We're excited to announce the research API for AI agents, with marketing, product, and insights teams at 60+ enterprises already using Gather.
How do companies like Datadog and Fortinet use Gather?
- Testing ads before they launch with simulated personas of their buyers
- Understanding where their competitors are weak, and using that in competitive deals
- Running real research to create thought leadership reports to drive top of funnel
GatherHQ makes customer calls, surveys, and past research queryable. When evidence is missing, its AI researcher designs studies and interviews real buyers, delivering quantitative and qualitative insights in days. Every interview strengthens a living buyer model that teams and agents can query to test messaging, pricing, and product ideas in minutes.
With 250,000+ interviews conducted and customers including Datadog, Fortinet, Envoy, and Staples, GatherHQ is already helping teams sharpen positioning, test campaigns, prioritize roadmaps, and create original research-backed content. Agents access the same capabilities through an API, making research a continuous part of their workflow. An API for how buyers think.
Today we're launching Gather's Customer Simulations. It moves GTM teams away from guessing what customers want and instantly simulates what they will say and do.
We’ve grown 10X in eight months and are lucky to be learning from dozens of incredible customers who are moving faster than ever before. Before you spend another dollar scaling an assumption, bring us the question behind it.
Simulate what’s next. Built on what’s real.
https://t.co/ESVTMXx6PH
engineering leaders in 2026:
trust in AI-generated code: up
production incidents caused by AI-generated code: 61% in the last 90 days
both true at the same time
@lightsilver323 Marketing teams guessing what their buyers want, because actually asking used to take six weeks and a five-figure invoice. We make it days.
https://t.co/ATodw3eWm6
@codyschneider Agreed on all of it. One addition: outbound and ads test the offers you already thought of. They can't surface the one nobody guessed. That's the difference between measuring appetite and discovering it.
Impressions are easy to screenshot. "We asked 200 buyers and they still can't tell us apart from our competitor" is not.
The second one is worth more and gets shown less.
We built Gather for the second kind of number.
https://t.co/ATodw3eWm6
a lot of founders are not buying marketing.
they’re buying screenshots they can send to investors.
and many marketing agencies are happy to sell exactly that.
impressions, trending charts and creator posts look great until someone asks what actually changed after the campaign.
same survey, a few questions apart:
"is your data governance ready for AI agents?" → 76% said yes
"what's your biggest obstacle to AI agents?" → data governance
we contain multitudes
@ItsAlexhere0 Neither wins. Both are guessing. The mediocre product doesn't know why people leave, the quiet one doesn't know why nobody arrives. Same missing information, different symptom.
Correct. Now do it for 600 buyers instead of the ten loudest ones on Reddit.
That's Gather. AI-led interviews with people who actually have the problem, then the content that answers it. Days, not quarters.
https://t.co/ATodw3eWm6
nobody cares about your product.
they care about their problem.
go where they're already complaining and reply with something useful. that's the entire marketing plan.
@victor_bigfield The complaints are the research. Most teams read a hundred of them, absorb none, then build messaging around what they wish customers cared about instead.
@theandreboso The old playbook let you skip knowing your buyer. Keyword volume did that work. Brand doesn't. Mentions, reviews, people repeating you, all of it requires knowing what's actually worth saying. The optional skill just became mandatory
@solopribuilds Plenty are. The harder question is what it should actually do. Most marketing SaaS gives you another place to put work in. The thing people would pay for takes work out. That's the bet we made with Gather!
@pmitu Autopilot works right up until the map is out of date. Execution automates fine now. Knowing what your buyers actually want this quarter is the part that keeps expiring
@DaveSchoools@rikin311 Fair. Product's hardest because the feedback arrives last. Engineers get a failing test in seconds, marketers get CTR by Friday. Product makes a call and waits two quarters. The only real fix is asking buyers before you ship, not after
@NickB2005 Report four, if the first three come back clean: ask the leads themselves. The CRM tells you what happened. It doesn't tell you why the ones who did answer said no. That's the one dataset that was never in the CRM to begin with!
Most teams talk to their customers once, then spend six months guessing what they'd say about everything else.
Grounded Personas fix that. Run a study on Gather and you get queryable archetypes built from your real interview data. Ask anytime.
https://t.co/ATodw3eWm6
@rikin311 Careful. The same logic that made engineering cheap is coming for marketing execution next. What stays expensive is knowing what customers actually want
@MicroLaunchHQ Most research tools stop at insights. Gather keeps going. Real buyer interviews in days, then the reports, posts, and decks that come out of them. One study, a month+ of content