Welcome back to Juniply Field Notes.

The more AI tools I use, the more I notice the same thing.

The outreach is faster. The tickets write themselves. The data is everywhere.

And the thing that still makes you stand out is a real conversation.

As AI gets more complex, humans become more human.

For the first 10 years of my career, no one invited me golfing.

My dad's generation sold on the golf course. My generation sold on Zoom. Online calls were quicker, and everyone was excited about them.

I told my parents I never needed to learn to play golf. For sales, it was a dying sport.

Now it has done a full 180 on me.

I'm getting invited to pickleball. To ICP dinners. And yes, to golf.

When I recorded this week's podcast with Maxwell Sherwood, GTM lead at Amplifier Health, we kept coming back to the same idea: as we use more and more AI tools, in-person and human connection is what makes you stand out. Maxwell's team is putting more time into conferences and face-to-face conversations for exactly that reason.

I've seen why firsthand. I once asked a team running 100,000 emails a month what their conversion rate was.

A third of a percent.

That isn't outreach. That's burning bridges.

When everyone can send a million messages, the message stops being the advantage. Showing up does.

We're going back to the traditional sales rep. The difference is that AI now handles the busywork around them.

HubSpot has started selling a new seats-and-credits pricing model to new EMEA customers. Email sends, workflows and AI agents all draw from a monthly credit pool.

AI metered work is the future. I'm seeing two shifts in how we pay for AI:

Pay per use is the more traditional method. But like anything else in life, it reminds me of the early days of playing video games: every action has a dollar amount.

Economically, this will be great. But I can see smaller teams not enjoying it, because every click costs something. In large organizations where a separate team handles finance, it's less of an issue.

That's my personal pain point with pay per use, and I expect to see more of it as costs rise.

The upside: the barrier to entry gets very low when you only pay for what you use.

SaaStr says one vendor quoted up to $240,000 a year to let its AI agent keep accessing its data. The agent's suggestion was to copy the data into a $5 database instead.

With MCPs and open APIs, everything is integrable. That makes data residency and ownership a much bigger challenge.

We're already seeing the pain, and the cost, of trying to keep these bills down.

What I think gets interesting is paying for results.

Where is the money actually going?

A better model doesn't mean better results. And results, at the end of the day, are what matter most.

With the market moving this fast, multi-year deals aren't as popular. Buyers don't want to lock in when the market shifts so much.

According to ICONIQ data highlighted by SaaStr, sub-one-year deals went from 4% of new logos in 2023 to 13% in 2026.

It's interesting to see that, instead of chasing the five-year deal, landing the one-year deal (rather than month to month) is now the big win.

This week SaaStr shared that an AI coding agent deleted a core part of its app, twice in under an hour, and then said it hadn't made the edit.

As more and more people become developers, methodology and best practices matter more than ever.

When I ask non-technical founders whether they have a staging and dev environment, many of them say:

"What do you mean? Everything goes to GitHub."

Learning how to set up a proper development life cycle is more important now, not less.

As things move faster, you need that redundancy.

More people are moving into product. And a lot of product meetings get extremely boring.

Sales says their ticket matters most. Marketing says theirs does. Customer success says theirs does.

So I built Roadmap Rumble.

Every ticket gets personified as a character, sized by how important it's expected to be. Click a character and you see the details: how many deals it's connected to, how often the issue comes up, linked Linear tickets, and three KPIs: calls, lost deals and dollars at stake.

Then the room votes, and the characters fight. Winners move on to the next round until you have a final ranking.

Now that everyone is automating tickets and creating tickets, you still need time to sit down and decide what matters most.

Ranking against a random rubric lets things fall through the cracks.

Asking questions and voting will matter more as we make work more autonomous. There comes a time to get back to in-person, real discussions.

One caveat: some big pain points aren't deep, just broad. Keep some oversight on the final ranking.

Just because it's work doesn't mean it can't be fun.

Watch the Roadmap Rumble demo

Explore the public repository: Explore the public repository

I sat down with Maxwell Sherwood, GTM lead at Amplifier Health, a voice AI company that reads health signals from how people speak.

One of his best tips is how he opens LinkedIn messages: "Hey Brendan, I know cold outreach is the worst, but..."

In this AI world, it's more important than ever to be truthful and get back to the idea of sales.

They know it's a sales call. Tell them it's a pitch. Call it out up front.

When everything else looks perfect, the human flavor matters more than ever.

We also talked about two things I keep coming back to:

Watch the full conversation

The build challenge is evolving.

I'm launching something new soon, built around what I love to do: running hackathons, building cool apps, and getting the community involved in learning all these new tools.

You'll be able to follow along, and you'll be able to compete too. There will be some nice swag.

More soon.

Have you built a useful workflow, agent, prompt system, or strange AI experiment?

Send it to me.

It doesn't need to be polished. It does need to solve a recognizable problem or teach us something useful.

Future editions will feature one community build, including how it works, who should try it, its limitations, and what its creator learned.

My new tool of the week is Buzz, the open-source workspace from Jack Dorsey's Block where people and AI agents chat side by side.

It will be interesting to watch. As AI gets better, more of the revenue comes from agent costs, while more and more tools become open source.

That means you can build off them to fit your exact needs.

I'm testing tools, building strange ideas, and learning what works in public.

Explore Juniply

Follow the 52 Weeks of Building challenge on YouTube

If you know someone who is also trying to understand where modern go-to-market is headed, forward them this edition.

As AI gets more complex, humans become more human.

Let's learn new things and do new things. Most importantly, let's experiment and have fun, because no one knows where the future is going.

The only thing you can do is bring your creativity to the world.

Until next time