The Outcomes Architect Owns Adoption and the Renewal

Notion built a department around one job title

Pull Notion’s public job board on 7 September 2026 and thirteen open roles carry Outcomes in the title. Eleven of them sit in a department named, literally, Outcomes Architect. That is not a team tucked under Sales, it is its own org. The postings span Dublin, London, New York, Paris, San Francisco and Tokyo, and there are separate manager roles to run it.

The same day, OpenAI lists seventeen Applied AI Architect openings, fifteen of them on a team called Technical Success, with locations from San Francisco and New York to Munich, Singapore, Delhi, Tokyo and Seoul. Cursor keeps an AI Deployment team inside Customer Success and is hiring both an AI Deployment Manager and a Regional Director above it. Gusto calls the role Agentic Operations Architect, Databricks calls it Senior Customer Enablement AI Programs Manager, Ramp calls it AI Solutions Strategist. The titles disagree; open the descriptions and the work overlaps.

From designing the system to carrying the renewal

Vendors already had customer-facing technical roles. A solutions architect draws the architecture and runs the proof of concept before the contract. A forward deployed engineer sits inside the customer and builds the thing. Both lean technical, and someone on the account team carries renewal and expansion.

The role splitting off now breaks that division. Notion’s New York posting describes it as a build-beside partner and hands one person workflow adoption, value delivery and renewal outcomes across a book of named accounts. The staffing priority is explicit: accounts that are flat or at risk go here first, and this person owns the commercial conversation on those renewals while partnering with solutions consultants on expansion strategy.

OpenAI draws the same line from the other side. The Applied AI Architect posting, Seoul included, calls the role the CTO of your book of business. The account director owns commercial strategy; this person owns technical strategy, the customer journey and the path from pilot to production. Between the people who sell and the people who build, a layer that owns the result has appeared.

The success metrics in the posting are the job description

Cursor’s AI Deployment Manager posting names three measures: customer retention, product adoption, revenue expansion. Everything listed underneath follows from them. Make the return on investment legible to CTOs and engineering VPs. Run enterprise rollouts end to end, from onboarding through internal champions to scaled adoption. Watch account health and usage data to push renewals and expansion. Turn one-off high-impact use cases into durable workflows that repeat across teams.

That list says what kind of job this is. The material is not a feature list, it is the customer’s working procedure. Which work stays with people, which work moves to the tool, and what evidence will show whether the move paid off, all decided with the customer’s own organization. More hours go into building agreement inside someone else’s company than into writing anything. The seat is a consultant’s, housed inside a vendor, and it differs from consulting in that the result is confirmed by a renewal rather than a deck.

Time spent inside a customer counts for a lot

Three paths lead in. The first runs from customer success or solutions consulting in enterprise SaaS, where carrying a book of accounts and a renewal number is already the job. The second runs from a consulting firm, where process redesign and executive persuasion are habits and only the product knowledge is new. The third runs from the buying side: someone who drove internal rollout at a company that adopted the tool and knows in their body where adoption stalls.

What all three need is hands-on feel for the product, the ability to argue in numbers in front of executives, and the patience to move another organization off its current habits. Technical background helps, but implementation belongs elsewhere, so reading code is enough and writing it is not the point.

Plenty is still unsettled. Titles differ enough that nobody knows what to put on a resume yet, and each company measures the outcome differently. Compensation structure is another open question, since the role sits between a quota-carrying seat and a technical one.

The cheapest thing to check right now is the boards themselves. Open the postings that carry Outcomes, Deployment or Enablement and read only the success metrics. Whether the section repeats retention, adoption and expansion, or lists technical deliverables, tells you which of the two branches that seat belongs to.