Focusing on what matters
Jona Glade
2025
Theories of change, OKRs and KPIs
Agenda
Intro
Goals for this session
Content part 1: Theories of change
Content part 2: OKRs
Content part 3: KPIs
Wrap-up and actionable next steps
Goals for this session
We strive to achieve three goals in this session while being aware that there is no one-size-fits-all approach
Gain shared understanding of ToC, KPIs and OKRs
1
Learn from each others experiences
2
Provide actionable next steps
3
Three goals include discussion of case studies and learnings from each other
No one-size-fits-all approach
Characteristics (see details later) of ToC, OKRs, and KPIs are very context-specific and vary a lot between organisations
These tools might not even make sense to prioritize depending on your specific context
Working on your ToC, OKRs, and KPIs will be a journey of constant updating and refinement with different priorities depending on what stage your organisation is in
This session will be rather superficial. Happy to follow-up with potential in-depth discussions
Anything missing?
Please post in the chat
What topic do you want to spend the most time on in 2023 (ToC, OKRs or KPIs)?
What are the 1-3 most relevant questions that come to mind when thinking about your priority topic?
Theories of change
"I am increasingly convinced that the difference between effective and ineffective people is their skill at developing a theory of change
Building a theory of change for your research –
aka: one tool for maximally kicking global problems in the butt
Aaron Swartz, Demand Progress
Michael Aird, Rethink Priorities
A non-profit’s ToC is analogous to a business model in the for-profit world. Just as you wouldn’t found a company without a clear business model […], one shouldn’t found or fund a charity without a strong theory of change.
In fact, ToC is more important for charities than business models are for businesses, because businesses have better feedback loops. If a for-profit is based on a bad idea, or has bad execution, it will see poor revenue and profits and it will soon go out of business. If a charity is ineffective on the other hand, it may limp along for years without having any impact, squandering limited funding and talent in the process.
Aiden Alexander, Charity Entrepreneurship
Simplified ToC of AMF to consider as example - in reality more sources of value with more complex interactions
Number and nature of steps differ between organisations
Input
Output
Impact
Activities
Outcome
$4 per mosquito net
N of nets distributed
N of lives saved
Distribution of mosquito nets
% Malaria infections reduced
Key assumptions, e.g., mosquito nets actually help against malaria
Key uncertainties, e.g., will people use mosquito nets in the right way
Quite often it makes sense to first create a long-list of main impact levers and then create a detailed ToC only for main levers
There are a four approaches to designing a ToC, choose by considering the purpose and available resources
For every project/major activity
Purpose
Example
Quick & dirty
Visually & detailed
Long & nuanced
Mix possible
Share learnings about ToC →Audience applies learnings →Better ToC →More good done in the world
To create joint clarity and facilitate high-level strategic thinking
Create clear flow-chart (e.g., see example on previous slide)
To elaborate on complex thinking
Create ~30-paged google doc showing high level of reasoning transparency
Get benefits from different formats by combining them
Mix and combine flow-charts, reports or quick ToC sketches
1
3
3
4
A theory of change can be developed and updated following a six-step process
e.g., expand program by sending job offers to alumni
Identify
value sources
1
Prioritize sources
of value
2
Create �flow-chart
3
Think �end-to-end
4
Collect data
5
Re-iterate
(as needed)
6
Consider the different ways your organization creates value by doing forward chaining based on comparative advantage and backward chaining based on what the world needs the most
Prioritize your sources of value by doing BOTECs1, collecting expert and field data as well as thinking through what would happen if your activity wouldn’t exist/the counterfactual
Visualize the results in a flow-chart, adding assumptions and uncertainties
Probe what part of the process of impact generation is currently neglected/has weak links
Normally there are lots of uncertainties. Make a plan to collect data to inform most important ones
Re-iterate steps 1-6 as needed (e.g. when new data emerges or as the organization develops)
e.g., motivation to apply for role in a field
e.g., impact of credential vs. having tested fit for work
