Why Vertical SaaS Beats Horizontal in Niche Markets

Veeva Systems and HubSpot finished their most recent fiscal years within $65m of each other in revenue. Veeva got there with 1,552 customers. HubSpot needed 288,706. That is a 190-fold difference in revenue per customer between a vertical software company and a horizontal one at the same scale, and it is the answer to the objection that vertical markets are too small.

The TAM objection, and the arithmetic that answers it

The objection goes: there are only so many pharmaceutical companies, construction firms or law practices, so a vertical product hits a ceiling. It is arithmetically true and strategically irrelevant, because addressable market is buyers multiplied by spend per buyer, and vertical companies compound the second term while horizontal companies compete on the first.

Veeva’s fiscal 2026, ended 31 January 2026, produced $3,195.3m of revenue, up 16%, of which $2,684.2m was subscription revenue, with GAAP operating income of $916.4m, up 33%. Across 1,552 customers that is roughly $2.06m of revenue per customer per year. HubSpot’s 2025 revenue of $3.13bn came from 288,706 customers, with average subscription revenue per customer of $11,683 in the fourth quarter. Veeva has told investors it passed a $3bn revenue run rate in early 2025 and is targeting $6bn by 2030.

Vertical companies expand along three axes rather than by adding logos.

1. Product count inside the same account

Procore’s 2025 results make the mechanism visible. Revenue was $1,323m, up 15%, from 17,850 organic customers. Gross revenue retention was 95% and net revenue retention 106%, on GAAP gross margin of 80%. The numbers that matter: 78% of ARR came from customers using four or more products, and 52% from customers using six or more. Customers paying more than $100,000 a year numbered 2,710, up 16%, and those above $1m numbered 115, up 34%. Procore is not winning by finding more construction firms. It is winning by selling six things to the same firm.

2. Payments and financial services attached to the workflow

This is the largest single lever, and the arithmetic is documented. Andreessen Horowitz’s Fintech Scales Vertical SaaS, first published in August 2020 and updated in September 2024, argues that “by adding fintech, SaaS businesses can increase revenue per customer by 2-5x.” Its worked assumption is a vertical small business spending roughly $1,000 a month across all vendors with only about $200 of that on software, leaving the balance available to a platform that can move the money. The authors cite Mindbody earning roughly $150 a month in software fees plus about $100 a month from payments per customer.

3. Acquisition, at the mature end

Toronto-based Constellation Software has industrialised the third route. Its 2025 annual results show $11,623m of revenue, up 15%, with organic growth of only 4% (3% adjusting for currency) and $1,579m spent on acquisitions during the year. The company describes itself simply as acquiring, managing and building vertical market software businesses. If a single vertical really does cap out, owning several hundred of them does not.

Embedded payments as a multiplier, with the margin caveat

Toast is the clearest public case. In full-year 2025 it reported $6,153m of total revenue across roughly 164,000 restaurant locations, of which $5,037m (82%) came from financial technology solutions and $936m from subscription services, on $195.1bn of gross payment volume, with annualised recurring run-rate above $2.0bn and a record 30,000 net new locations added in the year.

The margin structure is the part founders miss. Toast’s fintech revenue of $5,037m produced $1,146m of gross profit, a 22.8% margin, while $936m of subscription revenue produced $672m, a 71.8% margin. Payments revenue is largely pass-through interchange. It is still worth having, since fintech gross profit was 1.7 times subscription gross profit and is where most of Toast’s gross profit dollars now come from, but a founder who models an embedded payments line at software margins will be wrong by a factor of three.

Shopify shows the same shape at larger scale. Its 2025 results report $11,556m of revenue split $2,752m subscription solutions and $8,804m merchant solutions, on GMV of $378.4bn, with gross payments volume up 37%. Merchant solutions, largely payments, is 76% of revenue. The software subscription is the wedge; the transaction flow is the business.

