AI-Native Services Will Redraw the Line Between Software and Consulting

In May 2026, Gainsight¹ stood on stage at Pulse and told its customers something no major SaaS company had said before: don’t just buy our software, hire us to manage your renewals.

The offering is called Atlas, and Gainsight describes the model as retention-as-a-service. AI agents run the renewal motion across the long tail of a company’s customer base, from personalized outreach to contract negotiation. Humans step in for judgment, escalations, and edge cases. Gainsight signs outcome-based contracts and takes accountability for GRR and NRR.

That announcement gave a public reference case to a business model that has been forming quietly for two years: AI-native services.

What AI-native services are

The cleanest way to define AI-native services is by what the customer buys, not by the technology the provider uses.

AI-native services (AINS) are managed services where AI agents perform most of the day-to-day execution, humans provide oversight and judgment, and the provider is accountable for delivering a business outcome rather than providing software. Customers of traditional SaaS buy software and do the work themselves. Customers of AI-native services buy the result. The provider combines AI agents, software, playbooks, and human expertise into a service that owns execution and is measured on outcomes.

The contrast with traditional SaaS is structural:

None of this describes a software company adding a copilot feature. AI-native services invert the relationship between vendor and customer. The vendor stops handing over tools and starts taking over work.

Why SaaS companies are moving first

Gainsight’s move shows the logic. The company spent fifteen years building software, playbooks, and benchmark data for customer success teams. Its customers still faced the same coverage problem: thousands of long-tail accounts with no CSM assigned, renewals handled by nobody, churn discovered after the fact. Outsourcing to a BPO adds bodies without adding quality.

Atlas answers that problem with renewals pods that pair agents with humans. The agents execute outreach, risk assessment, and playbook actions across thousands of accounts. Renewal Agent Managers intervene where judgment matters. Everything reports through the platform, and the contract ties payment to results. Gainsight now sells three paths: buy the software and agents, build custom agents on the platform, or hire Gainsight to deliver the outcome.

Jake Saper of Emergence Capital² framed the stakes at the announcement: the most important strategic question facing every SaaS leader right now is whether to stop selling the seat and start selling the result.

The economics explain the timing. Seat-based pricing weakens as AI reduces the number of humans who need seats. At the same time, agents have become capable enough to execute real workflows, which means the vendor who owns the domain knowledge, the data, and the playbooks can now operate them instead of licensing them. For SaaS companies whose category has a measurable outcome and a repeatable workflow, the service business sitting on top of their software is becoming the bigger prize.

This path will suit some SaaS companies and destroy others. The model demands honest accountability for outcomes, and a vendor whose product only assists an outcome cannot credibly own it.

Consulting firms face the same shift from the other direction

Consulting sells judgment delivered through human labor, priced mostly by time. The traditional leverage model runs on a pyramid of analysts and associates doing execution work under partner supervision. Agents now do a growing share of that execution work at near-zero marginal cost, which breaks the pyramid’s economics.

Consulting firms hold exactly the assets AI-native services need: deep domain knowledge, tested methodologies, and client trust. A firm that codifies its methodology into agent-executable playbooks can sell outcomes at a price and speed that leverage-model competitors cannot match. A firm that keeps billing hours will find itself competing on rates against providers whose cost of execution keeps falling.

The strategic position of SaaS companies and consulting firms is converging on the same destination. Software companies are adding the service layer. Service companies are adding the software layer. Both end up selling operated outcomes.

Pure-play AI-native services companies will emerge

The third group starts with no legacy model to protect. Emergence Capital, which has invested over $150M in the category, has documented the pattern in its AI-Native Services Playbook²: new companies built AINS-first, targeting vertical service markets with high-volume, repeatable workflows. Insurance claims, fund administration, legal work, compliance, revenue operations. These companies carry no seat-based revenue to cannibalize and no partner pyramid to unwind.

Emergence attaches a warning worth keeping. Revenue growth alone proves nothing in this model, because strong growth can mask a business where humans quietly do the work behind an AI story. They call it Mirage PMF. The real test is whether AI performs a material share of the work at expanding margins while outcomes improve. Companies that fail that test are conventional services firms with better marketing.

Where the IP actually lives

The uncomfortable fact for anyone building in this category is that the AI models themselves confer no advantage. Every provider rents the same frontier models at the same prices.

The durable IP sits in three places. First, business knowledge: the playbooks, decision rules, and domain judgment that tell agents what good execution looks like in a specific context. Second, human orchestration: knowing where agents run autonomously, where humans intervene, and how to staff the boundary between them. Third, the optimization loop: every engagement generates data on what worked, and providers who feed that back into their playbooks get better with each customer while their imitators start from zero.

Compressed to a formula: AI execution + codified business knowledge + human orchestration = an outcome a customer will pay for.

The moat is operational. It lives in the playbooks and in the people who tune the system, which is why domain credibility decides who wins early customers in this market.

SaaS companies will start with the outcome they are built to own

Notice which outcome Gainsight chose to sell first: retention. Gainsight is a customer success company. Its customers already use the platform to reduce churn and improve retention, and many of them have cut their CS teams or never built one. Growing companies rarely have the time or the appetite to staff a customer success function. For Gainsight, running retention as a service was the logical extension of what its software, its playbooks, and its people already know best.

Expect the same pattern from every SaaS company entering this market. The first service to launch will be the top outcome aligned with the company’s technical and business-practice strengths, because that is where the playbooks run deepest and the credibility already exists. Retention carries an added advantage for anyone who owns it: the outcome is contractual, measurable, and already expressed in the metrics boards track. A provider can sign up for NRR in a way it never could for “brand awareness.”

The software industry spent twenty years teaching customers to buy tools and do the work themselves. The companies that define the next decade will be the ones that take the work back.

A6 Group is building its own AI-native services on this thesis, in the domains where we advise today. We will have more to say when we launch.

References:
1  Gainsight Atlas: https://www.gainsight.com/atlas/
2  Emergence Capital Partners: https://www.emcap.com/thoughts/the-ai-native-services-playbook