ANGI — Q2 2026 Financial Report Analysis

Angi, the home-services marketplace behind the Angi and HomeAdvisor brands, posted a $230.7 million net loss for the second quarter of 2026, almost entirely because it wrote down $235.2 million of goodwill and a trade name — accounting values left over from past acquisitions that the company now says are worth less, after its share price kept falling. Strip that out and the business itself was close to breakeven on an operating basis: revenue fell 11% to $248.0 million as homeowners asked for fewer big jobs and fewer contractors ("Pros") paid for leads, while advertising spending went up.

At a glance

• Revenue down 11% to $248.0 million. U.S. revenue fell 12%; the money Angi earns selling homeowner requests to Pros is shrinking faster than the number of requests coming in.
• $235.2 million non-cash impairment. It doesn't cost cash, but it is the company admitting its U.S. business is worth less than its books said — the trigger was the falling share price and market value.
• Average monthly active Pros down 17% to 106,000. Fewer paying contractors on the platform is the core problem behind the revenue decline, even though churn improved.

The numbers

NM = not meaningful. "Operating income excl. impairments" is our own arithmetic: the reported operating loss plus the $225.6 million goodwill and $9.6 million trade-name charges. Adjusted EBITDA is the company's own non-GAAP profit measure (it excludes stock pay, depreciation, restructuring and impairments).

How the marketplace works, and where it shrank

Angi makes most of its money by charging Pros — plumbers, roofers, cleaners — for "leads": a homeowner's request for a job, passed to one or more contractors. Two channels feed this. The proprietary channel is homeowners coming to Angi's own sites and apps; the network channel is requests sourced through partner sites.

• Proprietary revenue fell 10% to $198.1 million, even though proprietary service requests fell only 2% (4.04 million vs. 4.12 million). The 10-Q puts the gap down to "macroeconomic conditions causing a reduction in Pro spend and utilization of available Pro capacity," with homeowner demand shifting "toward lower-consideration categories" — in plain terms, more small jobs like cleaning or handyman work and fewer big-ticket projects like remodels, which Pros pay more to win.
• Network revenue fell 34% to $17.2 million, with network leads down 35%. The company attributes this to the continued shift in traffic after its "homeowner choice" change in January 2025 (letting homeowners pick their Pro rather than being matched automatically), which deliberately moved volume away from the network channel.
• Leads fell faster than requests (-13% vs. -6%): each homeowner request is being sold to fewer Pros. Revenue per lead still rose 1%, which the company credits to more Pros moving onto its subscription product.
• Large Pro and National Partnership revenue grew 20% — revenue from contractors committing at least roughly $50,000 a year and from franchise/trade-association deals. It is the one growing pocket, but it is a subset of U.S. revenue, which still fell 12% overall.
• International (European brands) was flat at $32.6 million. Over the first half it rose 3%, which the 10-Q says was driven primarily by stronger euro and pound exchange rates — so underlying international growth is weaker than that headline.

Where the money went

Angi is spending more to attract homeowners while cutting almost everything else:

• Consumer marketing rose to $102.6 million, 41% of revenue (35% a year ago). The 10-Q says the higher TV and online advertising is meant to grow proprietary-channel requests relative to the network channel. So far it has not stopped proprietary requests from slipping.
• Product development spending fell 54% to $10.9 million, and general and administrative expense fell 19% to $59.9 million, mainly from cutting about 350 jobs in a restructuring announced in January 2026. Restructuring charges total $28.5 million to date out of an expected ~$30 million.
• Depreciation doubled to $21.0 million (+105%). The 10-Q says this reflects more capitalized software spending and accelerated depreciation on software the company is retiring as it replaces its legacy technology platform — a cost that should fade once the old platform is gone.

What the headline numbers hide

• The $230.7 million loss is mostly an accounting write-down, but the write-down is a real signal. Goodwill is the premium paid in past acquisitions, carried on the balance sheet. Angi tested it because "the continued decrease in stock price and market capitalization since December 31, 2025 constituted a triggering event," and found the U.S. business's book value exceeded its estimated fair value by $225.6 million. No cash left the company, but management is conceding the U.S. business is worth less than recorded. The 10-Q also warns another impairment becomes more likely if conditions or performance worsen; $662 million of goodwill remains.
• Before impairments, operating profit nearly vanished. Add back the two impairment charges and operating income was about $1.5 million, versus $17.7 million a year ago. Adjusted EBITDA only fell 14% because it also excludes depreciation — and depreciation is precisely the line that doubled.
• Cash generation turned negative. For the first half, operating cash flow was $9.3 million against $54.0 million a year earlier, and after $30.7 million of capital spending, free cash flow — the cash left after investment — was -$21.4 million (vs. +$29.2 million). Restructuring payouts and paying down accrued compensation weighed on it.
• Paying down debt at a discount flattered the bottom line slightly. Angi bought back $73.4 million of its 2028 bonds in the quarter for $68.0 million in cash, booking a $5.6 million gain inside "other income." That gain, not operations, is why other income rose; interest income actually fell $2.2 million as the cash pile shrank.
• Net debt went up, not down. Cash fell from $303.7 million at year-end to $188.7 million, while bond debt fell from $497.7 million to $398.5 million. Debt minus cash rose from about $194 million to about $210 million, because the business itself consumed cash.
• Tax didn't cushion the loss. The $0.9 million tax benefit is tiny against a $231.6 million pretax loss because most of the goodwill charge isn't tax-deductible.
• Fewer shares, but no EPS boost to speak of. Weighted shares fell from 47.2 million to 40.5 million after 2025 buybacks; no shares were repurchased in the first half of 2026, as cash went to bond repurchases instead.

Takeaway: The impairment grabs the headline, but the real issue is underneath it — Angi lost 17% of its active Pros in a year and is spending a bigger share of revenue on advertising just to keep homeowner demand roughly steady, leaving operating profit before write-downs near zero and free cash flow negative.

What to watch next

The earnings release gives no numerical financial guidance for the rest of 2026, so there is no target to hold management to. What the filings do point to:

• New leadership. On September 22, 2026, an 8-K disclosed that CEO Jeff Kip left the company and the board; board member Michael Steib, previously CEO of TEGNA, Artsy and XO Group (The Knot), took over as CEO. Executive Chairman Joseph Levin stepped back to Chairman. A new CEO arriving weeks before the Q3 report raises the odds of a strategy reset, further cost changes or more write-downs.
• The AI product bet. In July the company began a beta test of "AI Front Desk," software that automates lead follow-up and scheduling for Pros, the first of a planned "Pro Chief Revenue Officer" suite. The case for Angi rests on tools like this making Pros spend more; whether it shows up in active-Pro counts or revenue per lead is the thing to check.
• Pro count vs. churn. Monthly churn improved to 6.0% from 6.8% and acquired Pros rose 13%, yet active Pros still fell 17%. If acquisition keeps rising and churn keeps falling, the active-Pro decline should start to narrow — Q3 is the test.
• Marketing payback. If consumer marketing stays above 40% of revenue without proprietary requests turning positive, profit will stay squeezed.

Our read: with no guidance, a fresh CEO and depreciation from the platform migration still elevated, Q3 is more likely to show further revenue decline than a turn. The balance sheet is not under immediate strain — $188.7 million of cash, $175 million available on its credit line, and no bond maturity until August 2028 — but negative free cash flow means that cushion is being used, not built.