Invoca

Invoca just wrapped an AI Passion Week: five days, 379 agents built by 100+ non-engineering teams across Sales, Marketing, CS, Finance, and Ops. Our finance team Aimee Johnson, Geneve Dromensk, Michael Ganesh, and Fermin C. created “Cash-GPT," a forecast engine agent that pulls from multiple data sources to improve forecasting and is already saving us hours per week. Every hour of internal efficiency we recover is an hour reinvested into the Invoca platform for our customers. This is what AI ROI looks like. The companies that win won't be the ones that bought the most AI software. They are companies that build real capabilities. #leadership #AI #ROI #CFO #AIPassionWeek

Marketers love to talk about AI Agents. My team spent a full week building them. As part of Invoca's AI Passion Week, our marketing team joined 100+ Invocans to spend five days building real AI workflows to help automate more of our day-to-day work. Here are some of the wins that came out of our department: Suzzette Giron & Natalia O. Burris Burris built an end-to-end display ad campaign builder that writes copy, creates layer names, and generates UTMs. This cut campaign creation from 8 hours to just 1. Mary Herkins created an agent that researches a prospect's website and industry, then generates a personalized pitch and script for SDRs and AMs. Every rep can now walk into their meetings a little more prepared than they were before. Carén B. built an agent that monitors closed-lost accounts in Salesforce and automatically alerts sales when a stalled account shows signs of life.  Sandy Pell created a deck builder that auto-converts off-brand presentations into on-brand ones.  Turns out when you give marketers the right tools and some protected time to build, they don't just talk about AI transformation, but they actually ship it. Big thank you to everyone on the Invoca team who participated!  #leadership #AI #AIPassionWeek #Marketing #DemandGeneration 

Automate the automatable. Preserve creativity for taste-making moments that fall outside the brand rules. Somewhere right now, your brand deck is going out with the wrong font, wrong hex colour code, and a logo stretched in ways nobody wants to describe. I can feel it. And so can you. So for our company-wide AI Passion Week, I finally did something about it. Here’s the thing about brand consistency: Designers fix it. Then it happens again. Fix. Repeat. Forever. AI should protect creative people’s time for the work only they can do. That should never be fully automated. Let design shine in the moments that matter. From May 4-8, 2026, non-R&D Invoca staff spent five days building real AI workflows to solve real problems. 100+ people. 379 new agents. 64,453 tool calls. 958,875 lines of code. Yes, those are real numbers. My contribution was a small one: a new shareable Claude skill that encodes our entire visual identity system directly into AI behaviour. Drop in an off-brand deck > Get an on-brand one back. No more heavy lifting. Free designers for the judgment calls only they can make. Honoured to have been selected as runner-up for best individual workflow for the creation of a new Claude Organization skill: / invoca-brand-guidelines I’m not a builder. I’m just a brand person who got tired of seeing the wrong hex colour code… Huge thank you to Matt Diederichs, Ali A., and the team for making this week happen. Which now has me curious... #brand and #marketing leaders, how are you tackling brand consistency at scale? #AI #innovation #AIPassionWeek

You can't just hire AI fluency. You have to spend time nurturing it, too. I've been thinking about this a lot since AI Passion Week wrapped at Invoca. We didn't mandate participation. We gave people protected time and the right tools, then got out of the way. 100+ Invoca employees across 7 departments participated. What stood out to me wasn't just the output (though 379 agents in five days is awesome), it was how people learned: → 61 support threads from Invocans who hit blockers and kept going anyway → Numerous live coaching sessions where people taught each other → 25+ teams forming across departments and functions Real AI fluency develops through tackling problems head-on, struggling a little, and teaching the person next to you what you just figured out.  Other People leaders, how are you approaching AI in your orgs? #leadership #AI #TalentDevelopment #FutureOfWork #AIPassionWeek

958,875 lines of code were written at Invoca in one week, and not one of them came from our engineering team. We intentionally kept our recent AI Passion week non-R&D only. Sales, Marketing, CS, Finance, Ops, Talent, IT. There were no engineers in the room, just the people who live in the workflows every day, given the same building power that engineers have had for years. Here's what happened: → 379 new AI agents built in five days → 64,453 tool calls across Sheets, Slack, Salesforce, Gmail, Gong → Tasks that previously took days, finished in minutes The best ideas for what we should build aren't always inside the Invoca product org. The leaders in the AI age will be the ones who apply AI across all departments, not just engineering.  #leadership #AI #Product #Innovation #AIPassionWeek

