Software Enabled Services Valuation: The SaaS Margin Gap

Two software companies can bill the same amount every year and still sell for very different prices. A lot of the time the reason has nothing to do with the software. It is how many of your own people have to show up to get each new customer running.

Take nCino, which sells software to banks. In the year that ended January 31, 2026, it brought in $594.8 million. After paying the direct cost of delivering all of it, the company kept 61 percent. On the software subscriptions by themselves, it kept 71 percent.

The rest of the business is the part where people install the software and get each bank using it. That work brought in $71.6 million and cost $85.1 million to deliver. It loses money on the way through.

Here is why that matters when you sell. A buyer does not pay you on your best line. They add up everything you bill, subtract what it costs you to deliver, and look at what is left, because that whole thing is what they are buying. The more your product needs people to make it work, the less of your revenue reads as software.

By the end of this post you will know how to measure how much of your revenue leans on your team, how a buyer decides whether you are selling a product or selling labor, and 3 things you can clean up before you go to market.

The number that drives software enabled services valuation

Your subscription margin is a line item. Your blended margin is the valuation.

Gross margin is simply what is left from a dollar of revenue after the direct cost of delivering it. For software that cost is hosting. For implementation it is people, and people do not get cheaper at scale the way servers do.

The 2026 benchmarks show the split clearly. In the Aleph and Benchmarkit 2026 SaaS and AI Performance Benchmarks, which covered 342 companies on full year 2025 data, the median software gross margin was 80 percent. The median blended margin across all revenue was 76 percent.

This is the same mechanism that decides why not all recurring revenue gets valued the same. Revenue that arrives without human effort is worth more per dollar.

The 10 point gap, one number at a time

Walk nCino’s fiscal 2026 income statement slowly. It is a genuine software business carrying heavy implementation, and it reports both lines separately.

Start with revenue. Subscription revenue was $523.1 million. Professional services and other revenue was $71.6 million. Together that is $594.8 million, so services are 12 percent of the top line.

Now the costs. Delivering that subscription revenue cost $149.6 million, which leaves a subscription gross margin of 71 percent. Delivering the services cost $85.1 million.

Read that pair again. The services brought in $71.6 million and cost $85.1 million to deliver. The services line loses money. It runs at roughly negative 19 percent gross margin, and pulls the blended margin down to the 61 percent the company reports.

Twelve percent of revenue moved the headline margin by 10 points. That is the pull a small services line has on a valuation. Buyers do not apply one multiple to the whole business. They value the subscription stream and the services stream separately, which is why $8 million of subscription plus $2 million of services does not price like $10 million of subscription.

The Onboarding Slope Test

Presence of services is not the problem. Buyers see implementation revenue at almost every vertical software company. What they test is whether that cost falls as you grow. I call it the Onboarding Slope Test.

Pull your last twenty new customers in the order they signed. For each one, record the internal hours spent between contract signature and the customer being live. Not the invoiced hours. All of them, including the engineer who fixed the data import on a Saturday.

Then compare the first ten to the last ten.

Treat the 20 percent figure as a working rule, not a law. Direction is what a buyer reads, and a flat or rising slope fails at any cutoff. The mechanism is simple. Software margin comes from building once and selling many times. Every hour of custom work at a new customer is an hour that did not get amortized across the base. A declining slope proves the product is absorbing the work. A flat slope proves the people are.

One 2026 analysis of SaaS revenue mix puts services in dilutive territory once they pass roughly 15 to 20 percent of total revenue, with professional services typically running at 20 to 35 percent gross margin against subscription margins more than twice that.

When services do not hurt your SaaS valuation

Here is the counter example, and it is the same company. nCino loses money on implementation, on purpose, and the market still treats it as a software business.

Three things make that work. The services share is bounded and shrinking, down from 13.2 percent of revenue in fiscal 2025 to 12.0 percent in fiscal 2026. The services exist to land and protect subscription revenue rather than to earn a profit. And the subscription line underneath is healthy enough at 71 percent to carry the drag.

Companies that require a long guided rollout carry structurally more implementation cost than self serve products, which is part of how buyer diligence differs for product led and sales led SaaS. A buyer will forgive a services line that meets those three conditions. What a buyer will not forgive is a services line that grows faster than subscription, because that is the signature of a company selling custom work with software attached. The direction of travel is the signal, not the presence.

What buyers flag, and what to fix first

Diligence here is not subtle. The analyst hunts for human work hidden inside a software cost structure.

Three moves change the picture, in this order. First, reclassify delivery labor honestly and restate at least eight quarters. Second, clear the onboarding backlog, because customers who signed but are not live are revenue a buyer discounts and churn risk they price. Third, turn your three most repeated custom builds into product configuration.

Do the reclassification first, even though it lowers your reported margin. An honest number you volunteer beats a higher one the buyer discovers, and it is the same discipline that decides how your gross margin gets read in a sale.

Frequently Asked Questions

Next Steps

Run the Onboarding Slope Test on your last twenty customers this week. A buyer will build their own version of that number, and knowing it first is the difference between explaining your margin and defending it in a retrade.