EverQuote, Inc. (EVER) - Deep Dive Research Report

Communication Services | Nasdaq: EVER | Report date: July 3, 2026

Most recent reported period: Q1 2026 (quarter ended March 31, 2026), reported May 4, 2026. Q2 2026 (quarter ended June 30, 2026) is due in early August 2026 and has not yet been released.

1. What the Company Does

EverQuote runs an online marketplace for insurance. A consumer who wants to shop for car, home, or renters insurance lands on EverQuote's site (or one of its partner sites), answers a series of questions about themselves, their vehicle, and their coverage needs, and gets matched to insurance companies and agents who want that person as a customer. EverQuote does not sell insurance and it does not carry any underwriting risk. It sells the consumer's attention and intent to the people who do sell insurance - carriers like Progressive and their local agents. The consumer pays nothing. The carrier or agent pays EverQuote.

That is the whole business in one sentence: EverQuote is a customer-acquisition engine for the property-and-casualty (P&C) insurance industry, paid on performance.

The mechanics matter, because they explain both the economics and the moat. EverQuote spends money on the front end buying traffic - Google and Facebook ads, search keywords, its own content sites, email lists, and increasingly placements on AI chat platforms - to attract people who are actively shopping for insurance. It calls this "high-intent" traffic: someone typing "cheapest car insurance in Texas" is worth far more than someone idly browsing. When that person arrives and fills out EverQuote's form, the company's machine-learning models score the lead in real time: How likely is this person to actually buy a policy? What is that policy worth? Which carriers and agents in EverQuote's network want exactly this kind of customer right now, and how much will they pay? EverQuote then routes the consumer - as a click, a phone call, or a data lead - to the highest-value match. The company earns revenue on a cost-per-click, cost-per-call, or cost-per-lead basis, meaning it gets paid at the moment of the referral, regardless of whether the consumer ultimately buys.

The genius and the fragility of the model both live in the same place: the spread between what EverQuote pays for traffic and what carriers pay for referrals. EverQuote calls the gross version of this spread "Variable Marketing Margin" (VMM) - revenue minus the cost of acquiring the traffic. Widen that spread with better matching, and profits compound. Have carriers cut their bids in a downturn, and the spread collapses.

That fragility was not theoretical. In 2022-2023 the US auto insurance industry fell into its worst underwriting-loss cycle in decades: inflation drove up the cost of repairing and replacing cars faster than carriers could raise premiums, so carriers were losing money on every new policy. Their rational response was to stop advertising and stop buying leads - if every new customer loses you money, the last thing you do is pay to acquire more of them. EverQuote's revenue cratered and it posted a $51 million loss in 2023. Then the cycle turned. Carriers repriced, returned to profitability, and flooded back into customer acquisition to rebuild market share. EverQuote's revenue grew 74% in 2024 and 38% in 2025. The company is, in effect, a high-beta play on the P&C carrier advertising cycle, with a structural growth story layered on top.

The founding story explains the DNA. The company was started in Cambridge, Massachusetts around 2011 (with roots in an earlier venture, AdHarmonics) by Seth Birnbaum and Tomas Revesz, MIT-connected engineers, backed by serial entrepreneur David Blundin and his Link Ventures. From day one it was a data-and-algorithms company that happened to operate in insurance, not an insurance company that bolted on some technology. It IPO'd on Nasdaq in June 2018. Founder-CEO Seth Birnbaum died in 2020, and president Jayme Mendal, who remains CEO today, took over. Blundin remains chairman, and his Link Ventures entities are the company's anchor shareholders through a dual-class structure (Class A trades publicly; Class B carries super-voting rights).

CEO Jayme Mendal's framing of the strategy, repeated across recent calls, captures the ambition cleanly: to be "the number one growth partner to P&C insurance providers by efficiently delivering better performing referrals, bigger traffic scale, and a broader suite of products and services." (Q4 2024 concall, Feb 24, 2025)

2. Business Segments

EverQuote reports as a single operating segment (an online marketplace), but it discloses revenue by insurance vertical, and management runs the business as two economically distinct engines plus an emerging third. Treating them as segments is the honest way to understand it.

Automotive (Auto Insurance) - ~91% of revenue

This is the core and the cash engine. In FY2025, auto insurance generated roughly $629.8 million of the company's $692.5 million total revenue, growing 41% year over year. In Q1 2026 auto revenue was $172.4 million.

