The Manager Tax

Why It Matters

The analysis shows that costly managerial layers erode fundraising effectiveness, prompting a shift toward outcome‑based pricing that could lower expenses and boost donor retention.

Key Takeaways

• Manager AI model cost four times market price, delivered poorer results
• Fundraising agencies pay for orchestration, not donor outcomes
• Lack of price signals forces reliance on activity metrics like send volume
• Switching to net retained value pricing aligns incentives and cuts costs
• AI specialist agents can outperform managers when priced per outcome

Pulse Analysis

The AI experiment outlined by Kevin Schulman exposes a stark inefficiency he calls the “manager tax.” Fifteen tasks were run in three setups: a single LLM working alone, a “manager” model delegating to specialist agents, and a market system where models bid for each subtask. The manager approach cost about four times more than the market solution and delivered poorer output. This mirrors fundraising agencies, where a senior director orchestrates direct mail, email, and telemarketing, inflating costs without improving the donor experience. The experiment underscores how coordination overhead can outweigh the benefits of specialization.

The problem aligns with classic economics from Coase and Hayek. Coase noted firms arise when market transaction costs are high; Hayek argued price signals convey dispersed knowledge. Fundraisers have embraced Coase’s side—building integrated agencies to avoid haggling—while ignoring Hayek, because there is no transparent price for the true campaign outcome. Agencies therefore track easy metrics like mail volume or gross response, creating a “volume trap” that rewards activity over effectiveness and drives falling donor retention. Consequently, agencies chase vanity metrics, mistaking volume for impact.

The fix is to price work on net retained value over a multi‑year horizon instead of on activity counts. Tying compensation to three‑year donor retention aligns agency incentives with nonprofit goals and gives AI specialist agents a clear outcome metric. With transparent price signals, the manager tax shrinks, costs fall, and donor experiences improve, allowing the fundraising sector to leverage AI without the overhead of costly orchestration. Adopting outcome‑based contracts also encourages continuous AI innovation to meet retention targets.