e.g., map how specifically careers are influenced
e.g., run a shorter fellowship as experiment
Steps 1 to 6
Activities per step
Example (Fellowship)
1 BOTEC = Back of the envelope calculation
e.g., continuously look for ways to create more impact
Number and nature of steps might differ between situations
Example | The Fish Welfare Initiative does experiments to confirm their ToC
You want to be strategic/proactive about testing key uncertainties and assumptions, e.g., via small experiments
FWI approach to ToC:
Source: Link
ToC of an organisation developed in different stages
1
2
3
4
Lifetime of an organisation
Think and talk a lot about ToC of your new idea until you hit diminishing returns, and you need to collect data
Do a lot of experimentation and/or research to confirm assumptions and reduce uncertainties
When convinced that your ToC is roughly robust one can start finetuning the activities and measure what you care about
Keep optimising activities and start exploring alternative activities to achieve same or different goal
Duration and content of steps may differ between organisations
ToC can be developed on org, team or individual level
Helps to always focus on value creation and think about different ways to create value/achieve the goal
Individual level
Team level
Organisational level
Common pitfalls to consider
No or too little time is spent on thinking through the ToC and improving it
The ToC is either too high-level or too complex
The ToC is not used for prioritizing activities, quarterly planning or internal and external communication
The current list highlights main points and is not exhaustive
The counterfactual component of the ToC is not considered enough
Key uncertainties and assumptions are not highlighted
Let's discuss
Name of your organisation
Input
Output
Impact
Activities
Outcome
TBD
TBD
TBD
TBD
TBD
Key assumptions, e.g., TBD
Key uncertainties, e.g., TBD
Do you have any questions for me or the others?
What worked well regarding ToC and what didn’t? When do you think was it useful/when wasn’t it useful?
What do you think is the biggest improvement lever for ToC for you/your org?
Homework | Consider these steps to put theory into action
Optional: Look at this collection of documents on ToC, if you want to learn more
Think through the main sources of value and build or improve your ToC based on recent learnings and example ToCs from this workshop or from here
Feel free to reach out in case of any questions during or after the program
Due to workshop context discussed material is high level, independent reading and thinking strongly encouraged
OKRs
What is the difference between OKRs and KPIs?
OKRs help you to track progress and achieve your (ToC-based) goals, e.g., save more lives
KPIs help you to constantly measure performance, e.g., donations / quarter
OKRs
VS
This list may not be exhaustive as definitions of OKRs and KPIs vary
Deep dive next slides
KPIs
AMF | Made up examples to illustrate OKRs
Objective:
Save more lives
Objective:
Improve communication with donors to increase funding
Key result 1:
Increase # of nets in current countries of operation by 40% in 2023
Key result 2:
Distribute nets in one new country of operation in 2023
Key result 3:
Raise 20% more donations in 2023 to finance increased number of nets distributed
Illustrative OKRs do not represent actual OKRs of AMF
Key result 1:
Establish 3 funder calls throughout the quarter
Key result 2:
Develop an impact report by the end of the quarter
Key result 3:
Have one podcast episode to share insights
1. Link
Defining OKRs can be combined with a forecasting exercise
Please post in the chat
If you have used OKRs before (yes/no)
Advice you would give others who want to use OKRs
Consider a few ideas for setting up OKRs
Objectives
Key results
1. Most common is 3 months
These objectives can be part of your ToC
Case study FWI | OKRs used to break down ToC
This is the source.
(FWI's ToC as a recap)
Summary of objectives
Breakdown of objectives into key results
Overall progress tracking
Progress tracking of key results
ToC broken down into different objectives
Accountable
Common pitfalls to consider
OKRs are not actively managed and updated
OKRs are probably not the right tool in highly unpredictable environments without clear goals e.g., start-ups
OKRs do not link to the high-level theory of change and strategy
OKRs focus on activities, not outcomes
The current list highlights main points and is not exhaustive
OKRs are valued too highly and lead to naive over optimization, e.g., performance reviews based on OKRs
Let's discuss
Do you have any questions for me or the others?
