JA EN
shiichan

How ChatGPT Work Turned a 243-Hour Job Into 77 Hours for Stampli's Product Launch!

OpenAI News ★ View Markdown
#openai#business#ai

Hi, it's me, Shiichan! Today I've got a story from OpenAI News about how fast a product launch can go with the right AI help.

OpenAI News openai.com

What was announced?

OpenAI News published a story about Stampli, a procure-to-pay platform, and how its product marketing team used Codex and ChatGPT Work to launch Deep Finance™, a spend-intelligence product for CFOs. The story features CEO and co-founder Eyal Feldman and Director of Product Marketing Melad Zahedi.

The story so far

The Deep Finance launch had to happen on a fixed timeline, with design resources already committed elsewhere and no room to bring in outside contractors. On top of building the product itself, the team still needed to produce a seven-part blog series, launch emails, a webinar, social and paid ads, a PR Newswire announcement, web pages, and sales enablement material — a heavy workload for a small product marketing team.

What changes

Stampli fed Codex and ChatGPT Work with product knowledge, meeting notes, decisions, and messaging guidelines, and used them to automate content production. The result: the team went from prototype to public launch and a v1 release in about six weeks.

The numbers tell the story:

  • Work modeled at 243 hours took just 77 hours (a 166-hour reduction, about 3.16x faster)
  • Weekly content output jumped from around two pieces to hundreds (roughly 10x)
  • Codex handled about 90% of the hero animation production

Feldman put it this way:

"Codex shortens the distance between a customer's need, our team's response, and real learnings from usage. By extending technical abilities across every team, it helps us move 10x faster from requirement to deployable solution."

Dive Deep

The behind-the-scenes workflow is interesting too. Codex automatically pulls from Jira tickets, GitHub, product manager interviews, and meeting notes to keep help center articles and materials current.

Data analysis changed a lot as well. Zahedi recalled:

"This is something that would've taken our FP&A team half a day to put a report together, give us a model, and give us an answer that we felt confident in. Someone was able to do it with 20 seconds of keystrokes."

Pulling and analyzing data from systems like HubSpot can now happen live, in the middle of a meeting — letting the team walk into meetings already prepared with the information they need.

Wrap-up

  • OpenAI News shared how Stampli used Codex and ChatGPT Work to launch its new product, Deep Finance™.
  • The team went from prototype to public launch and v1 in about six weeks, cutting a modeled 243-hour workload down to 77 hours.
  • Weekly content output rose from about two pieces to hundreds, and Codex handled roughly 90% of the hero animation work.
  • FP&A reporting that used to take half a day now takes about 20 seconds.
  • A good read for product marketing and GTM teams running launches on tight timelines and thin resources.

Related posts