whole ladder shifts down by ~20-25%. Current $49.45 is below even the reset bear.

❌ RETRACTED — one row, from a vendor trap

#4 · Price · Baseline showed "Forward P/E 6.3x" citing Yahoo's field. → Yahoo's fwd P/E still reads 6.12x because Yahoo's E number is the withdrawn $8.15 midpoint. On the actual $7.30 midpoint the true forward P/E is 6.78x. Baseline was correct at time of writing (the guide it inherited was still the $8.15 number) but the approach — pulling fwd P/E straight from Yahoo without cross-checking the E number — is exactly the pitfall this repo has already named. Flagged going forward: on any name that has cut guide in the trailing 90 days, compute fwd P/E from the guidance PDF, not from the vendor snapshot.

🔁 REFRESHED — figures moved, direction of the claim unchanged

#5 · State · FCF quarterly cadence (baseline flagged Q1's -55% YoY FCF collapse as the "single worst data point") → Q2 FCF $1.1B, 112% conversion — a strong recovery from Q1's $141M. FY cash-conversion target is ~90% and the quarter beat that. The Q1 FCF collapse now reads as timing (capex + working-capital lump), not structural. This is a force against the guide-cut narrative and is why the ledger does not retract more of the baseline: cash generation is intact even as reported margins deteriorate.

#6 · State · Leverage (baseline: Net Debt/EBITDA 3.1x drifting; downgrade trigger at 3.5x) → Q2 gross debt/adj EBITDA <3.2x, management targeting ~3x by year-end. Trigger did not fire. Deleveraging is on track, capital-allocation pivot from baseline (buybacks throttled 90%, modest debt paydown) is being honored. This is the strongest force against a deeper downgrade.

#7 · State · Insider buying (baseline: 6 insiders bought June 16-17 @ $48-51) → Third wave now on record: Director Fritz Lance $520k @ $51.95 on Aug 7, the day after the guide cut. That is three consecutive clusters (Dec 2025 CFO $1.06M @ $62; Jun 2026 six insiders @ $48-51; Aug 2026 Lance @ $52). Insiders continued buying through the guide cut, at higher prices than the June trough. This is a stronger signal at time-of-writing than at baseline. Weighting: still a mitigant, not a green light — the CFO and 6 directors were also wrong about Q2 relative to what they now know.

#8 · Price · Analyst consensus (baseline: 26 analysts, mean target $69.77) → Mean target $60.44, sell-side kept cutting through the print. Recommendation still "hold". The Street continues to price execution-continuity risk.

✅ CARRIED — re-tested, still true

  • #9 · Structural · Core banking moat (42% of US banks, high switching costs, 10-yr contracts). No client-loss disclosures, no material renewal misses in the Q2 8-K. Moat prior holds; a bad quarter inside an intact model.
  • #10 · Structural · ROIC on tangible capital ~97%. No re-computation warranted this pass; annual data unchanged. Reported ROIC 8.9% still masked by First Data goodwill.
  • #11 · Structural · Buyback-manufactured share count — 630M → 534M over 3yr, -15.3%. Q2 shares out 531.77M (was 534M at baseline). Buybacks throttled but pace still positive-for-EPS. See pattern-buyback-manufactured-eps-screens-as-growth — the Rev/share and FCF/share CAGRs remain flattering vs the underlying series.
  • #12 · Structural · Tangible book deeply negative (-$22.1B) from 2019 First Data goodwill. Unchanged. Not impaired.
  • #13 · Structural · Leadership continuity risk (CEO+President exits in 3 weeks pre-baseline). No further C-suite departures in the trailing 10 weeks. That specific risk did not compound.

📉 DRIFTED — moving against thesis but not yet breaking it

#14 · Structural · Clover/Merchant segment resilience — Q2 organic -5% (worse than Q1 -1% for Merchant, Q1 -6% Financial Solutions). The merchant/Clover leg was the growth story the baseline was still willing to underwrite. Breaks at: another consecutive quarter of Merchant organic negative, or an explicit Clover growth-rate print below 15% (baseline framework).

#15 · Structural · Argentina — baseline treated Argentina as a one-time October 2025 unwind that had been priced. Q2 guide-cut commentary re-cited Argentina as an active headwind. Breaks at: Argentina cited again in Q3, indicating it is not a one-off.

⏳ UNTESTED — could not corroborate this pass

#16 · Structural · STAR network sale as ~$15B call-option — baseline flagged as speculative, not underwritten. CEO says still exploring, but the September read is that Accel is more likely to be sold than STAR — the sale narrative that was 100% about STAR at baseline is now split. No terms, no filing, no confirmed counterparty. The frame has shifted (Accel over STAR) but no fact is dispositive enough to move a status. Called out: this is the second consecutive pass with STAR sitting in the ledger as speculative — if the Oct/Nov update still cannot corroborate, drop the row entirely rather than let a rumour accumulate seniority.

🆕 NEW — no baseline counterpart

  • Project Elevate — AI-driven enterprise transformation announced post-baseline. Currently a name without KPIs; the Nov print is where cost-out or margin-run-rate numbers should first appear. Structural rating: neutral until numbers land.
  • Q2 FCF conversion of 112% — was not in the baseline framework at all. Materially changes how to read the reported-margin decline: cash-generation quality is holding up better than the P&L suggests.
  • Michael Burry public "falling knife" — sentiment noise, not a force on any claim.

