1. Stop copying systems built for companies 100x your size

When People teams build a new program, the instinct is often to look at what Google, Meta, or another massive company is doing. But a process designed for hundreds of thousands of employees probably isn’t the right answer for a company with a few hundred.

Instead of asking, “What is best practice?” ask:

What does the company need today?

Will employees and managers find this useful?

Will this make their work easier or more effective?

“What was Google like when they were 300 people? Were they doing what they’re doing today? Absolutely not.”

Large-company systems can still provide inspiration, but they should not become the default playbook. Start with the problem in front of you, build the lightest solution that addresses it, and add complexity only when the company truly needs it.

2. Pick one or two things that will have an outsized impact

Hypergrowth creates pressure to build everything at once: onboarding, performance management, engagement surveys, career frameworks, manager training, and more. But trying to scale every People program simultaneously can spread the team too thin and produce a lot of activity without meaningful impact.

At Underdog, the team identified managers as one of the biggest organizational unlocks. The reasoning was simple: even the best HR programs cannot compensate for a poor manager experience.

“You could have the best HR team in the world. But if that person’s manager is not good, it almost doesn’t matter what the HR team does.”

The practical move is to identify the one or two constraints that most directly affect company performance. Invest disproportionately in those areas, even if the initial solution is manual, expensive, or difficult to scale.

3. Sometimes the right answer is to do something that doesn’t scale

Underdog’s early approach to manager development was intentionally high-touch: give every people manager access to some form of one-on-one coaching.

Rather than relying entirely on traditional workshops, managers could bring real problems from that week or month into a coaching conversation. That made development more practical, personal, and immediately applicable.

“If we really believe that managers are the unlock for us to have a great organization, what can we do to make sure that we are supporting them at all costs?”

As the company grew, the model evolved. Senior leaders received more white-glove coaching and 360 feedback, while other managers participated in functional or cross-functional cohorts. The lesson is not that every company needs an internal coaching team. It’s that People leaders should drive a valuable program until it breaks, learn from the breaking point, and then redesign it for the next stage.

4. Performance systems should reflect how different teams create value

Most companies use one performance process across engineering, sales, marketing, customer support, and every other function. That creates consistency, but it can also flatten the meaningful differences between how those teams operate.

The way an engineer creates value may be fundamentally different from the way a salesperson or marketer creates value. Over time, People teams should be willing to tailor performance and feedback systems to those realities.

“The way that we evaluate an engineer — should that be the same as the way we evaluate a sales team or a marketing team? The answer is obviously no.”

A common company-wide language may still be useful as a starting point. But once that foundation is established, challenge every element of the process. Different teams may need different questions, cadences, measures, or feedback mechanisms. Consistency should serve the business — not become the goal itself.

5. Shift the conversation from activity to impact

It is easy to celebrate a project because it took three months, involved a lot of effort, and ended in a polished launch. The harder question is whether that work actually changed the company.

At Underdog, the People team is moving from a broad focus on feedback toward a sharper focus on impact. That means managers should not only discuss what someone accomplished, but why it mattered.

Instead of asking:

What did this person complete?

Ask:

• What impact did this person have?
• Why did that work matter?
• How did it change the business, customer, or team?

“It’s not about what you did. It’s about why that mattered.”

This framing also changes the role of managers. Their job is not simply to assign work and give feedback. Their job is to direct talented people toward meaningful problems and help them increase the value of their work.

6. Treat HR like a political campaign

People teams often have dozens of priorities, but employees cannot absorb dozens of messages. Daniel’s framing is to run the People function more like a political campaign: decide the one, two, or three ideas everyone needs to understand, then align programs, communication, and resources around them.

“There are one or two or three things that I want to make sure everyone at this company understands that we care about and why we care about that.”

This creates a clear filter for prioritization:

Does this move us toward one of our stated goals?

Will people understand why it matters?

Will they find value in participating?

If the answer is no, cut it, redesign it, or place it on the backlog. The objective is not to make the People team’s work easier. It is to ensure every major initiative contributes to an outcome the company actually cares about.

7. Use pilots to create permission for experimentation

HR teams can become overly attached to getting a program perfect before launch. That slows the business down and increases resistance because employees assume every new process is permanent.

Daniel’s preferred approach is to label new ideas as pilots. A pilot lowers the perceived risk, creates space to try something different, and gives the team real evidence before making a long-term commitment.

“My oldest trick in the bag, if you want to sell anything to the business, is: ‘We’re just piloting.’”

The pilot should still have a clear hypothesis. Define what problem you are trying to solve, what success would look like, and what you will change based on the results. Many of Underdog’s pilots became the foundation for larger programs — but only after the team learned what worked in practice.

‍