LPL Financial Holdings LPLA

Reference table

1. What they actually do — and the segment data problem

LPL provides brokerage and investment-advisory infrastructure to independent financial professionals and to advisors affiliated with financial institutions. Per the company profile, this spans commission-based brokerage across annuities, mutual funds, equities, fixed income, insurance and alternatives; fee-based advisory platforms; retirement-plan support; trust and investment-management oversight; and practice-management tools.

Here is the problem: the segment data in this file is eight years old.

That is the entirety of the seg_prod block, and it is dated 2018-12-31. Revenue in FY2018 was a fraction of today's $16.99B. seg_geo is completely empty. So for a company where the revenue mix determines nearly everything about earnings quality — advisory fees are recurring and rate-insensitive, commissions are transactional, cash-sweep revenue is a pure interest-rate spread — we have no current mix disclosure at all.

This matters more here than in most names, and it is worth being explicit about why. Cash-sweep revenue is the classic source of both upside surprise and downside shock in this business model: it rises with short-term interest rates and with client cash balances, and it falls with both. In FY2018 it was $500M of roughly $5B of revenue — around 10%. If it is still 10% of a $17B revenue base, that is $1.7B of highly rate-sensitive revenue with almost no associated cost, i.e. a very large share of operating income. We cannot size it, and that is the single largest analytical gap in this dive. It is explicitly why the bear case is built around rate sensitivity.

What we can see instead is the consolidated income statement, and the growth there is not in doubt:

Read the last three rows together and the whole story is there. Revenue grew 37% in FY25 and GAAP net income fell 18.5%. Interest expense rose 47%. Operating margin has been flat-to-down for three years despite the revenue doubling. This is what a debt-funded acquisition looks like in GAAP accounts before the synergies arrive.

2. The GAAP-versus-adjusted gap — the central question

This is the analytical crux of the name, so it gets its own section with the full quarterly detail.

Three observations.

1. Q3'25 is the scar. Revenue jumped from $3,835M to $4,552M (+18.7% sequentially, the Commonwealth assets arriving) and operating income collapsed from $474.4M to $72.2M, producing a GAAP net loss of $29.5M while the adjusted line showed +$5.20 per share. A $5.57 per-share gap in a single quarter is an extraordinary adjustment, and it is the reason the market applies a discount to the adjusted series.

2. The conversion rate is improving in a straight line: 75%, negative, 72%, 79%, 81%. Excluding the Q3'25 charge quarter, GAAP has gone from 72% to 81% of adjusted in three quarters. This is the single most important trend in the dive and it supports the bull case. If it reaches 90%+ and stays there, the adjusted numbers are real and 12.0x FY27E is genuinely cheap.

3. Adjusted EPS itself has risen every single quarter — $4.51, $5.20, $5.23, $5.60, $5.84 — and beaten consensus every single time ($4.51 vs $4.23, $5.20 vs $4.49, $5.23 vs $4.92, $5.60 vs $5.48, $5.84 vs $5.39). Five for five, with an average beat of 8.3%. That consistency is worth something.

Trailing adjusted EPS is $21.87 (Q3'25 through Q2'26), putting the stock at 16.5x trailing adjusted and 33.0x trailing GAAP ($10.92 FY25 basis). Both numbers are correct; they describe different companies. The investment case is a bet on which one you will be holding in two years.

3. The balance sheet — where the risk actually lives

FY25 year-end: cash $1,037M plus short-term investments $91.5M = $1,129M; total current assets $8,814M; goodwill $2,645M; intangible assets $3,331M; total assets $18,493M; total current liabilities $3,642M; long-term debt $7,259M (no short-term debt); total liabilities $13,148M; equity $5,344M.

• Our computed net debt: $7,259M − $1,129M = $6,130M, giving an enterprise value of $34.95B. The vendor reports EV of $33.59B, implying $4,764M of net debt — a $1.37B discrepancy, most plausibly because the vendor uses a trailing cash figure that includes client-related balances. We use the balance-sheet figure and flag the alternative.
• Leverage: $6,130M of net debt against trailing EBITDA of $2,593M = 2.36x. The vendor's netDebtToEBITDATTM says 1.84x on its lower net-debt figure. The honest range is 1.8x to 2.4x — manageable for a business with recurring revenue, but not trivial for one that printed a GAAP loss nine months ago.
• Goodwill plus intangibles of $5,976M EXCEEDS total equity of $5,344M. Tangible book value is therefore negative, which the vendor confirms at -$4.48 per share. This is normal for an acquisitive service business and abnormal for a financial institution. It means there is no asset backstop under the equity — the value is entirely the earnings stream.
• Debt has grown with the deals: $3,963M (FY23) → $5,748M (FY24) → $7,259M (FY25), funded by net debt issuance of $1,019M, $1,758M and $1,776M in those three years. The acquisitions were debt-financed, essentially in full.
• Interest coverage 5.1x and a debt-service coverage ratio of 5.4x. Comfortable, but every additional deal tightens it.

