TREE — LendingTree, Inc. · 2026-09-10 · Verdict: WATCH · Conviction 3
Price $26.56 (screen row, 2026-09-10 run; no live price) · Mkt cap $373M · EV $652M · EV/normalised after-tax profit 8.0x · LTM FCF yield 19.8% on the cap · Net debt $279.5M · ADV $12.8M Sources read: 10-K 2026-03-09 (Items 1, 1A, 7), 10-Q 2026-07-31, DEF 14A 2026-04-27, 8-Ks 2026-04-30, 2026-05-18, 2026-06-18, 2026-07-29, Form 4s (12m), Q2 2026 earnings call transcript.
Desk stats - Revenue trend: FY2025 $1,117.3M, up 24%, with Insurance supplying $163.2M of the $217.1M increase on 22% volume and 7% revenue per consumer (10-K, Item 7). Q2 2026 $313.4M, up 25%: Insurance +42% (22% volume, 16% price), Home +9%, Consumer -4% (10-Q, Item 2). - Normalised after-tax operating profit: LTM to 30 June 2026 GAAP operating income $103.9M (FY25 $64.8M less H1 25 $13.8M plus H1 26 $52.9M). Add back litigation settlements and contingencies of $1.2M LTM (Mantha matter, $15.2M of which fell in Q1 2025) and restructuring and severance of $3.3M LTM; no impairments, no discontinued operations. Normalised operating income $108.4M, taxed at 25%, gives $81.3M, against FY2025 GAAP net income of $151.3M. Stock compensation of roughly $20M a year stays as a cost. - EV / normalised after-tax profit: EV $652.3M (cap $372.7M on 14,034,240 shares, plus $386.4M long-term debt and $3.9M current portion, less $110.8M cash, all from the 30 June 2026 balance sheet) divided by $81.3M = 8.0x; 8.5x including $42.0M of non-current lease liabilities. - Leverage: net debt $279.5M / LTM adjusted EBITDA $153.6M = 1.82x, matching the CEO's "leverage down to 1.9x" (Q2 call); about 2.1x charging the excluded stock compensation. The $397.0M term loan matures 21 August 2030, with a $75.0M undrawn revolver (10-Q). - Is the growth sustainable? Fully organic and cash-backed, but bought: Q2 revenue rose 25% while segment profit rose 4% and variable marketing margin fell from 33% to 28% of revenue, and the full-year adjusted EBITDA guide was cut on 29 July from $152-162M to $145-152M. - What the screen got wrong: everything that made TREE screen. The 2.5x P/E, the negative tangible book and the 163.5% book-per-share growth all trace to one line, the $149.5M non-cash tax benefit releasing the valuation allowance set in 2022 (10-K, Item 7). FY2025 pre-tax income was $21.0M, not $151M. The screen also put a lead-generation advertiser in the financial lane on SIC 6163 and scored it on P/E and price to tangible book; TREE holds no loans and carries no credit risk, so those metrics say nothing. Minor: its EV omitted the $3.9M current portion of debt.
1. What the business actually does
LendingTree runs comparison-shopping marketplaces. It buys consumer traffic, mostly on search, collects request forms, and sells each request to up to five of about 770 Network Partners for a match fee (10-K, Item 1). It lends nothing and underwrites nothing. Three segments: Insurance, which is QuoteWizard and ValuePenguin and was 64% of FY2025 revenue and 67% of Q2 2026 revenue; Consumer, which is credit cards, personal loans and small business loans; and Home, now mostly home equity because the mortgage market has sat at roughly four million existing home sales for three years (10-Q, Item 2). One customer, Progressive Casualty Insurance, was 27% of total consolidated revenue in FY2025 (10-K, Item 1).
2. Why it is mispriced — the edge case
There is a real, dated, identifiable seller, and it is a reason for the price to be low rather than evidence it is wrong. Founder, chairman and CEO Douglas Lebda died on 12 October 2025 (DEF 14A 2026-04-27). Every Lebda-related holder filed a Schedule 13G/A on the same day, 4 March 2026: Lebda Family Holdings LLC at 1,325,000 shares (9.5%), the Estate of Douglas R. Lebda at 798,705 (5.7%), and Megan Greuling, who departed the company in February 2026, at 922,410 (6.6%). Against 13,953,018 shares outstanding, that is more than a fifth of the company sitting in estate settlement on $12.8M of daily volume, held by parties genuinely indifferent to price.
But an overhang pays you only if you can see it ending, and here you cannot: no filing shows a single share sold, no block trade, no 10b5-1. The counter-evidence is unambiguous. Across 39 Form 4 transaction rows in twelve months there was not one open-market purchase by any insider, at any price, through a 65% decline. Directors and officers own 4.0% as a group and the CEO owns 1.1% including 22,228 option shares. Nobody who knows this business is buying it here.
3. Unit economics and growth
There is no impairment to reverse; the distortion is entirely in the tax line, handled above. What matters is that the growth is being bought at a falling price. Q2 2026: revenue up 25%, total segment profit up 4%, variable marketing margin up 4% to $87.3M, adjusted EBITDA up 11% to $35.2M. Margin fell in all three segments year over year: Insurance 27% to 24%, Consumer 51% to 46%, Home 32% to 26% (Q2 press release, segment table). Management is explicit about why. On Insurance: "competition for policies is intense. So that helps us in partner demand, but it does pressure immediate costs" (CFO, Q2 call). On traffic: "it is fair to say that marketplaces like Google are more expensive today than a year ago" (CEO, Q2 call). Cost of revenue is only 4% of sales, so the entire cost structure is a media auction the company does not control.