What it does: connects consumers shopping for car insurance to auto carriers (Progressive is the archetypal large-scale, performance-marketing carrier) and to the thousands of local independent agents who sell auto policies. Auto is the ideal lead-gen vertical because it is a legally mandatory, price-shopped, annually-renewing product bought by nearly every adult - a deep, recurring pool of high-intent shoppers.

Core capability: EverQuote's edge is the accumulated data from, in management's words, "hundreds of millions of historical insurance shopping events." Every consumer who has ever gone through the funnel teaches the models which lead attributes predict a purchase, what a policy is worth, and which carrier will pay the most for which profile. That data flywheel is genuinely hard to replicate - a new entrant starts with zero shopping history and therefore worse matching, worse monetization, and a worse bid to offer carriers.

Competitive position and priority: auto is where EverQuote competes head-on with MediaAlpha and QuinStreet. It wins on the depth of its data and its direct, owned-and-operated consumer traffic (versus MediaAlpha's more open, transparent exchange model). It is not the growth-rate leader within the group; it is the scale and profit leader. Management's priority for auto is to raise monetization per shopper (better matching, more carriers bidding) rather than simply pushing more raw volume.

Home and Renters Insurance - ~9% of revenue

The declared growth vector. Home/renters revenue was $62.7 million in FY2025 (up 20%) and $18.5 million in Q1 2026 (up 33%).

What it does: the same marketplace mechanics applied to homeowners and renters policies. Structurally harder than auto: home insurance is more geographically concentrated (catastrophe-exposed states like Florida and California can see carriers withdraw entirely), the shopping frequency is lower, and carrier appetite is more volatile. But it is a large, under-penetrated adjacency that reuses EverQuote's entire technology stack, sales team, and carrier relationships.

Why it exists separately and its priority: it is the same machine pointed at a second product, giving EverQuote a second growth lever that does not depend on the auto cycle. Management has explicitly said it expects home to "outpace auto growth in the medium term" (Q4 2025 concall, Feb 23, 2026) as it expands the roughly 90/10 auto-to-home mix. This is the strategic option - low base, higher growth rate, and a hedge against auto cyclicality.

Health, Life, and Other - de minimis today

EverQuote historically also operated in health and life insurance and once ran a Medicare-focused agency business, but it exited or de-emphasized those lower-quality, lower-margin verticals to focus on P&C (auto + home). Today they are immaterial to the model. The strategic posture is P&C-first.

3. Products and Business Detail

EverQuote's "product catalogue" is really two things: the consumer-facing marketplace, and the increasingly sophisticated set of tools it sells to the buy side (carriers and agents). The second is where the story is evolving.

Consumer marketplace (the front door). EverQuote.com and a network of partner and content properties collect insurance shoppers. The consumer experience is a guided quote-request funnel. The output for the consumer is a set of matched carriers/agents to contact or a set of quotes to click. This is the traffic-acquisition layer, and it is being actively re-engineered for the AI-search era: EverQuote is investing in three new consumer-traffic channels - paid placement on AI/LLM platforms, AI-assisted content generation, and technical integrations so that AI assistants route insurance-shopping queries through EverQuote's marketplace (Q4 2025 and Q1 2026 concalls). Management expects LLM-sourced traffic to "become a channel of substance in 2026."

Referral products (what carriers actually buy). EverQuote sells clicks (consumer clicks through to a carrier), calls (a warm phone transfer to an agent's call center), and data leads (the consumer's information delivered as a lead). Pricing is performance-based - cost-per-click, cost-per-call, cost-per-lead - so the carrier only pays for measurable outcomes and can compute its own return on ad spend.

Smart Campaigns - the AI bidding platform. This is the most important product development of the last two years. Smart Campaigns is an automated, machine-learning bidding and budget-management tool that lets carriers and agents set an objective (e.g., a target cost per acquisition) and let EverQuote's algorithms optimize their bidding across the marketplace automatically, rather than managing keyword-level bids by hand. EverQuote launched Smart Campaigns 3.0 in Q3 2025, and reported that a customer migrating from 2.0 to 3.0 saw a 7% improvement in ad-spend efficiency (Q3 2025 concall, Nov 3, 2025). By the Q4 2025 call, "most carriers" had adopted Smart Campaigns. The strategic point: the more of a carrier's spend that runs through EverQuote's automated tools, the stickier the relationship and the more of the value chain EverQuote captures.