Invoca's AI Passion Week Readout: GTM Team Coding! The GTM team spent a week building AI agents while they continued to execute on their "day jobs". Here's why that's the best ROI decision we made this quarter. GTM orgs aren't usually who you picture when you hear "AI hackathon." We support our customers, close deals & progress pipeline, but we don't often see the entire team "code" with AI. However, last week we did. As part of Invoca's AI Passion Week, our GTM team joined 100+ Invocans across 7 departments to create real AI Agents & workflows. Below are some examples: Kyleigh O., Caitlin Ristau (Ferrara) Ristau, and Jojo Cush built a shared AI brain for account managers that pulls customer data and analyzes it at scale. Robin Shrestha built an agent that analyzes unanswered calls to surface callback rates, attrition, and time-to-retry insights, helping partners identify lost leads and take action. Brian Coulombe created an agent that reduces keyword spotting signal build time from over a week to under 20 minutes using Claude and customer transcripts. Kyle Paklaian and Brandon McCarty built a secret shopper that measures how companies respond to inbound leads across channels, benchmarking speed-to-lead. This intel can then be used by our sales team during outreach to show pain points and areas for improvement.   Christopher C., Ian Hamilton, Erick Bright, and Charlie Conner created an AI agent that cuts platform summary creation from 4-8 hours down to 20 minutes by automatically pulling and formatting Invoca data for CSM reviews. When GTM teams deeply understand what AI can do firsthand they become a completely different kind of partner for our customers. Invoca customers rely on our AI to drive revenue. It matters that our GTM team got their hands dirty building with it too. Sandy Pell Derek Andersen

When we started planning Invoca's AI Passion Week, I genuinely didn't know if anyone would show up 🫠 I kinda thought I might have to personally go find "volunteers." Would people sign up on their own? Would they actually build something real, or just go through the motions? Yeah... nah. About HALF of our non-R&D employees signed up to build things. Like, a LOT of things. In less than a week: → 379 new AI agents → 958,875 lines of code generated by people who don't write code → 64,453 live tool calls into Salesforce, Slack, Google Workspace, Gong The results are in the slides. But the part I actually want other operators to steal is how we set people up to succeed: → Help people think, not just build. We gave everyone a custom Claude Skill to guide them through design thinking, product scoping, and agent architecture before they wrote a single workflow. Most people had never thought about AI this way, this helped give them confidence to start. → Protecting the time. Monday design session, Tuesday and Thursday fully blocked for building. Managers clearing meetings for their teams to build. We did training & enablement on our toolset for weeks prior (shoutout Gumloop Luke Ierullo Max Brodeur-Urbas for being amazing partners in this project). → Enablement and support. Ali A. & our AI Ops team didn't just sit on the sidelines. They spent the entire week in the trenches with people; unblocking, coaching, building alongside them. That presence made the difference between people giving up & people shipping. Overall I was absolutely blown away by the creativity & engagement of the Invoca team. It's amazing what you can accomplish when you give people the space & tools to get going. #AITransformation #Leadership #AI

Our customers don’t pay us to be busy. They pay us to keep raising the bar on what’s possible with AI. That only happens when innovation is not locked inside one team. It has to become how the whole company operates. Our recent AI Passion Week illustrated this point. In five days, 100+ non R&D  Invoca employees across Sales, Marketing, Customer Success, Finance, Operations, Talent, and IT stepped away from the day-to-day to build real AI workflows. Here’s what they shipped: → 379 new AI agents built → 64,453 tool calls into the systems we use every day → Tasks that took 8 hours now take 1. → Workflows that took a week now take 20 minutes. → Projects already running in production by Friday Every one of those numbers turns into something our customers feel. Faster answers. Deeper insights. Better execution. A partner that is continuously compounding its AI capability. AI adoption cannot be a side project or a center-of-excellence exercise forever. At some point, it has to move from the lab into the operating rhythm of the company. That’s how teams get faster. That’s how customers get more value. That’s how companies keep raising the bar. Curious how other leaders outside of Invoca are approaching this: how are you helping AI move from experimentation to everyday execution? #AIPassionWeek #leadership #AI #innovation

Thank you to Tim Finnigan, MBA for helping me accomplish a key career objective Poolside podcast! 🌴🌴💧💧 🤪🤪