🔎 Session Summary

This session walked through FilmHub’s real journey building a transparent compensation model.

Not a polished framework built in isolation.

Not something that appeared fully formed, or ever finished

But a startup that recognised its compensation approach needed more structure, and committed to improving it step by step.

Drew shared the decisions they made, the trade-offs they accepted, and the adjustments they’ve continued to make along the way.

💡 The big theme running through it all:

Start with something clear.

Expect it to evolve.

Be willing to adjust as you learn.

🧭 Phase 1: The Starting Point

When Drew joined, compensation looked like many growing startups:

• Offers built case by case
• Different data sources used at different times
• Location logic applied inconsistently
• No shared level framework
• Equity and salary not clearly linked

Nothing unusual. There were a lot of head nods in the room.

But as the company grew, it became harder to explain decisions clearly and consistently.

When they stepped back and reviewed the organisation as a whole, it became clear they needed stronger foundations.

💡 What you could do next

• Review your last 10 hires and look for patterns or inconsistencies
• Ask whether you can clearly explain why each person is paid what they are
• Identify where decisions rely on memory rather than documented logic

🧱 Phase 2: Establishing a Baseline

They needed structure.

🧠 Decision 1: Choose a primary data source

They selected Carta as their main dataset, with Pave used as a secondary sense-check.

Within Carta, they had to decide which peer group to anchor to:

• Revenue
• Valuation
• Employee count

Each gave different outputs.

They ultimately chose employee count because:

• It provided a broader dataset
• It was more stable over time
• It reduced distortion from valuation assumptions

🧠 Decision 2: Set a clear geographic anchor

They chose San Francisco as their 100 percent benchmark.

All other locations were calibrated downward from that baseline.

In the US, the lowest factor was 0.8.

Internationally, the lowest factor was 0.5.

This created guardrails.

Not perfection.

But consistency and defensibility.

💡 What you could do next

• Select one market dataset and commit to it
• Define your anchor location
• Set explicit floors for location adjustments

🌍 Phase 3: Rethinking Location as Seniority Increased

Over time, market reality challenged the model.

Senior engineers in lower-cost geographies were receiving highly competitive offers elsewhere.

💡 Realisation: A flat location factor across all levels no longer reflected the market.

🔧 So they adjusted:

As levels increase, compensation moves closer to a global midpoint.

Entry-level roles remain more regionally priced.

Senior roles compress toward global norms.

This was one of their most significant structural changes.

It required meaningful salary increases for some employees.

It had real cost implications.

But it improved retention and internal alignment.

💡 What you could do next

• Review how location factors apply across levels
• Analyse retention risk by geography and seniority
• Model cost impact before implementing structural changes

🔄 Phase 4: Refining Equity Approach

Early on, they offered candidates flexibility:

• Higher salary, lower equity
• Lower salary, higher equity

It felt flexible and candidate-friendly.

Over time, it created complexity:

• Salary bands began to align
• Equity differences persisted
• Internal disparities became harder to reconcile

They moved to standardised equity targets by level.

🌱 Lesson learned:

Optionality early can create operational debt later.

💡 What you could do next

• Audit early equity agreements for long-term imbalance
• Set clear equity targets by level
• Avoid multiple mix-and-match models unless you can manage the downstream impact

🔓 Phase 5: Making It Public

They built a public compensation calculator.

🤷 Why?

Primarily for trust 😇.

Internally:

• Employees could see how pay was determined
• Differences had clear explanations
• Conversations became more grounded

Externally:

• Candidates understood level expectations earlier
• Negotiation became more structured
• Offers were easier to explain

Transparency removed the ability to quietly adjust pay behind closed doors.

It increased accountability.

💡 What you could do next

• Share ranges internally before going public
• Pressure-test clarity with employees
• Decide what level of transparency aligns with your culture

🛠 Phase 6: Building the Public Tool

Initially, everything lived in Google Sheets.

It worked, but it was difficult to navigate.

So Drew built a custom calculator himself using AI tools and deployed it via Vercel.

The lesson was not technical sophistication.

It was ownership and velocity.

Rather than waiting for engineering capacity, he prototyped something usable and iterated.

💡 What you could do next

• Improve usability before adding complexity
• Prototype scrappily before waiting for perfect resources
• Focus on clarity for candidates and employees

⚠️ The Ongoing Challenge: Performance and Compensation

Managers can adjust levels anytime.

In theory.

In practice, twice-yearly “circuit breakers” ensure progression conversations actually happen.

The aspiration is continuous evolution.

The reality still requires structure.

This remains an area they are actively refining.

💡 What you could do next

• Decide whether you truly want continuous comp adjustments
• Train managers to own progression conversations
• Introduce structured checkpoints if needed

🎯 The Big Themes from the Journey

1. It is never too early to start
2. It will change multiple times
3. Market dynamics will force iteration
4. Transparency strengthens retention and trust
5. Structure reduces unnecessary negotiation cycles
6. Fixing inconsistency may require real financial investment

And perhaps the most important:

🌱 Start with something.

Even if it is a strawman.

You can evolve it.

You cannot evolve what does not exist.