Original Position Impact Estimation

	B	C	D	E	F	G
1	Original Position - Impact Estimation Model
2
3	This model estimates the impact of Original Position over a two-year period under expected and conservative assumptions. It provides all of the assumptions underpinning the claim that Original Position could achieve a >15X giving multiple within two years, with current confidence levels for each. I encourage you to make a copy of this model and play around with the inputs to assess your own confidence level. Please also refer to this report from Founders Pledge for more information on the underlying methodology, and this report and model from Giving What We Can to see how this method is used by an established effective giving organisation (and a huge thank you to both for making this work public!). In plain English, the model estimates that at a cost of $460k over two years, I can build an offering sufficiently compelling that 280 users fund OPDAF accounts with an average initial balance of $5,400, 3% of their pre-tax income. They will aspire to increase this by 0.75 percentage points each year until they hit a giving goal of 8%. The average user saves 40% of their donation each year to earn 5% interest, and distributes 50% of the remainder to the most effective charities. This would have a counterfactual present value impact of $7.1m, constituting a 15.5x multiple on the $460k cost, which I have ~60% confidence in achieving. I have 90% confidence of hitting at least a 3x multiple based on the more conservative set of assumptions. A simplified model of how the process works: A user opens a DAF account and funds it with a certain % of their gross income They decide what proportion of that balance they wish to save for that year, which they earn interest on They then decide where to donate the remainder of the balance between effective and less effective orgs They repeat this process once per year
4
5	Cell format legend	Input	Calculation/Output
6
7	Key Metrics	Conservative	Expected	Importance	Confidence	Rationale/Commentary
8	Necessary audience (#)	7500	11200	Medium	Medium	Difficult to estimate due to exponential growth path from network effects. Conservative estimate assuming low growth rate during period. V2 marketing operation estimated to be ~40% more effective
9	Conversion rate to funded account (%)	2.0%	2.5%	Medium	Medium	Average donation page CVR 16% (UK, 2021; Source); average eCommerce CVR: 3.0-4.2% (2022; Source)
10	# Users with funded accounts	150	280	High	Low	21 months of operation (functional offering from Sep 1). Expected to add 12 users/month plus a converted waitlist of 28. Conservative: 7/month and 3 waitlist
11	Ave pre-tax income of users ($)	$150,000	$180,000	Medium	Medium	Experience filters out those not in the target demographic ($100k-$1m income). $180k conservative estimate of mean and median income within demographic (IRS, 2021)
12	Ave real income growth rate	2%	2%	Medium	Medium	Conservative estimates given lack of readily available data. ChatGPT estimate from various sources of 2-4%. Real median household income grew 1.9% per year from 2012-23. Evidence to suggest that real incomes of higher earners have grown more quickly. Potential negative impact of white-collar job automation from AI within the next two years cannot be understated
13	Ave existing pre-tax giving %	2.50%	2.50%	N/A	N/A	Current average donation of ~2.5% of gross income for target demographic (IRS, 2021). Does not directly factor into model, rather just informs the counterfactual coefficient (which will be higher if the pre-existing giving % is lower)
14	Ave initial pre-tax account funding (%)	3.00%	3.00%	Medium	Medium	Conservative estimate from initial MVP testing suggesting that an initial outlay above $5,000, and incremental giving budget above $1,000 are mental barriers for the average user. This is less of a barrier the greater the income of the user
15	Ave initial account funding ($)	$4,500	$5,400	Medium	Medium	Calculated values - see above
16	Ave annual increase in account contribution (percentage points)	0.50%	0.75%	Medium	Medium	Captures the proposed 1 percentage point increase each year weighed against those who won't increase or at a slower rate. V2 estimated at 50% more effective.
