A new industry analysis published via Barchart’s GetNews platform is drawing attention for its detailed breakdown of how a new category of AI product – so-called “answer engines” – is changing the fundamental business model of search and discovery across multiple sectors, including online gambling.
The report, titled “The Business Model of Answer Engines: How Vertical AI Changes Unit Economics,“ argues that the financial significance of these tools is not their interface design, but their ability to fundamentally alter the cost and revenue dynamics of discovery-led businesses.
What Is an Answer Engine?
According to the analysis, answer engines differ from traditional search tools in one critical way: instead of returning a list of links, they resolve the user’s question directly, then continue the conversation to refine intent and drive a decision.
The business model implication is significant. Traditional search monetisation depends on ads sold against traffic, affiliate commissions driven by clicks, or subscriptions for research tools. Answer engines, the report argues, can re-bundle this value in a way that increases pricing power — because they move users from browsing to resolution faster, changing what companies can charge for and what it costs to serve each user.
The analysis identifies three conditions under which vertical AI tools consistently outperform traditional search:
- The domain is complex and changes frequently
- Users have constraint-heavy questions
- Mistakes are costly — financially, legally, or reputationally
“That is why legal and healthcare vertical AI have commanded premium pricing,” the report states, “and why regulated consumer categories like iGaming are investing in structured, domain-specific search.”
The Numbers Behind the Model
The analysis benchmarks two established vertical AI companies to illustrate the unit economics at play.
Harvey, the legal AI platform, is cited with an estimated $75 million ARR as of April 2025 – up from $50 million at the end of 2024 – at a pricing of roughly $1,200 per lawyer per month, with minimum contract sizes in the tens of seats. The report describes this as “classic high-ARPA SaaS economics,” where a relatively small customer base can generate meaningful revenue at high gross margins.
Abridge, a healthcare AI tool for clinical documentation, is cited with an estimated $100 million ARR as of May 2025 at pricing of approximately $2,500 per clinician per year, supported by a reported $300 million Series E raise at a $5.3 billion valuation in June 2025, with deployments across more than 150 health systems.
The report then examines how the same unit economics apply in consumer discovery markets – specifically iGaming, where monetisation has historically been driven by affiliate commissions rather than subscriptions.
It uses marvn.ai, an AI-powered search engine developed by Marlin Media Ltd., as its primary case study for the “outcome-led lead generation” model.
According to the analysis, the financial logic is straightforward: traditional affiliate economics are noisy and often opaque. An iGaming Business audit cited in the report found that advertised revenue share deals can compress dramatically after fees, with average net revenue share falling to 23.91 percent and dropping as low as 8 percent in the worst cases. That kind of compression, the report argues, puts pressure on everyone in the chain — publishers chase volume, operators compete for placement, and users become sceptical of ranked lists.
An answer engine changes the funnel by monetising what the report calls “resolved intent” – users who have already filtered by jurisdiction, payment preferences, bonus type, and wagering conditions before clicking through to an operator. Fewer clicks, but significantly better quality.
Marlin Media, the analysis notes, is explicit that it operates as a lead generation business, and that marvn.ai – which draws on partnerships with more than 500 licensed iGaming brands — functions as an AI-powered complement to its existing publisher portfolio. The platform is free to end users, consistent with standard lead generation economics where the value sits downstream in conversion quality rather than direct user fees.
Why the “Discover” Feature Matters Financially
The report also examines marvn’s Discover section, launched in January 2026, which functions as a news and knowledge hub — scanning reputable sources and generating topic overviews, with follow-up query capability built in. From a unit economics standpoint, the analysis argues this matters for two reasons. First, it drives higher session frequency: users return for updates rather than only when they have a specific search query. Second, it improves top-of-funnel activation by giving undecided users guided entry points before they have a specific question to ask.
Both effects, if they hold, lower effective customer acquisition cost over time by building organic retention and repeat usage — the most durable form of traffic for any discovery platform.
The Four Questions Investors Should Ask
The analysis closes with a framework for evaluating any answer engine as a business:
- Can it raise revenue per user session? Through higher conversion rates, better partner economics, or subscription ARPA.
- Can it lower the effective customer acquisition cost? Through repeat usage, direct traffic, and reduced dependence on platform algorithms.
- Can it keep marginal costs under control? Data freshness and verification workflows — not model inference costs — are typically the margin driver at scale.
- Can it compound a defensible asset? The moat in vertical AI is the dataset and the feedback loop it generates, not the underlying language model.
“The winners will not be the tools that generate the most text,” the report concludes. “They will be the ones that turn domain complexity into measurable business leverage, then do it at scale.”
The full analysis is available via Barchart’s GetNews platform.