Agent tools and multi-product adoption. Historically EverQuote sold mostly leads to its ~5,000 local-agent customers. It is now extending its AI-powered products - smart campaigns, advanced bidding, return-on-ad-spend optimization - to agents for the first time (Q1 2026 concall). The metric management watches is products-per-agent: average adoption rose to about 1.4 products per agent from roughly 1.0 eighteen months prior, with over 40% of agents now using multiple products (Q4 2025 concall). More products per customer means more wallet share and higher switching costs.

The "manufacturing" process is software and data. There is no factory. The constraints are (1) the cost and quality of the traffic it can buy, (2) the accuracy of its matching/scoring models, and (3) the number and bid-aggressiveness of carriers on the platform. A structural feature of the last three years is extreme operating leverage: management states it has "more than doubled revenues since 2023 while keeping operating expenses essentially flat" (Q4 2025 concall), driven by pushing AI/automation through its own operations. Revenue per employee has risen nearly 3x over three years, and the engineering org has moved to an "agentic-first" software-development lifecycle (Q1 2026 concall).

Geography: EverQuote is essentially a US-only business. There are roughly 70 carriers and about 5,000 agencies on the platform. It is not an export story; its addressable market is the US P&C distribution system.

4. Customers

EverQuote's paying customers are the buy side of insurance: large P&C carriers and their local independent agents. There are two very different buyer types, and the mix matters.

Enterprise carriers. These are the large national auto and home insurers - Progressive is the model buyer, and management refers repeatedly to the "top 25 carriers." The decision-maker inside a carrier is the performance-marketing or customer-acquisition team, which runs EverQuote as one channel in a portfolio of acquisition sources and holds it to a hard return-on-ad-spend target. The buying criterion is almost purely economic: does a dollar spent on EverQuote leads produce more than a dollar of profitable new premium? Carriers can turn spend up or down almost instantly, which is why EverQuote's revenue tracks the carrier underwriting cycle so tightly. Concentration here is real but two-sided: a handful of top carriers drive a large share of spend (management noted a single top-five carrier "more than doubled its expected spend" late in Q1 2026), which is a risk if one pulls back, but also a sign of quality when they lean in. Crucially, management's recurring bull point is that "80% of top 25 carriers are still below peak spend" (Q3 2025 and Q1 2026 concalls) - the biggest customers have not yet returned to their prior appetite, implying embedded upside.

Local independent agents (~5,000 agencies). These are small businesses buying leads and, increasingly, campaign tools to fill their own sales pipelines. The decision-maker is the agency owner. The criterion is close-rate and cost-per-acquired-policy. Agents are stickier and more numerous than carriers, and the shift to selling them multiple products (leads + Smart Campaigns + bidding tools) raises switching costs - an agent who runs their whole acquisition workflow through EverQuote's dashboard does not casually leave.

Switching costs and revenue predictability. There are no multi-year lock-in contracts here; this is spend-when-it-pays performance marketing, so in principle a carrier can leave overnight. The switching cost is not contractual but economic and operational: EverQuote's data-driven matching delivers a better return than a rival marketplace can (because of the shopping-history flywheel), and its automated tools become embedded in the customer's workflow. That makes revenue "repeat-driven" rather than "contracted." The upside is high incremental margin when spend returns; the downside is low visibility - the model has no subscription backlog to cushion a downturn.

5. Competitive Landscape

This is a concentrated, four-horse public race for US insurance customer-acquisition budgets, plus the carriers' own in-house direct-acquisition machines, which are the real long-term competitive threat.

The publicly traded pure-plays are EverQuote, MediaAlpha, QuinStreet, and LendingTree. Together the auto-lead-aggregator segment runs at roughly $1.5-2.0 billion of combined annual revenue.

• MediaAlpha (NYSE: MAX) is the closest comparable - an open, transparent programmatic exchange for insurance (and health) referrals, powering close to 10 million consumer referrals a month. Its bet is that it becomes the neutral routing rail for insurance shopping, including AI/ChatGPT-routed traffic. EverQuote's contrast is that it owns more of its consumer traffic directly and holds deeper proprietary shopping data, whereas MediaAlpha is more of a marketplace intermediary. They overlap heavily in auto P&C.
• QuinStreet (Nasdaq: QNST) is a diversified performance-marketing company spanning financial services and home services, with a large insurance segment. It competes in the same auto vertical but is less pure-play than EverQuote.
• LendingTree (Nasdaq: TREE) is historically a mortgage/loan marketplace that has pushed aggressively into insurance; insurance is now a major growth driver for it. It competes for the same carrier budgets.
• The carriers themselves are the structural competitor that does not show up in a peer table. Progressive and GEICO run enormous in-house direct-acquisition operations. Every dollar a carrier can acquire efficiently on its own is a dollar it does not need to spend through EverQuote. As carriers get better at direct acquisition (a genuine trend), they can compress aggregator margins over time.