The practical consequence is that a vertical company with a payments attach can afford a customer acquisition cost several times higher than a pure software company selling into the same buyer, which is precisely why the incumbent horizontal vendor cannot follow it down.

The Canadian cohort

Canada has produced an unusual concentration of vertical software at scale, and payments sit near the centre of most of it. Constellation and Shopify are two of the country’s largest technology companies by revenue, and both fit the pattern above. The most instructive newer case is Clio, the legal practice management company based in Burnaby, British Columbia.

TechCrunch reported in June 2025 that Clio had reached $300m of annual recurring revenue as it agreed to acquire legal data company vLex for $1bn, having been valued at $3bn a year earlier and $1.6bn in 2021. Clio confirmed in November 2025 that the acquisition had closed alongside a US$500m Series G and a US$350m debt facility at a US$5bn valuation, serving hundreds of thousands of legal professionals in more than 130 countries.

Clio’s product set now combines practice management, invoicing, electronic payments and legal research: three of the expansion axes above, running simultaneously, in a vertical whose buyer count is essentially fixed. Nobody is creating new law firms at venture scale.

Why vertical wins where it wins

Three structural advantages, none of them about features. Distribution is cheap, because the buyers read the same three trade publications, belong to the same association and attend the same annual conference, which compresses customer acquisition cost in a way no horizontal category allows. The competitive set is small and usually consists of on-premise software written in the 1990s plus a horizontal tool being used badly. And regulatory or workflow specificity creates switching costs that a horizontal platform cannot replicate without effectively becoming a vertical company itself, at which point it has to build and support a separate product line for every industry it serves.

That last point is why the horizontal-platform-plus-vertical-app model has largely given way to full vertical stacks. The vertical layer is where the pricing power sits, so the companies that started as applications on someone else’s platform have steadily moved to owning the platform too.

When vertical is the wrong choice

• No money moves through the workflow. Without a payment, payroll, lending or insurance flow to intermediate, you lose the 2x to 5x revenue-per-customer lever and are left selling seats into a fixed buyer count.
• The buyer count is small and ACV cannot be large. Three thousand buyers at $10,000 a year is a $30m business at full penetration. That can be an excellent company and a poor venture case, and pretending otherwise just delays the reckoning by a funding round.
• The vertical has no shared workflow. Some industries look like a vertical from outside while every firm inside operates differently. You end up building bespoke software and charging SaaS prices for it, which works until the second customer.
• The incumbent is modern and competent. Verticals with a well-run current-generation incumbent are the hardest to enter, precisely because the switching costs that would protect you are currently protecting them.
• You cannot price above the churn cliff. A vertical product priced like a consumer app inherits consumer retention. If the buyer will only pay $30 a month, the specificity that makes vertical software defensible never gets paid for.

The test to run before you commit to a vertical

Estimate three numbers for your target industry. First, the count of firms that could plausibly buy. Second, the annual dollars each currently spends on software for this workflow. Third, the annual dollars that flow through the workflow itself: payments, payroll, financing, claims.

Multiply the first by the second for your software market, then apply a realistic take rate of 50 to 100 basis points to the third. If the second number is small and the third is large, you are looking at the shape that produced Toast and Clio. If both are small, you have a good business and a bad venture-scale bet, and the honest move is to capitalise it accordingly rather than to argue the market size upward in a deck.

Sources

• Veeva Systems — Fourth Quarter and Fiscal Year 2026 Results
• Procore — Fourth Quarter and Full Year 2025 Financial Results
• Toast — Fourth Quarter and Full Year 2025 Financial Results
• HubSpot — Q4 and Full Year 2025 Results
• Constellation Software — Results for the Year Ended 31 December 2025
• Shopify — Fourth Quarter and Full Year 2025 Financial Results
• Andreessen Horowitz — Fintech Scales Vertical SaaS
• Clio — Completes vLex Acquisition and Series G at US$5bn Valuation