LLenny's Podcast
← All frameworks
StrategyDan Hockenmaier (Faire, Thumbtack, Reforge)

The Growth Model Building Blocks

Build a spreadsheet formula of how your business grows, layer by layer, to find your real levers

Difficulty
Advanced
Time to result
~weeks to results
Steps
6
Confidence
95%

A growth model is an analytical representation of how a business grows, built in a spreadsheet, that forces you to understand which levers actually matter. Start with three core building blocks and layer complexity based on business type. It is an opportunity-assessment tool, not a forecasting tool.

Origin

Developed by Dan Hockenmaier across 20-30 growth models built at Thumbtack, Faire, and his consulting firm Basis One. He credits Reforge (Brian Balfour) with building this into a formal discipline.

Core principles

  • 01A spreadsheet is hard to fake — forcing assumptions to link and line up forces real understanding
  • 0250% of the value is simply building it yourself from first principles; templates negate this learning
  • 03It is not a forecasting tool — output is highly variable because you stack assumptions; it is for opportunity assessment
  • 04Non-linearity (virality, reinvested contribution margin) is where the interesting levers appear

How to run it

  1. 1

    Model acquisition channels

    For each channel (paid marketing, sales, viral referrals) set assumptions for traffic or spend and conversion rate.

    Pro tip Partner with a smart analyst or finance person and just start building — do not wait for a perfect template.

  2. 2

    Model retention

    Define activation rate and a monthly retention/survival curve so cohorts stack over time.

    Watch out Growth is far more sensitive to retention than you would intuit, but retention is the hardest metric to move because it is the culmination of the whole product experience.

  3. 3

    Model monetization

    Add how customers pay — a monthly/annual fee for SaaS, or transactions-per-month times AOV for transactional businesses.

  4. 4

    Layer in business-type complexity

    For transactional businesses add unit economics (COGS, contribution margin) since marginal costs are high. For marketplaces add supply acquisition/retention and model how supply and demand interact.

    Watch out As soon as you stack assumptions you become highly sensitive to how many assumptions you make and whether you know how to make them — marketplaces create junk-in/junk-out risk.

  5. 5

    Make it non-linear

    Explicitly link virality (customers referring customers) and reinvested contribution margin (earnings funding the next acquisition) so the compounding loops are visible.

    Pro tip Linking these makes it obvious why payback period is a better measure of paid-marketing performance than LTV/CAC — speed of getting money back drives growth rate.

  6. 6

    Split into one macro model plus team mini-models

    Keep one simple high-level conceptual model of the whole system, then give each product pod / go-to-market team its own mini-model for the piece they own and its North Star input.

    Pro tip Do not try to stitch everything into one master model — that is a very difficult task; the dual benefit is understanding the whole plus each team's specific levers.

In the wild

Thumbtack's repeat-rate revelation

The first model Hockenmaier built, with Thumbtack's finance team, made it immediately obvious the business was exceptionally sensitive to the repeat rate of new customers. Almost all traffic came from targeted SEM/SEO for a specific service, so upsell into other categories drove LTV, which fed back into how much they could pay to acquire. They had shipped hundreds of experiments on the initial flow but done much less on lifecycle cross-sell.

The model convinced them to shift resources from top-of-funnel to deeper in the stack, building a much better customer journey.

Common mistakes

Treating the model as a forecast

It is an opportunity-assessment tool; output is deliberately variable because you are flexing assumptions. Using it to replace finance's projections leads to false precision.

Downloading a template instead of building from scratch

Half the value is the painful process of building it yourself — a template hands you the artifact without the understanding of how your specific business works.

Is it for you?

Best for

Founders and growth/strategy leaders who need to weigh opportunities across teams and allocate resources during quarterly or annual planning.

Not ideal for

Teams wanting a reliable revenue forecast, or very early startups without enough data to make credible assumptions.

From the transcript

the analytical representation of how the business grows and is typically built in a spreadsheet which has a really nice feature of being very hard…

06:00

I think 50 of the value you get from it is simply building the model right

06:30

it is not a forecasting tool

06:30

it makes it really clear why thinking about something like payback period is a much better measure of paid marketing performance than LTV to CAC

09:00

one very basic high level conceptual mob

12:00

each product pod each go to market team they should have their own kind of Mini model

12:30

From the episode

Developing a growth model + marketplace growth strategy

Dan Hockenmaier (Faire, Thumbtack, Reforge)