Why EverQuote wins or loses. It wins on data depth and matching accuracy - the shopping-history flywheel produces better monetization per shopper, which lets it bid more for traffic and pay carriers a better return, a self-reinforcing loop. It wins on owned traffic and its emerging AI-search positioning. It loses, or is exposed, when carriers pull spend in a downturn (it has no contracted revenue to lean on) and when carriers get good enough at direct acquisition to route around aggregators.

Barriers to entry. Moderate and data-based rather than absolute. The barrier is the accumulated shopping data plus the carrier/agent relationships plus the traffic-buying scale - all of which a new entrant would need years and a lot of capital to build. But there is no patent, no regulatory license moat, and no physical asset. The AI-search transition is the wildcard: if consumer insurance shopping migrates to AI assistants, the incumbent traffic advantages could reset, which is exactly why all four players are racing to position there.

Market caps are peer-size references only, as of the dates shown; they move daily.

6. Industry

What drives demand. EverQuote's demand is derived from two things: how many consumers are shopping for insurance, and how much carriers are willing to pay to acquire them. Both are elevated right now. Consumers shop more when premiums spike (which they did sharply in 2023-2025), and carriers pay more to acquire when their underwriting is profitable and they want to grow. The single most important industry variable for EverQuote is the P&C underwriting cycle: profitable carriers advertise aggressively; unprofitable carriers go dark.

Size and trajectory. The US insurance industry is projected to spend over $14 billion on digital advertising in 2026, and auto-specific digital insurance advertising was expected to roughly double toward $4 billion by 2025. The broader US digital-insurance market has been growing at roughly a 20% CAGR. The structural tailwind underneath all of it is the slow migration of insurance shopping from offline (call your local agent) to online/performance channels - a shift performance-marketing players have argued for years is the future of auto insurance distribution.

A consumer-behavior nuance that shapes the model: roughly 74% of consumers research insurance online, but only about 25% purchase online; insurance remains a complex, often phone-closed purchase (about 78% of shoppers call a business after searching). That is precisely why EverQuote monetizes clicks, calls, and leads that hand a warm consumer to an agent to close - it does not need the consumer to buy online, only to shop online.

Regulation. Insurance is state-regulated in the US, and the lead-generation industry faces evolving rules on consumer consent and telemarketing (TCPA and related one-to-one consent standards for how leads can be contacted). Tightening consent rules can raise compliance costs and reduce the resale value of a lead, an industry-wide headwind that periodically reshapes lead economics.

Cyclicality. This is the defining feature. EverQuote is highly cyclical - not to the general economy, but to the P&C carrier profitability cycle. The 2022-2023 hard market (carrier losses) took the whole aggregator group down; the 2024-2026 recovery took it up violently. Anyone underwriting EverQuote is underwriting a view on where carriers are in that cycle.

7. Growth Triggers

All items are drawn from the six most recent earnings calls and attributed to the specific call.

• Carrier spend still below peak - embedded recovery runway. Management has repeated across multiple calls that roughly 80% of the top 25 carriers remain below their historical peak quarterly spend. (Q3 2025 concall, Nov 3, 2025; repeated Q1 2026 concall, May 4, 2026)

  "80% of top 25 carriers are still below peak spend." A single top-five carrier "more than doubled its expected spend during the back half of the quarter." (Q1 2026 concall, May 4, 2026)

• Home and renters insurance to outgrow auto in the medium term. Home grew 33% in Q1 2026; management is deliberately expanding the ~90/10 auto/home mix. (Q4 2025 concall, Feb 23, 2026; Q1 2026 concall, May 4, 2026)

• Smart Campaigns roll-out deepening across carriers, then agents and new verticals. Smart Campaigns 3.0 launched in Q3 2025 (7% ad-efficiency gain on migration); now adopted by "most carriers" and being "extended to local agents and new verticals." (Q3 2025 concall, Nov 3, 2025; Q4 2025 concall, Feb 23, 2026)