When do you think OKRs are useful?
What do you think is the biggest improvement lever for OKRs at your organisation?
Homework | Consider these steps to put theory into action
Optional: Learn more via What Matters and Mooncamp as deep dive reads (slightly different definition of OKRs)
Use FWI's or Mooncamp's OKR Excel Template as a starting point to define your own OKRs for the next three months. If you find it useful you can test them with your team and eventually role it out across orgs
Evaluate your OKRs after 1-3 months. Do not hesitate to reach out to your peers or me in case of any questions or challenges along the way
Due to workshop context discussed material is high level, independent reading strongly encouraged
KPIs
Recap | What is the difference between OKRs and KPIs?
OKRs help you to track progress and achieve your (ToC-based) goals, e.g., save more lives
KPIs help you to constantly measure performance, e.g., donations / quarter
OKRs
VS
This list may not be exhaustive as definitions of OKRs and KPIs vary
KPIs
Deep dive next slides
Please post in the chat
Which KPIs you track at your organisation (if any at all)
Which ones you should be tracking (vague definition ok)
CART framework helps for high-level KPI definition – how to measure anything is a good read for finetuning the KPI
Credible
C
Credible monitoring systems collect high quality data and analyse these data correctly.
Actionable
A
Actionable systems collect data that could actually affect how the organisation works. The collected data should be analysed and used to be able to take concrete action.
R
Responsible
Responsible data collection systems strike the right balance between money spent on data collection and analysis and money spent on implementing programs.
Transportable
T
Transportable systems generate knowledge that can be useful for other programs. This can also help other organisations to improve.
Source can be found here
Common pitfalls to consider
The counterfactual component of the KPI is not considered enough
KPIs are used when they shouldn‘t be used e.g., when it is hard to capture the value or the sample size is too small
KPIs are valued too highly and lead to naive over optimization e.g., performance reviews should not be solely based on KPIs
KPIs aren‘t combined with qualitative data e.g., case studies to understand concrete mechanisms better
In general, downside risks of specific KPIs aren‘t mapped out and countermeasures aren‘t defined
The current list highlights main points and is not exhaustive
Let's discuss
Do you have any questions for me or the others?
Is there anything you would like to share about KPIs (e.g., doubts or questions, support, new ideas)?
What do you think is the biggest improvement lever for KPIs at your org?
Homework | Consider these steps to put theory into action
Optional: Read the Goldilock Challenge, the LessWrong forum post How to Measure Anything, and the What Matters post to learn more
Use recent learnings as a starting point to define/review one KPI, which you think captures your main source of value best (this can be extremely hard and maybe not worth the effort)
Evaluate your KPI regularly and potentially adjust the KPIs if the organisational context or goals change. Do not hesitate to reach out in case of any questions or challenges along the way.
Some example KPIs for the non-profit sector can be found here
Due to workshop context discussed material is high level, independent reading strongly encouraged
Wrap-up and actionable next steps
Your ToC, KPI and OKR development will be a journey of constant re-iteration
This is the source.
At some point you probably want to do impact evaluation, which will probably go beyond the KPIs and OKRs you measure and include, e.g., a one-off survey to capture all sources of value
Plan
Track
Evaluate
Workshop's main objectives serve as reflecting point
Gain shared understanding of ToC, KPIs and OKRs
1
Learn from each others experiences
2
Generate actionable ideas for improvements
3
Three objectives included discussion of case studies and learnings from each other
Any remaining questions or required information/material based on our three objectives for this workshop?
Homework | Various resources and proposals for concrete activities to choose from
KPIs
ToC
OKRs
motivational picture/quote
Let‘s kick global problems in the butt (and reduce the chance that we are kidding ourselves)
THANK YOU