4. Cash flow — why the negative number is not what it looks like

The negative operating cash flow is largely a broker-dealer artifact and should not be read as a going-concern signal. LPL's balance sheet carries large client-related receivables and payables that swing with market activity and asset levels; the working-capital line went -$1,195M, -$1,646M, -$2,610M across three years of rapid asset growth, which is what happens when a custody platform's client balances expand. The vendor's freeCashFlowYieldTTM of -3.66% and priceToFreeCashFlowRatioTTM of -27.3x are therefore economically meaningless for this business model and are excluded from the valuation.

But two things are genuinely negative and should not be waved away:

1. Buybacks have collapsed — $1,100M (FY23) to $170M (FY24) to $134M (FY25) — precisely as the acquisitions ramped. Capital that was returning to shareholders is now buying advisor practices and servicing debt. Shareholder yield is now essentially the dividend alone: $94M, a 0.33% yield, on a 9.5% payout ratio.

2. Capital expenditure is real and rising — $570M in FY25, 3.4% of revenue and roughly 88% of depreciation. This is a technology platform that requires continuous investment; it is not an asset-light royalty.

Net position: capital allocation over the last two years has been debt-funded acquisition, not shareholder return. That is a defensible strategy in a consolidating industry, and it is also why the GAAP numbers look the way they do.

5. Valuation — priced in or room?

At $360.35 (market cap $28.82B, EV $34.95B), trailing revenue is $19.61B and trailing adjusted EPS is $21.87.

\*FY26E growth measured against FY25's $16.99B.

Note the coverage asymmetry: only 2-4 analysts contribute revenue estimates while 6-11 contribute EPS. The revenue series should be treated as indicative; the EPS series is the one with real support and is what the valuation uses.

6a. What today's price assumes (the inversion)

At $360.35 — 12.0x FY27E consensus adjusted EPS and 33.0x trailing GAAP EPS — the price embeds roughly:

• Adjusted EPS compounds at ~26% from the trailing $21.87 to $36.29 by FY28, i.e. earnings grow 66% in two and a half years. (Source: consensus, 6-11 analysts.) Falsifier: any quarter where adjusted EPS fails to exceed the prior-year quarter. The streak has been unbroken for five quarters.
• The GAAP-to-adjusted conversion rate continues climbing from 81% toward the high nineties, i.e. Commonwealth integration charges genuinely stop. (Our computation from the reported quarterly data.) Falsifier: any quarter with a conversion rate below 75%, or any new large acquisition that restarts the charge cycle.
• The market keeps paying only ~12x forward. At an unchanged price the multiple rolls mechanically from 15.2x to 9.9x. The price is currently assuming a permanent discount for earnings quality — i.e. that "adjusted" is not fully believable. Falsifier: a sustained re-rating above 15x forward would mean the market has accepted the adjusted series.
• Leverage stays contained — net debt around $6.1B against rising EBITDA, keeping the ratio inside 2-2.5x while interest expense of roughly $400M annually does not rise further. (Balance sheet plus income statement.) Falsifier: total debt exceeding $8.5B, or quarterly interest expense above $120M.
• THE MOST FRAGILE ASSUMPTION: that cash-sweep revenue holds. This is the assumption we can least verify and the one most likely to break. Cash-sweep is a pure interest-rate spread on client cash balances — it was $500M of roughly $5B of revenue in FY2018, the only disclosure we have — and it carries almost no cost, so it drops nearly entirely to operating income. A rate-cutting cycle compresses it directly, and a bull market in equities reduces the client cash balances it is earned on. Neither is in any consensus number we can see. Falsifier: any quarter where revenue growth materially lags asset growth — the signature of spread compression.

6b. The return bridge (why the multiple moves)

Expected return ≈ EPS growth + multiple drift + shareholder yield.

Our base case of $420 is constructed as 14.0x FY27E adjusted EPS of $29.95 = $419. Equivalently, that is 11.6x FY28E EPS of $36.29 — which is below today's 12.0x FY27E multiple. In other words: the base case is essentially today's forward multiple, rolled forward one year, with a modest 2-turn expansion on the nearer year.