• AI-powered products extended to local agents for the first time. Smart campaigns and advanced bidding platforms rolling out to the ~5,000-agent base to lift products-per-agent (already up from ~1.0 to ~1.4). (Q4 2025 concall, Feb 23, 2026; Q1 2026 concall, May 4, 2026)

• LLM / AI-search traffic as a new consumer channel. Investing across three vectors - paid AI-platform advertising, AI content generation, and technical LLM integrations - expected to "become a channel of substance in 2026." (Q4 2025 concall, Feb 23, 2026; Q1 2026 concall, May 4, 2026)

• Agentic-AI operating leverage. "Agentic-first" software-development lifecycle and AI tools across the employee base; revenue more than doubled since 2023 with opex essentially flat, revenue per employee up nearly 3x. (Q4 2025 concall, Feb 23, 2026; Q1 2026 concall, May 4, 2026)

• The $1 billion revenue ambition - organic, no M&A. Reiterated repeatedly: reach $1 billion of annual revenue within two to three years through organic growth alone, while expanding EBITDA margin. (Q3 2025, Q4 2025, and Q1 2026 concalls)

  CFO Joseph Sanborn: the "$1 billion revenue path remains one where we need no M&A." (Q1 2026 concall, May 4, 2026)

8. Key Risks

• Carrier-spend cyclicality (high probability, high impact). This is the dominant risk and it is proven, not hypothetical. EverQuote has no contracted revenue; carriers can cut acquisition spend to near zero when underwriting turns unprofitable, exactly as they did in 2022-2023, when the company swung to a $51 million loss. Any renewed hard market - a spike in claims inflation, a bad catastrophe year, adverse loss trends - would compress carrier margins and pull spend, and EverQuote's revenue would fall fast with little to cushion it. Management's own recovery narrative (carriers "still below peak") is the mirror image of this: the recovery that is powering growth can reverse.

• Customer concentration in a few top carriers. A meaningful share of spend comes from the largest carriers, and management openly attributes swings to individual accounts (a single top-five carrier doubling spend late in Q1 2026 was called out as a driver). The upside quote is also the risk: if that same carrier pulls back, the quarter turns the other way.

• Carriers building direct-acquisition capability (structural, slow-burn). The largest carriers - Progressive, GEICO - run huge in-house acquisition machines. To the extent they get more efficient at acquiring customers directly, they need aggregators less, which pressures both volume and the price EverQuote can charge. This is a gradual erosion risk, not a cliff, but it caps long-run pricing power.

• AI-search disruption of the traffic model (uncertain, potentially large). EverQuote's advantage rests partly on owning consumer traffic acquired through today's channels (Google, Facebook, content). If insurance shopping migrates to AI assistants that answer "what's my cheapest policy" directly, the incumbent traffic-buying advantages could reset and the referral could disintermediate. Management is investing to be the marketplace those assistants route through - which is both the mitigation and an admission that the risk is real. Rival MediaAlpha is making the same bet, so this is an industry-wide reordering, not an EverQuote-specific one.

• Regulatory tightening on lead generation. Evolving consent and telemarketing rules (TCPA-style one-to-one consent) can reduce how leads may be contacted and resold, raising compliance cost and lowering lead economics across the industry.

• Governance / dual-class control. Chairman David Blundin and Link Ventures hold super-voting Class B stock, so public Class A holders have limited control. Notably, part of the buyback has been used to repurchase shares directly from Blundin-controlled Link Ventures (a $21 million block in August 2025), which is a related-party dynamic worth watching - capital is being returned partly to the controlling insider rather than only to the open market.

9. Walk the Talk

The six calls used: Q4 2024 (Feb 24, 2025), Q1 2025 (May 5, 2025), Q2 2025 (Aug 4, 2025), Q3 2025 (Nov 3, 2025), Q4 2025 (Feb 23, 2026), and Q1 2026 (May 4, 2026). The most recent is well within 90 days of today.

The through-line across these six calls is a management team that has, so far, consistently under-promised and over-delivered - but against the tailwind of a violently favorable industry cycle, which flatters the record and should be weighed accordingly.

Start at Q4 2024. Management framed 2024 as a breakout - revenue up 74% past $500 million - but was pointedly cautious about 2025:

CFO Joseph Sanborn: "the rate of increase in auto insurance premiums are forecasted to return to more normalized levels in 2025, which we expect will lead to our revenue growth rates also normalizing after the first quarter." (Q4 2024, Feb 24, 2025)

That was a deliberate expectation-lowering. It is also notable for what was absent: management did not yet put out the $1 billion target. The guidance for Q1 2025 was $155-160 million.

