13 tactics for company-wide AI adoption

You’re reading Release Notes, a weekly newsletter covering product lessons from scaling beehiiv and building AI products in the creator economy. (Read the full archive here.)

If you're new here, I'm Jake, a product expert in the creator economy, currently leading new product initiatives at beehiiv.

A few weeks ago, Tyler (beehiiv’s CEO) claude-pilled himself over the weekend, and sent an all-hands email that kick-started one of the fastest internal transitions and tech adoption in the company’s history.

The email in question addressed one of the main concerns for founders and leadership teams right now: the need for AI adoption.

Two months on, we have >90% daily adoption across the business.

And by adoption I don’t mean “uses AI chat for Q&A, basic copywriting or idea generation”.

Adoption at beehiiv means creating AI workflows, skills and automations that are saving employees hours every day. i.e. using AI to create genuine operational leverage and significantly increase output without increasing headcount.

How did we get there in just two months? We followed some simple steps that you can replicate.

This call-to-arms also sent me on a mission to find out what we could do better: I spoke with some of the teams at the most AI-forward startups like Zapier, Linear, Reforge & Equals to find out exactly what worked.

This newsletter covers the 13 most effective tactics I encountered for driving adoption.

So if you’re a founder / startup leader and you’re thinking about these questions

Then this is for you.

[sponsored]

Most AI creative tools fall short for one simple reason. You can generate tons of ads, but they aren’t up to par.

Refining copy, adjusting layouts, or nudging a CTA into place shouldn’t require rewriting prompts over and over. It slows teams down and breaks the creative process.

With Hightouch Ad Studio, AI gets you 90% of the way there. For the final 10%, use a built-in editor to quickly refine copy and design, or export directly to Figma for seamless collaboration with your design team.

Move faster without losing control. Every ad, exactly how you want it.

See how Ad Studio works

1. Issue a "code red"

When GPT-4 launched, Zapier CEO Wade Foster called a company-wide halt. No normal work for a week. A playbook published. A full-company hackathon. Adoption went from under 10% to over 50% in seven days.

While an actual code red isn’t necessary, what IS required is a top-down directive from the CEO that this is a #1 priority starting today and explaining what needs to happen. This was Tyler’s Sunday email for beehiiv.

Note: This does not work at the individual team level, it must come from the top.

2. Make it clear why adoption matters

Most AI mandates fail because they're framed as exciting ‘extra work’ rather than necessary and existential to the business. They also typically lack the context on ‘why it matters’.

Tyler's note worked because it was contextualised with our real revenue, burn and hiring numbers, plus his vision for the company and the role that operational leverage (via AI adoption) played in getting us there.

Don’t underestimate/skip this step.

3. Define what "good" looks like (by role) and be clear on your expectations

Zapier is arguably the best at this (although this was referenced in every conversation I had with other companies).

Zapier maps AI fluency across four levels for every function in the company: Unacceptable, Capable, Adoptive, and Transformative. Critically, what "adoptive" means for an engineer (chaining LLM calls, adding eval tests, using Claude Code) is completely different to what it means for someone in marketing (A/B testing AI copy, optimising prompt libraries). Give people a target they can actually aim for.

In beehiiv’s case, Tyler spelled out his expectations team-by-team, making it clear that it’s not about being able to do 2 tickets instead of 1, it’s about seeking opportunities for real leverage and trying to figure out how you can go from 1 → 10.

4. Lead by example

Similarly to the code-red only working if it comes from the CEO, company-wide adoption only works if the CEO, leadership team and management are all leading from the front and pushing their teams.

At beehiiv, Tyler and the rest of the leadership team are some of the most active users and (importantly) some of the most frequent contributors to the “AI show and tell" channel (check out #11).

At Whoop, when a PM brings the Head of Product a problem, she says: "Want me to show you how I solve this with AI?" Then she shares her screen. Things like this, or responding to questions with “have you tried using AI yet to answer/solve that”, are easy ways to drive top-down momentum.

5. Remove cost constraints as a barrier for adoption

Don’t underestimate how much being cost-conscious with AI tooling can undermine your adoption efforts.

Why? The cheapest and lowest tiers typically have the worst models, the slowest responses and the lowest daily/hourly limits. The limits are what kill momentum and adoption, but the worse models also prevent your team reaching that ‘lightbulb’ moment when it gets something complex right.