3. How the close calls were decided

The central tension of the pass: the trough thesis broke (baseline's whole gating question), yet FCF and leverage — the two hard financial rows — actually improved from the baseline pass. Weights:

  • The baseline's own downgrade trigger fired mechanically. "Q2 organic -4% or worse for a second consecutive quarter → [4.5]". Q2 came in -5%. The July report wrote that rule when it had the least emotional attachment to the outcome, which is exactly when trigger rules are supposed to be honored. Overriding a rule that just fired because the FCF number softens the blow would be moving the goalposts.
  • But the FCF and leverage improvement is a genuine force, not cover. It is why the ledger settles at [4.5] and not [4.0] or worse. A company generating $1.1B/quarter of cash with leverage on-plan to 3x is not the shape of a business in operational distress.
  • The insider buying continuing through the guide cut (Aug 7 Lance @ $52) is the third consecutive cluster and the second at a higher price than the last. Weighting: real, but insiders who bought at $48-52 through Q2 are now sitting on the same guide-cut information as the analysts who kept cutting targets — they have not been vindicated yet.
  • The muted price reaction to the guide cut (-3.7% net over ten weeks) cuts two ways: either the market had already priced trough-failure and moved on, or there is a pool of forced-sellers waiting for the next negative catalyst. On the reset fair-value ladder, current $49.45 is still below reset bear ($58) — the market pricing is doing the same work as our reset math, which raises confidence in the direction but not conviction in the compounder.
  • Structural rows all held. The moat did not break in Q2. What broke was the pace of getting through the transition, not the ability to. This is why the verdict stays HOLD rather than TRIM — the model is intact, the timeline is not.

Where forces balanced, the baseline stood. Moat, ROIC, buyback dynamics, negative TBV — all carried without argument. This is what the ledger is supposed to look like on a name where the operating trend rerated but the business did not.

4. Thesis persistence and conviction delta

Persistence: Of 5 Structural + 3 Trend/State ledger rows carried into this pass, 6 held (CARRIED/REFRESHED-in-direction) and 2 broke (SUPERSEDED: trough, guide-affirmation). Persistence ~75% — moderate. The structural thesis (moat, capital allocation, cash quality) held; the operational-timing thesis (trough is Q2, guide holds) did not.

Conviction 5.5 → 4.5, driven by: - Row #1 (trough SUPERSEDED) — the specific claim that gated the baseline verdict. - Row #2 (guidance SUPERSEDED) — the second consecutive downward reset within the CEO transition window. - Row #14 (Clover DRIFTED) — the growth-lever softened. - Offset by Row #5 (FCF REFRESHED positive), Row #6 (leverage on-plan), Row #7 (third insider cluster).

Net -1.0 point. The baseline's own trigger written on 2026-07-09 read "downgrade to [4.5]", and honoring an ex-ante rule is worth more than a fresh weighting today.

5. Updated verdict — full standalone

Verdict: HOLD [4.5]. Do not add; do not trim below fair value; wait for Q3 to test whether the guide-cut is the trough or the second-of-three cuts.

  • Fair value: base $73–$88 (10-12x FY27E adj EPS $7.50-$8.00 recovery, one turn below prior model); bear $58 (8x $7.30 cut guide); bull $126-$135 (14-15x FY28E $9.00 if Project Elevate delivers a real margin step). Graham on refreshed TTM EPS $5.42 · BVPS $50.67 = √(22.5 × 5.42 × 50.67) ≈ $78.6. Analyst mean $60.44.
  • Entry zone: $45-$52 — full downgrade trigger has fired, so the "wait for Q2" gate from baseline is retired; new gate is Q3 turn (§triggers). Do not initiate above $52 without a Q3 print.
  • Trim as multiple: 13x fwd (formerly implied ~14x on baseline base). On the cut guide midpoint that dollarizes to ~$95; the site rebuilds the level as EPS moves.
  • Break triggers (downgrade to [3.5]): Q3 organic worse than -3% (third consecutive negative quarter); FY guide cut a third time; Net Debt/adj EBITDA prints above 3.5x; further C-suite exit; Merchant/Clover growth prints below 15%.
  • Upgrade triggers (back toward [5.5]): Q3 organic flat or better; Project Elevate delivers a named margin-run-rate KPI; STAR or Accel divestiture with disclosed terms and clear proceeds-use; adj op margin returns to 33%+; insider selling absent for two more quarters while buying continues.

6. What this pass did NOT test — flagged for next run

  • STAR/Accel divestiture — UNTESTED for the second consecutive pass. If Q3 update still has no filed terms, drop from the ledger rather than carry a rumour with growing seniority.
  • Project Elevate cost-out KPIs — NEW row without numbers. Nov print is the first-look.
  • Q2 organic-by-segment corroboration — pulled from press release + earnings-call summary. Not cross-checked against the 10-Q line items filed 2026-08-something. Any status change here on Merchant/Clover growth deserves a filing read in Q3.
  • Argentina segment sizing — cited by management but not sized in this pass. If cited a third time in Q3 without a run-rate, escalate to a filing pull.
  • Tangible ROIC (~97%) — carried by inheritance, not recomputed. Should be re-derived at the next annual report.

A portfolio-specific passage was removed from the public build.

Sources

Baseline: analyze-2026-07-09.md · Prior-prior: analyze-2026-06-04.md