What actually happened in Q1 2025: revenue came in at $166.6 million, above the top of the guide. The "normalization" management warned about did arrive as slower growth than the 74% blow-out, but the absolute beat was clean. Through Q2 2025 ($156.6 million, +34%) and Q3 2025 ($173.9 million, +20%), the pattern held: growth decelerated from the post-crash bounce, exactly as guided, while each quarter beat its own forecast and set records on EBITDA and cash generation.

Somewhere in mid-2025 the ambition crystallized into the headline promise that now anchors the story:

The commitment to "achieve $1 billion of revenue while continuing to expand the cash generation of our marketplace" within two to three years, growing "20% and expanding adjusted EBITDA margin by 100 to 150 basis points per year on average." (Q3 2025, Nov 3, 2025; reiterated Q4 2025 and Q1 2026)

This is a trackable, dated, quantified commitment, and it has been repeated verbatim across three consecutive calls without walking it back - a good credibility sign. On the operating-leverage promise, the numbers back the talk: management's claim of "more than doubled revenues since 2023 while keeping operating expenses essentially flat" (Q4 2025) is consistent with the reported EBITDA expansion (FY2025 adjusted EBITDA up 62% to $94.6 million on 38% revenue growth). Saying you have operating leverage and then showing EBITDA growing far faster than revenue is walking the talk.

The clearest single beat came at Q1 2026. Management had guided Q1 2026 revenue to $175-185 million on the Q4 2025 call; actual revenue was $190.9 million, above the top end, and the stock rose sharply (reported +64%). The Smart Campaigns claims have also been specific and falsifiable rather than vague - "7% ad-spend efficiency improvement" on the 3.0 migration (Q3 2025), "most carriers adopted" by Q4 2025, products-per-agent up from ~1.0 to ~1.4 - the kind of concrete, checkable metrics that build trust.

Where should a skeptic push back? First, the guide-low-beat pattern is easy to sustain when carriers are climbing back from a trough; the real test of this management's credibility will come when the cycle stops helping. Second, the $1 billion target is still a promise, not an outcome - it depends on carriers returning to peak spend and on home insurance and AI-search traffic materializing as claimed. Third, the related-party buyback from the chairman's entity, while disclosed, sits slightly awkwardly against the "returning value to shareholders" language.

Net assessment: this is a management team that says specific, datable things and has hit or beaten them for six straight quarters, with reported operating leverage that matches its rhetoric. The record is genuinely good - but it has been earned entirely inside a rising-cycle, and has not yet been stress-tested by a downturn.

10. Shareholder Friendliness Index

Dividends. EverQuote has never paid a cash dividend and pays none today. It has no dividend policy, no per-share payout, and no signal of initiating one. For a company that only returned to sustained profitability in 2024, this is unremarkable; capital is being retained and partly returned through buybacks instead.

Buybacks and dilution. The board authorized a $50 million Class A repurchase program in July 2025 (expiring July 22, 2026), the company's first meaningful capital-return step (SEC 8-K, July 2025). Execution has been steady: a $21 million block repurchased from chairman David Blundin's Link Ventures entities in August 2025 (900,000 shares at $23.33, a small discount to market), roughly $30 million cumulative by the Q4 2025 call (including ~$9 million in early 2026), and a further ~$19.9 million (about 1.1 million shares) in Q1 2026, per the Q1 2026 results. The MoatMap database shows zero open-market buybacks recorded in the trailing ~90-day window (since April 4, 2026), consistent with the program approaching its authorized limit and its July 2026 expiry - so the buyback activity is concentrated in the July 2025-March 2026 window, not the most recent quarter. Against this, option and RSU vesting continually creates new shares, so the net share count has been roughly flat rather than meaningfully shrinking; the buybacks have functioned largely to offset dilution rather than to retire a large chunk of the float. One flag: a material portion of the buyback went to the controlling insider rather than the open market, which softens the "return to public shareholders" read.

Verdict: Neutral - the company has begun returning capital via a modest buyback and carries a fortress balance sheet (~$178.5 million cash, no debt), but it pays no dividend, the buyback has mostly offset dilution rather than shrunk the count, and part of it benefited the controlling shareholder.

