ChatGPT Work Cuts Campaign Email Production From Two Weeks to Two Days
OpenAI’s marketing team argues that ChatGPT Work can compress campaign-email production by bringing briefs, brand guidance, reusable product use cases and Figma templates into a shared workflow. Rather than replacing marketing judgment, the team says the setup produces on-brand drafts and finished template work more quickly, shifting review from coordinating handoffs to decisions on messaging and creative. In the demonstration, a process previously said to take roughly two weeks was reduced to about two days.
Campaign production shifts from handoffs to review
A marketing workflow that once required a written brief, a kickoff with lifecycle and creative teams, rounds of draft review, and a final handoff was described as taking roughly two weeks end to end. In the setup shown here, that process was compressed to about two days.
The change is not simply that ChatGPT Work drafts email copy. The workflow brings together the materials that normally sit across separate stages of production: a brief, brand guidance, a library of use cases, and Figma email templates. That shifts team meetings away from “the work around the work”—coordinating inputs, interpreting briefs, and passing drafts between functions—and toward decisions on messaging, copy, and visuals.
The Figma comparison made the intended difference concrete. An earlier set of templates appeared as gray wireframes with placeholder text and image boxes. The newer campaign set contained completed visual treatments, graphics, calls to action, and copy including “Need to get something done? Ask ChatGPT Work to help” and “Connect the tools you use every day, keep work moving in ChatGPT Work.”
With ChatGPT Work connected to Figma, the team created a more dynamic series of templates that it said could be automated end to end.
The workflow still leaves marketing judgment with the team. Its purpose is to get to a tangible, on-brand starting point faster, so review can center on whether the campaign says the right thing and looks right.
A prompt spreadsheet becomes a usable campaign library
The workflow begins upstream with reusable source material rather than a one-off campaign request. The team maintained a spreadsheet titled “ChatGPT Work use case & prompt list,” organized around fields including feature or capability, prompt type, in-product prompt, and deep link. Visible examples included “Build a spreadsheet to compare debt payoff plans” and “Pull together my performance review packet.”
ChatGPT was asked to turn that spreadsheet into an interactive site: a dashboard through which teams could find use cases to feature in marketing. The resulting interface, headed “Let ChatGPT Work do it,” lets a marketer choose a capability—such as Sites, Artifacts, Scheduled Tasks, Mobile, Model Intelligence, or web-based tasks—and then narrow by life situation or audience.
In the example shown, selecting scheduled tasks and knowledge workers surfaced tiles derived from the spreadsheet. A marketer could click a prompt, copy it, and use it in campaign work. The spreadsheet remains the underlying library; the dashboard is the browsing layer that makes relevant examples easier to locate than a long list of prompts.
The site was described as having been created in a couple of minutes. Its role is not to generate a campaign strategy on its own, but to make an existing collection of product examples available at the point a marketer needs to choose a feature, audience, or scenario.
Reusable prompts can support both campaign selection and product discovery
The same use-case material was also adapted into a public-facing discovery surface. The team scanned Instagram for Reels mentioning ChatGPT Work and assembled them into a social-media-style application, where users can scroll through examples and see the relevant prompt alongside each video.
One screen pairs the heading “Find your next job before everyone else” with a request to build a personal job-tracker dashboard on Sites and continually scan Indeed, job boards, and company career pages for newly posted roles. A “Try in ChatGPT” button lets the user act on the prompt.
For campaign production, the significance is that a reusable prompt library can serve more than one audience. Internally, it helps marketers select use cases worth featuring. Externally, it turns those same use cases into examples that prospective users can browse and try. The team said the public application was created using Sites.
A shared email plugin makes the workflow repeatable
For email creation, the Figma plugin turns the workflow into a shared package of skills. Its panel, titled “OpenAI Email Engine – Example,” describes its purpose as building “governed, on-brand emails from the approved Figma EDS.” The listed skills include Brand Guidelines, Brief Strategy, Campaign Orchestrator, Campaign Research, Email Love Production, Email Marketing, Figma Knak Handoff, and Knak Email Automation.
The package contains about a dozen skills that break email marketing into distinct tasks: reading a brief, adhering to brand guidelines, connecting to Figma, and using the organization’s templates. Once installed and shared, it gives anyone on the team access to the same materials and constraints for writing campaign emails.
The demonstrated request was specific: “Create 5 connector emails for knowledge workers using the content we have in our use case library and tell me when they are in Figma.” ChatGPT Work would then use the connected files for a couple of minutes, with particular connections available to specify directly.
The model is governed reuse rather than generic prompting. Instead of beginning with a blank text box and hoping the output matches the brand, a marketer invokes a shared system of guidelines, campaign strategy, templates, and use cases. The resulting emails land in Figma, where the team can assess and refine the creative work.