Consumer is the swing. Small business, the 40%-a-year growth engine, broke in Q2 on what management calls temporary merchant sentiment; the CFO now says Consumer "might be flat to down" for the year, and the guide assumes only the lender-side recovery already visible. Home is no recovery either: revenue grew 9% while segment profit fell 14%, as direct-to-consumer lenders bid up media against a fixed pool of borrowers. LTM cash conversion is good, $86.1M of operating cash flow less $12.2M of capex for $73.9M of free cash flow, and the H1 accounts payable build of $40.6M was roughly offset by a $38.7M receivable build, so the number is not flattered.
4. Balance sheet and capital allocation
Clean and improving. The 0.50% convertibles were repaid in cash on 15 July 2025 and the capital structure was refinanced on 21 August 2025 into a single $400.0M term loan due 2030 (10-K, Item 7). Net debt of $279.5M is 43% of enterprise value, so the equity is levered about 1.75x to the operating result. The $114.7M deferred tax asset is real: the CFO put total attributes at "around $300 million of offset capacity" against future federal tax (Q2 call), so cash taxes will run well below the 25% charged above. There is no share repurchase program anywhere in the 10-K or 10-Q; what exists is a $50.0M ATM equity program, never used, while share count rose 3.1% year over year on equity awards. The CEO says leverage now gives "capacity for debt paydown, buybacks and accretive M&A," but nothing has been authorised, and pay is measured on adjusted EBITDA, the metric that excludes the stock compensation doing the diluting.
5. Management: what they said vs what they did
On 30 April 2026 management raised full-year adjusted EBITDA guidance to $152-162M and guided Q2 to $38-40M on $93-97M of VMM. Thirteen weeks later Q2 delivered $35.2M and $87.3M, below the low end of both, and the year was cut to $145-152M. The CEO's framing was "our eighth straight quarter of double-digit year-over-year adjusted EBITDA growth" and "if we would have hit that original budget for the entire year, we would be performing at the high end of the previous guidance." Both are true and neither is the point: the budget was missed. Separately, the company filed an arbitration demand against Google on 17 July 2026, having directed "about $2.8 billion to Google through the impacted period dating back about a decade" (CFO, Q2 call). That is unquantified optionality and belongs in no base case.
6. Valuation
Base (50%): FY2026 adjusted EBITDA lands at the $148.5M midpoint, 2027 is flat as Insurance plateaus and small business partly recovers, normalised after-tax profit about $82M, 8x EV against roughly $200M of net debt by end-2027 gives about $32. Bear (30%): the auto shopping cycle turns, Insurance revenue falls 20% at a 24% margin, small business does not recover, adjusted EBITDA returns to about $105M, and 7x on roughly $48M of normalised profit against $260M of net debt leaves about $9. Bull (20%): small business regains the Q1 record, Insurance keeps compounding, 2027 adjusted EBITDA $180M, normalised profit $100M at 10x less $200M net debt gives about $56, with any Google recovery on top. Probability-weighted $29.9, about 13% above price, with a bear case 66% below it. Reverse DCF: at $26.56 the $652M enterprise value against $81.3M of normalised after-tax profit implies, at a 10% discount rate, a permanent 2.5% annual decline in owner earnings. That is not absurd for a business whose traffic comes from a search box AI is rebuilding, and whose own Item 1A names "use of AI in search rather than search engines" as a risk.
7. Catalysts and timeline
Q3 2026 print in late October: whether small business revenue grows sequentially as promised and whether Insurance VMM holds near 24%. Any capital allocation announcement following the CEO's flagged flexibility. A damages figure in the Google arbitration. Any 13D/G amendment or block trade showing the estate clearing.
8. Risks and pre-registered kill criteria
- Insurance segment profit declines year over year in any quarter. It was 56% of total segment profit in Q2 and grew 25%; a decline means the carrier shopping cycle has turned, and Progressive alone is 27% of revenue.
- Consolidated variable marketing margin below 28% of revenue for two consecutive quarters. Q2 printed exactly 28% against 33% a year earlier.
- Full-year 2026 adjusted EBITDA guided below $145M at the Q3 print, a second cut in two quarters.
- Net leverage above 2.5x, or use of the $50.0M ATM or a revolver draw for anything other than an acquisition. Thesis-confirming instead: a first open-market insider purchase, or an authorised buyback funded from the roughly $80M of annual free cash flow.
9. Verdict and one-paragraph summary
WATCH, conviction 3. LendingTree screens at 2.5x earnings for one reason, a $149.5M non-cash release of a deferred tax valuation allowance that turned $21.0M of 2025 pre-tax income into $151.3M of net income, and correcting it still leaves a business that is not expensive: 8.0x normalised after-tax profit, about $80M of annual free cash flow against a $373M market value, leverage down to 1.8x and a $300M shield against future cash taxes. There is even a real, dated seller, since founder and CEO Douglas Lebda died in October 2025 and Lebda-related holders control more than a fifth of a 14 million share company. What stops it being an idea is that the estate is a reason for a lower price rather than a wrong one, no filing shows it actually selling, and not one insider bought a single share in the open market through a 65% decline across 39 Form 4 rows. Underneath, the 25% revenue growth is purchased: Q2 segment profit rose 4%, variable marketing margin fell from 33% to 28% of revenue, margin compressed in all three segments, the small business engine that grew 40% a year is now guided flat to down, and management missed its own ninety-day-old guidance and cut the year on 29 July, all while two thirds of revenue rides an auto insurance advertising cycle whose leading indicator, on the CEO's own account, is carriers eventually giving rate back. The Q3 print is the dated test worth waiting for.
Research for discussion, not investment advice. Positions and sizing are the reader's decision.