11. Insider Activities

Source and mode. EverQuote trades on Nasdaq, an open venue, so the MoatMap database block (US SEC Form 4 data) is the spine, cross-checked for the most recent two weeks. The block is current to July 2, 2026 and is not marked stale.

The picture: routine, compensation-driven selling, and zero open-market buying. Over the trailing twelve months the block records 22 transactions across nine insiders - 0 open-market buys, 10 open-market sells, and 12 "other" (grants and equity actions). Every "sell" is small, executed by officers, and consistent with routine RSU-vesting tax coverage and pre-planned disposals rather than any conviction signal. There is no cluster of buying and no single large purchase - the strongest bullish signal in this section (an insider buying on the open market) is simply absent.

Buys - the signal. There are none. No insider made an open-market purchase in the trailing twelve months. That is the notable absence: despite management's public conviction in the $1 billion target, no officer or director backed it by buying stock on the open market.

Sells - the why. The sells are uniformly small and confined to officers (CFO Sanborn, CAO Ayotte, CTO Brainard) and one modest director sale (Neble). The size and cadence - a few hundred to a few thousand shares, clustered right after the April 1 and May 20 grant/vesting dates - are the fingerprint of routine tax-and-diversification disposals tied to equity vesting, not signals about the business. The June 4 director "$0 / Other" lines are the annual board RSU grants (equal 9,105-share awards to all five outside directors), not sales. Reasons are not individually disclosed in footnotes here, but the pattern is textbook compensation housekeeping.

Net assessment. Insiders are modest net sellers by transaction count, but the dollar amounts are trivial relative to the company's size and the activity is entirely explained by grant/vesting mechanics rather than any change in outlook. The read is neutral: no bullish open-market buying to lean on, but also no large or unusual selling that would raise a red flag. The one governance note that carries more weight than these small Form 4 trades is the separate $21 million buyback of chairman Blundin's Link Ventures stake in August 2025 - a partial insider cash-out executed through the corporate balance sheet rather than the open market.

12. Scenarios

Bull case. The P&C carrier cycle keeps improving and the 80% of top-25 carriers still below peak spend climb back toward - and eventually past - their prior appetite, pouring budget into EverQuote just as its matching gets smarter. Home and renters, growing faster than auto off a small base, becomes a genuine second engine and de-risks the auto concentration. Smart Campaigns and the new agent-facing AI tools embed EverQuote's software into carriers' and agents' daily workflows, lifting wallet share and making the platform sticky in a way a pure lead vendor never was. The AI-search transition, feared by everyone, turns into EverQuote's advantage: its LLM integrations make it the marketplace that AI assistants route insurance shoppers through, and its hundreds of millions of historical shopping events give it a matching edge no chatbot can replicate. Operating leverage does the rest - revenue crosses $1 billion organically within the two-to-three-year window while EBITDA margin keeps expanding, and the company is a far more valuable, more diversified, more software-like business than the lead-gen shop it started as.

Base case. The recovery continues but at a decelerating pace, roughly as management has guided - high-teens-to-20% growth rather than the 74% post-crash bounce. Auto stays the overwhelming majority of revenue; home grows faster but stays a single-digit-to-low-teens share of the mix for a while. Smart Campaigns adoption deepens and the agent-tools push modestly lifts products-per-agent, giving EverQuote incremental monetization without transforming the model. AI-search traffic becomes a real but supplementary channel rather than a step-change. Management keeps its guide-low-beat cadence and inches toward the $1 billion target, and buybacks roughly offset option dilution. It looks like a well-run, cash-generative, still-cyclical marketplace that is executing its plan without either a blow-up or a re-rating catalyst.

Bear case. The auto underwriting cycle rolls over. A claims-inflation resurgence, a heavy catastrophe year, or adverse loss trends push carriers back toward unprofitability, and - as they did in 2022-2023 - they slash acquisition spend. Because EverQuote has no contracted revenue, the top line falls quickly, the variable-marketing spread compresses, and the operating leverage that looked so good on the way up works in reverse. Concentration bites: one or two top carriers pulling back takes a disproportionate chunk of revenue. Underneath the cycle, the structural threat plays out too - the biggest carriers keep improving in-house direct acquisition and route around aggregators, while AI assistants begin answering insurance-shopping queries directly and disintermediate the referral entirely, resetting EverQuote's traffic advantage before its own AI-search bets have scaled. The $1 billion target quietly disappears from the script, and the stock re-rates back to a cyclical lead-gen multiple.