Decomposing the ~17% base-case return: adjusted EPS grows 26.4% from FY26E to FY27E; the multiple expands from 12.0x to 14.0x (+17%); dividends contribute 0.33% annually. Because EPS growth exceeds the total return, this base case is actually assuming the multiple does LESS work than the earnings do over the horizon. The 2-turn expansion is attributed to one specific mechanism: GAAP convergence. As the conversion rate rises from 81% toward the nineties, the gap between the P/E the market pays on GAAP (33x) and on adjusted (16.5x) narrows, and the market becomes willing to pay a normal-quality multiple on the adjusted series.

That expansion is the fragile leg and it is named as such. If the market never accepts the adjusted numbers — because charges keep recurring — the multiple stays at 12x and the FY27E fair value is $359, exactly today's price. In that scenario you earn the earnings growth from FY27 to FY28 and nothing else. That is not a disaster, but it is a materially worse outcome than the base case, and it is why this is Stage-In rather than Buy.

Shareholder yield is 0.33% and falling in relevance. Buybacks collapsed from $1.1B to $134M as the acquisitions ramped. Do not underwrite any part of the return to capital return.

6c. Variant perception (where we differ, what would surprise)

• We are more bullish than the price and roughly aligned with the sell side, which is itself informative. Consensus targets average $392.67 (+9.0%) in a band of $387 to $401 — a $14 spread on a $360 stock, which is not a distribution of opinion but a handful of analysts anchored to the same near-term model. Our $420 is above all of them, because we think the street is underwriting the next twelve months while the actual repricing event — GAAP convergence — takes two years.
• Our specific variant perception: the market is applying an earnings-quality discount that the Q2'26 data has already begun to invalidate. The conversion rate has gone 72% → 79% → 81% in three quarters. Nobody is talking about that trend, and it is the single cleanest observable that resolves the entire bull/bear debate. This is the edge in this dive.
• Where we are MORE cautious than consensus: rate sensitivity. No consensus number we can see accounts for cash-sweep compression, and the file does not disclose the line's size. We treat this as the primary source of negative surprise and it is the reason the bear case is set 21% below spot rather than 10%.
• Positive surprise that would force a repricing: two consecutive quarters with GAAP-to-adjusted conversion above 90%, or a resumed buyback at scale. Either would signal management believes the integration is complete. Watchable at: 2026-10-29 and the following quarter.
• Negative surprise: a new large debt-funded acquisition. This would restart the charge cycle, push leverage past 2.5x, and defer convergence by another two years — and the market would be right to keep the discount. Watchable at: any deal announcement.

Synthos fair values

• Bear ~$285 (-21%). Cash-sweep revenue compresses in a rate-cutting cycle, Commonwealth advisor attrition runs above plan, and the market refuses the adjusted series. FY27E EPS lands at the analyst low of $28.35 and the multiple compresses to 10x = $284. This is roughly the 52-week low of $265.86 plus 7%, so it is a scenario the stock has traded near within the last year.
• Base ~$420 (+17%). 14.0x FY27E adjusted EPS of $29.95, equivalent to 11.6x FY28E. Assumes convergence continues and a modest quality re-rating.
• Bull ~$545 (+51%). 15.0x FY28E adjusted EPS of $36.29 = $544. Requires convergence to complete, the acquisition cadence to stop, buybacks to resume, and the market to price LPL as the platform utility it is rather than as a levered broker.

6. Knowledge base — zero substantive claims

A systematic search on LPLA, LPL Financial and LPL returns exactly one hit, and it is not about the company.

compound_and_friends (2026-02-13, conviction 65, neutral): "AI 'tape bombs' — a karaoke penny stock's AI-freight claim, private altruist vs Schwab — wipe tens of billions off incumbents on unverified, often false headlines." The entity list includes Algorithm Holdings, JB Hunt, C.H. Robinson, Schwab, LPL and Raymond James. LPL appears only as one of six named incumbents in an observation about market structure and false headlines. There is no view on LPL Financial the business or the security.

We therefore record kb_claim_count: 0 and kb_breadth: 0. This dive is fundamentals and quant-driven only — an accepted and precedented outcome. The one hit is documented here rather than counted, in the same spirit as the name-collision discipline applied elsewhere in this batch: a mention is not a claim.

