Three of the most consequential investors in Netomi's Series C are not financial backers. Accenture is committing hundreds of consultants to the deployment playbook. Adobe is wiring Netomi into Brand Concierge so the platform sits inside the digital experience layer of most of the Fortune 100. Jeffrey Katzenberg's WndrCo, joining the board, brings the consumer media and entertainment perspective that pure enterprise rounds usually lack. Reuters reported the round at $110 million on April 30, 2026. The cap table is the smaller story. The bet inside it is the larger one.

The bet is composable. Adobe owns intelligence and content. Netomi handles agentic execution in regulated, high-stakes environments. Accenture supplies the integration capacity. None of these three companies, on its own, claims to be the customer experience platform. Together, they are arguing that no single vendor will be, and that the assembled stack outruns the monolithic one because each piece is the best at the layer it owns.

Anjul Bhambhri, the senior vice president at Adobe overseeing the integration, has framed the partnership as a recognition that customers move across channels faster than any single platform can follow them, which is why Adobe is wiring Netomi into Brand Concierge rather than building the execution layer itself. That framing matters. Adobe is the company most exposed to the question of whether intelligence and content can be unbundled from execution, and the answer it is giving with this round is yes.

That argument cuts directly across what Salesforce, ServiceNow, and Creatio are each, in their own way, building toward.

Netomi exists because customer service breaks under pressure

Founder and chief executive Puneet Mehta started the company as msg.ai in 2015, before the transformer architecture that powers today's large language models had been published. The team stitched together earlier model types to make a chatbot work for a Sony Pictures film promotion, then went through Y Combinator's winter 2016 batch. The company renamed to Netomi in 2018 and raised a $14.7 million Series A from Index Ventures the next year, followed by a $30 million Series B led by WndrCo in 2021.

Mehta's career before Netomi is the part that explains the product. He spent roughly seven years building low-latency trading systems at JPMorgan Chase, Merrill Lynch, and Citi, where reliability was not a goal but a precondition. He has cited that work as the design constraint for Netomi: the architecture has to hold up when traffic spikes, when the customer is angry, and when a wrong answer creates real liability. The company describes its platform as a dual-agent design that pairs deterministic action agents, which execute defined transactions, with reasoning agents driven by language models, which interpret intent and shape the conversation. The deterministic layer is what keeps the probabilistic layer inside the lines.

That architecture is why United Airlines, Delta Air Lines, MetLife, DraftKings, the NBA, and Paramount are named customers. Netomi reports zero failures, broken guardrails, or brand violations across deployments. The figure is vendor-supplied and unaudited, but the customer list itself is the more useful proof point. Airlines during weather events and sportsbooks during peak betting windows are the environments where customer service infrastructure tends to publicly fail.

Three platforms, three answers to one question

Do you assemble the customer experience stack, or do you own it? That is the question the Netomi round forces. Salesforce, ServiceNow, and Creatio each answer it differently, and the differences matter more than any feature comparison.

Salesforce is betting that ownership is defensible if the data sits in one place. Headless 360, which I covered earlier this month, makes the platform reachable to any agent without requiring a user interface or a per-seat license. The Agentforce Experience Layer separates the agent action from the surface where it renders, which means Salesforce no longer needs to control the front end to deliver the value. The CCaaS launch in March pulled telephony into the same data record. The pattern is consistent: every customer touchpoint feeds the layer Salesforce already owns. If the bet works, ownership becomes irreversible because the data gravity makes leaving prohibitive.

ServiceNow is betting that ownership is inevitable if the back office is where the work actually happens. Context Engine, announced in April, treats institutional decision history as the asset that makes agents reliable. The argument from president Amit Zavery is that agents fail at scale because they lack the context to apply policy correctly, and that context lives in workflows rather than in conversations. The Moveworks acquisition gave ServiceNow an employee-facing front door. The Build Agent Skills SDK lets developers code in Cursor or Claude Code and deploy into ServiceNow Studio with governance attached on the way in. The platform is positioning itself as the system of action behind whatever surface the customer sees.

Creatio is betting that the question is irrelevant for the part of the market that cannot afford the answer. The Growth tier at $25 per user per month is a fraction of comparable Salesforce pricing. The no-code agent building lets a business analyst, not a systems integrator, modify workflow logic. The company reports about 900 employees and 40 percent annual growth, both vendor-supplied. For mid-market organizations and lean information technology teams, time to value beats architectural elegance. Creatio is not trying to win the platform argument. It is trying to make the platform argument unnecessary.

Three vendors, three stakes. Salesforce loses the argument if interoperability lowers the cost of leaving the data layer. ServiceNow loses if customer experience leaders, not just chief information officers, refuse to grant the company brand permission to own the customer-facing experience. Creatio loses if accessibility stops being enough, because the rest of the market has standardized on protocols Creatio has not adopted.

One protocol decides which bet survives

The Model Context Protocol was donated by Anthropic to the Linux Foundation in December 2025, where it became a founding project of the new Agentic AI Foundation alongside Block's goose framework and OpenAI's AGENTS.md. The platinum members of the foundation include Amazon Web Services, Google, Microsoft, Bloomberg, and Cloudflare. The protocol is no longer Anthropic's. It is becoming the default way agents from different vendors discover, communicate with, and act on each other's tools.

That single fact reshapes the question every customer experience leader is trying to answer. If composability is cheap, ownership is harder to defend on architectural grounds and has to be justified on data gravity, governance, or industry depth. If composability is expensive, the monolithic platform wins by default because integration tax exceeds platform license cost.

Salesforce and ServiceNow have both publicly committed to MCP and to the Agent2Agent protocol. Their bet on ownership now rests on something other than integration friction, because the friction is being engineered out of the market. Adobe supports MCP endpoints and partners broadly across language model providers, which is consistent with the composable thesis the Netomi round expresses. Creatio has not engaged publicly with either protocol. For a platform whose entire pitch is accessibility and speed of deployment, absence from the interoperability conversation is not a feature gap. It is a strategic exposure.

The Netomi round is the first significant capital deployment that takes the new protocol environment seriously. The investors are not betting on Netomi as a future platform. They are betting that the platform is no longer the unit of value, and that an assembled stack of best-in-layer components will outrun a single vendor trying to do all of it.