17	Ave final pre-tax giving (%)	6.5%	8.0%	High	Medium	increasing to an average final pledge of 8% for V2
18	Ave proportion of account balance 'saved' (%)	50%	40%	Low	Medium	Strategy and communication will focus more on in-year distributions vs saving, so will expect to be lower than benchmarks of All DAFs: 76% (2023, AP News); Daffy: 62%
19	Average annual interest earned on 'saved' balances (%)	4%	5%	Low	Medium	Historic average rate of return for investments that are 'locked-up' for 1-year, as assumed in this model
20	Effectiveness coefficient of donations (%) AKA Ave proportion of donations distributed to GiveWell Top Charities	40%	50%	High	Low	The advisory experience is designed to encourage 80% allocation to GiveWell's Top Charities, though this will likely start from a lower level and increase over time, decreasing the point average. Equivalent GWWC conservative estimate is 67% (2020-22)
21	Counterfactual coefficient	0.3	0.4	High	Low	If Original Position were to only increase the rate of adoption of both a higher giving % and a greater % to effective charities, and not even effect what those higher levels are (suggesting people are inherently effective givers, they just would take longer to come to that realisation and behaviour change) would imply a counterfactual coefficient of 0.7. In reality, the experience will likely increase both of these top rates, which could achieve counterfactual coefficients of >0.9. Equivalent GWWC conservative estimate is 0.2 (2020-22)
22	Average user lifetime (years)	25	25	Low	High	Target demographic age 30-60. Marketing will be focused on 30-40 bracket, giving an average 25 years before average retirement age for income bracket of 60
23	Compound nominal discount rate	0.90	0.91	Medium	Medium	Conservative estimate based on a discount rate of 4.5% (US federal nominal rate at 1/1/25, which includes expected inflation), and a churn rate of 4.5-5.5%
24	Counterfactual present value of a new account	$9,268	$25,500	High	Low	Calculated values: see relevant 'Donation Model' sheet. Donation growth model will make this difficult to evaluate with high confidence without 3+ years of cohort data
25	Total counterfactual present value of accounts	$1,390,182	$7,139,877	High	Low	Calculated values
26	Total Cost ($)	$460,000	$460,000	High	High	High-end of estimated budget range for each period. Incorporates build of product and running costs for the two-year period
27	Giving Multiple	3.0	15.5	High	Medium	Calculated values. ~50% confidence in the claim that Original Position could achieve a >15x Giving Multiple over this two year period
28
29
30
31
32	Operation
33	Ave proportion of donations distributed to Original Position (%)	1.0%	2.5%	Medium	Medium	Advisory experience will recommend 20% of donation budget allocation to Original Position if the Giving Multiplier has been proven sufficiently high with sufficiently high confidence. Before then, comparison to Daffy which costs $36/year, and legacy DAFs that charge 0.6% of fund balance (equivalent to a 1.2% donation rate on a 50% savings rate)
34	LTV to Original Position per user	$772	$3,187	High	Medium	Calculated values
35	Ave distribution to Original Position per user within period	$61	$218	Medium	High	Less important than Giving Multiplier (assuming foundations are willng to donate to a high-confidence, high-multiplier offering if it is not also independetly financially sustainable) but still a medium-term goal. Initial giving percentages may be lower, decreasing this amount, though lower than standard one-off donation for this demographic and given users will be a captive and engaged audience, confident it will not be lower than these
36	Total operational income	$9,127	$60,981	Medium	Medium	Calculated values
37	Total acquisition costs	$300,000	$300,000	High	High	Assumes only 20% of my time to be spent on non-user acquisition activities. Biggest risk is that product development requires greater cost and my time than budgeted, reducing acquisition budget. CAC is likely to be consistent under this scenario, but will reduce Giving Multiple due to lower total user acquisition
38	CAC	$2,000	$1,071	High	High	Benchmark Consumer Tech CACs: Fintech $202, EdTech $264. I estimate Daffy's CAC to be <$350. It is hard to imagine Original Position requiring 3-6x this cost to acquire each user. To be validated with marketing experts and landing page/waiting list testing
39	LTV/CAC ratio	0.4	3.0	Medium	Medium/High	2.5-3.0 LTV/CAC ratio generally considered optimal for Consumer SAAS startups. Below this suggests unsustainable operations, above this suggest under-investment in growth
40
41
42		Backlog to improve model
43		Accounting for % of this cohort creating accounts in Y2 and discounting accordingly
44		Expected increase in % donated to effective charities over time (rather than point average)
45		Expected increase in % donated to original position over time (rather than point average)
46		Adding a terminal value to account for the distribution of final funds at the end of the 'lifetime' period stated
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100