Control-and-Impact Filter
Concentrate resources on a few decisions you can control that change results
- Difficulty
- Easy
- Time to result
- ~days to results
- Steps
- 6
- Confidence
- 94%
The Control-and-Impact Filter narrows a crowded operating agenda to decisions that the business can influence and that materially change outcomes. Begin with the result that needs to improve, then list the initiatives and decisions competing for resources. Judge each by how much control the team has and how significant the likely business effect is. Select only a handful, define ownership and measures, and say no to the rest. The mechanism counters the founders' earlier pattern of agreeing to expansion, travel, and new markets without enough discipline. It also supports their proposed 'day one again' reset: retain the valuable product, brand, customers, team, and experience, but rebuild priorities as if designing the next phase from scratch.
Origin
Extracted from The Foundr Podcast
Core principles
- 01Control over numbers, teams, and outcomes supports better execution
- 02A small number of consequential decisions beats scattered activity
- 03Saying no protects attention for work that matters
- 04Current business evidence matters more than entrepreneurial image
- 05A reset can reuse existing brand, product, customers, and experience
How to run it
- 1
Define the needed result
Choose the business outcome that must improve, such as profit, delivery performance, or sales in a priority channel.
Watch out A vague ambition such as growth cannot distinguish high-impact work.
- 2
Inventory active decisions
List the launches, markets, hires, campaigns, and operating changes currently using attention or money.
- 3
Test control
Assess whether the team can directly influence the inputs and execution behind each decision.
Pro tip Separate controllable operating actions from hoped-for external responses.
- 4
Test impact
Estimate how strongly each decision could change the chosen result using current evidence rather than excitement.
Watch out Activity that looks impressive may have little economic effect.
- 5
Choose the handful
Commit resources to the few decisions that combine meaningful control with material impact. Pause or reject the rest.
Pro tip Make the stop list explicit so rejected work does not quietly continue.
- 6
Measure before expanding
Review the selected actions against their intended results before adding another initiative.
Watch out Do not confuse completion of an initiative with impact on the business.
In the wild
The founders recall saying yes to rapid UK expansion and taking the team to America because something appeared to be happening there. Adam says they spent money and effort without getting a useful result, which prompted a move toward fewer consequential decisions.
→ The experience supplied the case for filtering initiatives by control and impact.
The founders describe resetting the next phase while retaining a proven product, brand, customers, some capital, and a more experienced team. The reset is intended to rebuild priorities rather than discard the company.
→ Existing assets become inputs to a more focused operating phase.
Common mistakes
Saying yes to momentum
Expansion can feel justified by recent wins even when the team has not identified a controllable path to material results.
Spreading attention thin
Many small decisions consume time without producing the progress available from a few consequential ones.
Managing the founder image
Choosing work for the story it creates can displace disciplined attention to what is actually happening in the business.
Is it for you?
Best for
It is best for founder-led teams correcting overexpansion, weak margins, or excessive simultaneous initiatives.
Not ideal for
It is not ideal when the immediate constraint is unknown and no reliable operating evidence exists.
From the transcript
“we need to just make decisions that have”
“Control and impact, that's what that's what good businesses have.”
“We are going to say no to most things.”
From the episode
521: They Made $20M selling Perfume