One indirect observation is worth extracting from that claim, however. It notes that unverified headlines "wipe tens of billions off incumbents." LPL is precisely such an incumbent — a low-beta, widely-held financial platform. With an RSI of 77.9 and 9 hold ratings against 15 buys, this is a name where a single negative headline into an extended technical position could produce a disproportionate move. That is a timing argument, and it reinforces the staged entry.

7. Technicals — the reason this is Stage-In and not Buy

• RSI 77.9. The most overbought reading in this batch by a wide margin (next highest is 64.3). Readings above 70 are conventionally extended; above 75 they are rare.
• Price $360.35 (+1.88% today), -8.4% from the $393.25 52-week high, +35.5% above the $265.86 52-week low. Maximum drawdown from peak is only -9.7% — this stock has barely corrected.
• +19.0% above the 50-DMA ($302.94) and +9.2% above the 200-DMA ($329.95). The 50-DMA is now above the 200-DMA — a fully repaired trend structure, and a strong one.
• MACD +13.18 — strongly positive momentum.
• Relative performance is the interesting wrinkle: 3-month +13.1% (against SPY +5.1%) but 12-month -8.9% (against SPY +19.9%). LPLA spent most of the last year going nowhere or down and has re-rated hard in the last quarter. The 19% gap between price and the 50-DMA is what a violent re-rating looks like, and gaps like that typically close at least partly.
• Beta 0.507 — by far the lowest in this batch. Average volume 991,500 shares.

The technical read is unambiguous: excellent trend, terrible entry. A 19% gap above the 50-DMA with an RSI near 78, four days after a beat, is the definition of a name to buy patiently. The 50-DMA at ~$303 would be a -16% move and would be an outstanding second-tranche entry with the fundamentals unchanged.

8. Moat & competitive position

LPL's moat is scale plus switching costs in a fixed-cost service business. Building and maintaining custody, clearing, compliance, trade processing and advisor-facing technology is expensive and regulated; doing it for tens of thousands of advisors spreads that cost across an enormous asset base. Once an advisor is on the platform, leaving requires re-papering every client relationship — disruptive to the advisor's own book and therefore rarely done casually. Client assets are sticky by construction: they belong to the end investor, and the advisor is reluctant to introduce friction.

Evidence in the data: operating margin sustained in the 13-16% range while revenue doubled; interest coverage of 5.1x maintained through a doubling of debt; and consistent earnings beats through a period of heavy integration. Businesses without pricing power do not do that.

Three real competitive constraints:

1. The peer set is formidable. Raymond James ($34.1B market capitalisation) and Stifel ($12.7B) are named peers with comparable models, and the largest custodians compete for the same advisors.

2. The moat is bought as much as built. $2.81B of acquisitions across FY24-25 is the primary growth mechanism. A roll-up's moat is only as good as its integration, and this file provides no advisor-count or retention disclosure to verify it.

3. Rate sensitivity is a hole in the moat, not a moat. Cash-sweep revenue is not a competitive advantage — it is a spread that every custodian earns and that the Federal Reserve controls.

Verdict on the moat: real, durable, and unglamorous — with a rate-sensitive revenue line bolted onto the side of it.

9. Data integrity — what we rejected and why

Vendor data corruption checking is standard for these dives. For LPLA:

