Jeremy (@JHurst12) header
Jeremy (@JHurst12) profile picture

Trying my best to be mindful & intentional

Joined June 2009
Photo by Jeremy (@JHurst12)Photo by Jeremy (@JHurst12)Photo by Jeremy (@JHurst12)Photo by Jeremy (@JHurst12)Photo by Jeremy (@JHurst12)Photo by Jeremy (@JHurst12)Photo by Jeremy (@JHurst12)Photo by Jeremy (@JHurst12)Photo by Jeremy (@JHurst12)Photo by Jeremy (@JHurst12)
As I got more senior, I kept hearing "work ON the business, not IN it". AI has turned that up to 100. But every time I stop to harden the system, I'm not losing a day. I'm buying back every day after it.
8
In a post-AI world, we have all the context we could ever ask for. Call recorders catch every customer interaction. Relying on the rep as your primary input mechanism no longer makes sense.
11
Every time I demo Swan AI, I'm demoing a second AI product at the same time. It's Wispr Flow - an AI voice dictation app. It changes how I think when I'm working with AI. I describe what's actually in my head, even when it's not fully formed yet.
1
36
It's almost always better to double or triple down on what's working than to go try something new. For us, the 20% was LinkedIn. Outbound just had to show up next to it.
11
We scaled Datorama $0-60M ARR in 5 years. Salesforce acquired us in 2018. What I learned on the inside: their own forecast doesn't run on Salesforce. So the forecast lived in a Google Sheet. And the humans on the call filled in the missing context.
8
v1 of my follow-up system: HubSpot tasks. Every follow-up became a task, on top of the tasks our call recorder created after every meeting. Within weeks I had a wall of overdue tasks I wasn't getting to. The CRM turned into a guilt list.
2
1
21
7 months ago I took my biggest career bet. I walked away from a $450K VP Sales role to take a job with a made-up title: “VP of Autonomous GTM” at Swan AI. I built a fully autonomous one-person GTM org. I hit $1M+ of pipeline a month with just me & AI agents.
2
21
Most folks in GTM think you want more pipeline coverage, not less. But Manny Medina (Outreach co-founder) told me the best AI-native companies he speaks with average 1.7x pipeline coverage. Historically, we've always been told 3-5x.
1
18
Two thirds of our MQL alerts. 19% of pipeline. Deals 65% larger than average. All from signals we were already getting. The best signal-based outbound I've built didn't start with a new signal. At Swan AI, our best organic inbound source is LinkedIn.
1
12
Every new sales rep gets a ramp. I gave the agent the same grace. I won't ever know where the boundary is if I don't push the limit.
17
My follow-up system went through 5 versions this year. The first four each broke in a new way. v5: My own follow-up app. Under 10 minutes to clear the day. Turns out, the busy work was never the writing. It was everything around it.
10
So I had to treat the AI like a 4 year old - to ask questions I'd never asked software before. What failed? Why? How do we prevent this next time? Maybe we break it into smaller chunks? Weird questions to ask a machine. But the machine had answers.
16
"Very unprofessional Jeremy." A CEO emailed me that after I cancelled our meeting. He ended with "Everyone's time is valuable." He was right. It's also why I still cancel meetings. I don't need more meetings & opportunities. I need fewer, higher-quality ones.
17
We're generating $1M+ of pipeline per month with 1 GTM person (me) + AI agents. Gold leads went fully autonomous. The outreach agent gets it right 70-80% of the time. If 70% quality means processing 2 or 3x more leads, the net is massively positive.
9
AI isn't a 1 shot technology. It doesn't just work. The people on the other side of this revolution aren't the ones AI worked for on the first try. They're the ones who kept going when it didn't. That's the rep. That's the work.
6
First week at one of the hottest AI GTM startups. Full pipeline. No team. What was my first move? Turn the AI off.
1
14
Turns out, you can't automate judgment you haven't developed yet. Automation is a force multiplier. It multiplies whatever you feed it. Including the broken stuff. Automate last, not first.
1
1
The irony: the best AI outbound I've ever seen was built by doing zero AI outbound first.
2