Whether you go the Reforge route of unlimited spend for employees on AI tools (whilst using good judgement), or the beehiiv route of setting up an enterprise workspace that gives employees effectively limit-free access to the latest models, this is a non-negotiable.

Limiting spend is the biggest own-goal you can do during this process.

Note: If you are offering an unlimited budget but everything requires approval, then this isn’t going to work well.

If you need predictability in your spend, go the enterprise route or give a fixed (but generous) budget. Eg Duolingo gave each employee $300/month to experiment with tooling at first.

6. Give everyone dedicated, protected time.

"Not enough time" is consistently the #1 reason people don't try new tools or make enough progress to add it to their workflow. Zapier replaced three all-hands meetings with an async one-hour work block for learning and building.

Meeting-free focus days (beehiiv has two/week) are a great way of facilitating this too.

7. Assign a procurement owner and fast-track approvals (or centralize your AI Ops)

At Zapier, they assigned a PM to own the process for vetting and approving different AI tools - this included working with tech and legal to make sure everything was safe to introduce into the ecosystem.

At beehiiv we took a different approach, centralizing AI operations within a Claude enterprise account, but then having an engineer to approve the 3rd party app connections into the ecosystem. One advantage of this? If a connection or MCP was missing, the engineer could just make something custom for the team.

8. Run a company-wide hackathon (for every function, not just engineering)

This is the single highest-leverage action you can take. Zapier's first hackathon had 360 people, 150 ideas, and 60+ submissions, with people from support to HR to marketing.

Zapier's marketing team ran their own version: everyone submitted a screen recording of something they'd built in Cursor, Codex, or Claude to automate a repetitive task, with cash rewards for the best. The message: get hands on keyboards, build something real, learn together. Fear of AI fades the moment people actually use it.

9. Shared Workflows, Connections and Skills

One of the biggest impacts of company-wide adoption is the compounding effect from being able to leverage the work done by others.

This can be simple things like when one person adds Linear to Claude, everyone gets access to linear, or more complex setups with sharing skills and plugins.

eg. At Equals an engineer built a skill that flagged if any new ticket was a duplicate, that could be used across the team.

At beehiiv, managers work with their team to develop skills for their specific function, and then share these as ‘official team skills’ that are available to the entire team and company.

Note: While this is much easier within a team account on Claude / ChatGPT, this can be achieved by a centralized GitHub repo or skills shared into a Slack channel. The more hacky the solution → the worse the discoverability → the lower the compounding effect.

10. Optimize for workflow

The more you can reduce friction and expose your team to helpful workflows, the more you will get adoption. Bringing tools into Slack/Teams is one of the easiest ways to do this.

Cristina Cordova (Linear COO) describes this dynamic perfectly: “We tend to think of AI adoption less as a training problem and more as a workflow design problem. If AI is fast, useful and embedded in everyday tasks, people naturally start reaching for it.”

eg. Equals hooked up their own internal tool Analyst to Slack so anyone can ask detailed questions about pipeline, deal status or revenue directly in a thread.

11. Create a dedicated show-and-tell channel.

Arguably the easiest thing on this list to put into action: Just a Slack channel where people drop things they've built, time they've saved, workflows that surprised them.

Not only does this drive home the impression that everyone is participating (which drives more participation and engagement) but it completely opens your eyes to new ways of solving problems as your colleagues use their expertise to approach the same challenge in a completely novel way.

12. Select your “AI champions” for each team

This is essential for kick-starting the flywheel on contributions in the show-and-tell channel. Select one person per team, and task them with sharing one team-focused use case into the channel each week. This will be the necessary spark to ensure that people feel the need to get involved.

13. AI office hours

To maximise adoption across the entire company you need to make sure you have processes in place to help people get started, get unblocked and ask the ‘stupid questions’.

AI office hours were particularly effective for adoption at beehiiv: it was a 60-minute session led by engineers/PMs who would do basic demos, show people how to get their account set up, unblock billing/access issues, explain terminology, set up connections etc.

Having this ensures that there are no barriers to entry for any employee.

None of these are hard to do. It’s just a question of focus and prioritization.

Let me know what I missed or if there was anything you tried that was really effective.

Cheers,
Jake

PS. If you’re considering launching your own newsletter, you can get a free month & 20% off by clicking here.