• REJECTED: the sellingGeneralAndAdministrativeExpenses series. The quarterly values are $321.5M (Q2'25), $606.8M (Q3'25), $708.9M (Q4'25), $400.8M (Q1'26), $777.8M (Q2'26). A near-halving followed by a near-doubling in consecutive quarters, on stable revenue, is not an operating pattern — it is an inconsistent mapping. The annual series shows the same problem: $1,040M (FY24) to $2,476M (FY25), a 138% jump against 37% revenue growth. Excluded from all margin analysis; we use operating income and gross profit only.
• REJECTED for valuation purposes: all free-cash-flow-based ratios. freeCashFlowYieldTTM of -3.66%, priceToFreeCashFlowRatioTTM of -27.3x, priceToOperatingCashFlowRatioTTM of -76.8x. These are arithmetically correct but economically meaningless for a broker-dealer whose operating cash flow reflects client-balance swings rather than earnings. The FY25 working-capital line was -$2,610M against $863M of net income — that is client money moving, not value destruction. Documented in section 4.
• FLAGGED: net-debt discrepancy. The vendor's EV of $33.59B implies $4,764M of net debt; the FY25 balance sheet shows $6,130M ($7,259M debt less $1,129M cash and short-term investments). The $1.37B gap most plausibly reflects a trailing cash figure that includes client-related balances. We use $6,130M and an EV of $34.95B, and disclose that leverage is therefore 2.36x rather than the vendor's 1.84x.
• MAJOR GAP: the segment table is from FY2018. seg_prod contains four line items dated 2018-12-31 totalling under $2.0B, against FY25 revenue of $16.99B. seg_geo is entirely empty. We have no current revenue mix — no advisory-versus-commission split, and critically no cash-sweep revenue disclosure. This is the largest analytical gap in the dive and directly drives the bear case construction.
• GAP: no operating metrics. No advisor count, no assets under administration, no net new assets, no advisor retention rate. For a platform business these are the four numbers that determine everything, and none are present. We cannot separate organic growth from acquired growth.
• FLAGGED: consensus EPS is on an ADJUSTED basis and the gap is large. FY25 estimate EPS of $19.81 against reported GAAP diluted EPS of $10.92 — the consensus series runs at roughly 1.8x GAAP. Every forward P/E in this dive is a non-GAAP multiple and is labelled as such. The trailing GAAP multiple of 33.0x is presented alongside it throughout, deliberately.
• FLAGGED: the price-target band is implausibly narrow. High $401, low $387, consensus $392.67, median $390 — a $14 spread. Against 25 rating contributors (15 buy, 9 hold, 1 sell), a $14 target range implies very few analysts actually publish targets. Treat the $392.67 as weakly-supported.
• NOTE: dividendYieldTTM of 0.33% and dividendPerShareTTM of $1.20 reconcile with the $94.4M of FY25 dividends on ~80M shares. Clean.
• NOTE: insider activity is immaterial. The largest transaction in the file is 308 shares sold at $306 (2026-06-17); the rest are in-kind tax withholdings of 44-232 shares and director awards of 3-19 shares. No open-market purchases and no meaningful sales. No signal.

10. Verdict, kill-criteria & flip conditions

Stage-In. LPL runs the custody-and-platform infrastructure for independent financial advisors — a fixed-cost business with genuine switching costs — and it has compounded revenue at 25.4% annually since FY22 to $16.99B, beaten the consensus earnings line in five consecutive quarters with adjusted EPS rising in a straight line from $4.51 to $5.84, and now trades at 12.0x FY27 consensus adjusted earnings and 9.9x FY28 with a 0.507 beta. That combination — mid-twenties earnings growth, low volatility, low multiple — is the cheapest genuine grower in this batch.

The discount is legible and it is closing. FY25 GAAP diluted EPS of $10.92 against ~$19.81 adjusted, a GAAP loss in Q3'25, $7.26B of debt against $5.34B of equity, negative tangible book, and -$993M of FY25 free cash flow are all real. But the GAAP-to-adjusted conversion rate has gone 72% → 79% → 81% across the last three quarters, which is the mechanism by which the discount closes. Base fair value $420 (+17%), built as 14.0x FY27E — barely above today's forward multiple.

The reason it is Stage-In and not Buy is entirely the entry point. RSI 77.9, the stock 19% above its 50-DMA and 4% from its 52-week high, four sessions after a beat, with a consensus target only 9% higher. Nothing about the business argues for waiting; everything about the chart does.

Staged entry (build a 2-4% core position in 3 tranches):

1. Tranche 1 — now (~1/3 of target), at ~$360. 12.0x FY27E on a 0.507-beta compounder with five straight beats. You are paid to own this even from an extended level.

2. Tranche 2 — on a pullback toward the 50-DMA (~$303), which would be -16% with the fundamentals unchanged and would take the FY27E multiple to 10.1x. This is the tranche to be patient for. Alternatively, add after the 2026-10-29 print if the GAAP conversion rate exceeds 85%.

3. Tranche 3 — on confirmation rather than price: two consecutive quarters with conversion above 90%, or a resumed buyback at scale. This tranche is gated on the thesis being proven, not on the stock being cheaper.

Pre-registered KILL / avoid-adding criteria:

• A new large debt-funded acquisition that restarts the charge cycle and pushes net leverage above 2.5x. This is the single most likely way the thesis breaks, because it is management's own recent playbook.
• GAAP-to-adjusted conversion falling back below 75% in any quarter — evidence the charges are structural rather than transitional.
• Adjusted EPS failing to beat consensus, breaking a five-quarter streak, or failing to exceed the prior-year quarter.
• Revenue growth materially lagging asset growth — the signature of cash-sweep spread compression, the risk we cannot size.
• Interest expense above $120M in a quarter (currently ~$100-106M), indicating rising cost or rising debt.

Pre-registered FLIP TO HIGH-CONVICTION CORE (upsize toward 4%):

• Two consecutive quarters with GAAP conversion above 90% — the thesis proven, and the trigger to pay a full multiple.
• Buybacks resuming above $500M annually, signalling the integration is funded and complete.
• Any current segment disclosure, particularly sizing cash-sweep revenue — this would remove the largest analytical gap in the name and could move the bear case up materially.
• The multiple re-rating above 15x forward, confirming the market has accepted the adjusted series.

Where LPLA fits in the Synthos Framework Portfolio. The financial-platform / low-beta compounder sleeve, at a 2-4% core weight. Its distinguishing characteristic in a portfolio context is the 0.507 beta — this is a genuine volatility dampener that still grows earnings in the mid-twenties, which is a rare combination and makes it complementary to the high-beta names in this batch (RDDT at 1.938, SYM at 1.933, GFS at 1.796). Sized as a portfolio stabiliser that happens to grow, not as a financials-sector bet. Logged as a tracked Synthos call (Stage-In) as of 2026-08-04 at $360.35.

Single biggest risk: a 45% GAAP-to-adjusted earnings gap on a balance sheet with $7.26B of debt and negative tangible book value. If the Commonwealth-related charges prove structural rather than transitional — because management keeps acquiring, which is exactly what it has done for two years — then the "$29.95 of FY27 earnings" the valuation rests on does not exist, GAAP earnings of roughly $17-20 are the real number, and the stock is trading at 18-21x rather than 12x with 2.4x leverage attached. The most fragile assumption in today's price is that cash-sweep revenue holds through the rate cycle — an unsized, near-zero-cost, pure-spread line that drops almost entirely to operating income and that this file gives us absolutely no visibility into.

Provenance & disclosures

• Traceability: ZERO substantive knowledge-base claims. A systematic search on LPLA, LPL Financial and LPL returned one hit — compound_and_friends (2026-02-13, conviction 65) on AI-driven "tape bombs" — in which LPL appears only as one of six named incumbents with no view expressed on the company. Recorded as a mention, not counted as a claim. kb_claim_count: 0, kb_breadth: 0. This dive is fundamentals and quant-driven only — an accepted, precedented outcome.
• Data as-of: fundamentals 2026-06-30 (Q2'26, filed 2026-08-03) · estimates 2026-08-04 · prices 2026-08-04 ($360.35, +1.88%; 50-DMA $302.94, 200-DMA $329.95, RSI 77.9, beta 0.507) · KB search 2026-08-04. Sole data source: scripts/deepdive/vti_data/LPLA_data.json.
• Rejected data: the SG&A expense series (values swinging from $321.5M to $777.8M across adjacent quarters on stable revenue, and $1,040M to $2,476M annually against 37% revenue growth — inconsistent mapping); all free-cash-flow-based valuation ratios (economically meaningless for a broker-dealer whose operating cash flow reflects a -$2,610M client-balance working-capital swing).
• Corrected data: net debt restated from the vendor's implied $4,764M to the balance-sheet figure of $6,130M, raising enterprise value to $34.95B and leverage to 2.36x from the vendor's 1.84x. Both figures disclosed.
• Disclosed gaps: the seg_prod revenue mix is dated 2018-12-31 and covers under $2.0B against FY25 revenue of $16.99B — no current advisory/commission/cash-sweep split; seg_geo entirely empty; no advisor count, assets under administration, net new assets or retention disclosure, so organic and acquired growth cannot be separated; revenue estimates carry only 2-4 analysts; the price-target band is $387-$401, a $14 spread implying very thin target coverage.
• Non-GAAP note: every forward P/E in this dive uses the consensus ADJUSTED EPS series, which runs at roughly 1.8x GAAP (FY25: $19.81 estimated versus $10.92 reported GAAP diluted). The trailing GAAP multiple of 33.0x is shown alongside the trailing adjusted multiple of 16.5x throughout, deliberately, because the gap between them is the investment question.
• Fair-value caveat: the $285 / $420 / $545 anchors are scenario multiples applied to consensus adjusted EPS, labelled at each step. The base case assumes a 2-turn multiple expansion (12.0x to 14.0x FY27E) driven specifically by GAAP-to-adjusted convergence, and that expansion is named as the fragile leg. This is scenario arithmetic, not a discounted-cash-flow model.
• Not investment advice. Independent research, educational and informational only, never personalized. Nothing here is a recommendation to buy or sell any security.
• Version: 2026-08-04-full.