Across six lab and field experiments, their 2008 participants underestimated how many strangers would agree to a direct request by roughly half. Science found the same pessimism at planetary scale in June: more than 100,000 people in 125 countries, 69% cooperating against an average prediction of 47%.

Now the finding most fundraisers have known in their bones for thirty years without a number attached to it. Mahdi Roghanizad and Vanessa Bohns had people make an identical request two ways, face to face and by email. The in-person ask was 34 times more effective (Journal of Experimental Social Psychology, 2017). Not 34 percent. Thirty-four times. And askers never see it coming: in a follow-up across roughly 2,650 people testing in person, video, phone and email, help-seekers expected the channels to perform about the same and badly overestimated the mediated ones (Social Psychological and Personality Science, 2022). Every experienced fundraiser already believes the meeting beats the message. This is the research that says so out loud, and it adds what we did not know: we misjudge the gap, then build our calendars on the misjudgment.

Xuan Zhao and Nicholas Epley found the reason underneath all of it. Across studies with 2,118 adults (Psychological Science, 2022), askers underestimated how willing people were to help, underestimated how good helping would feel, and overestimated the imposition. The asker is solving a compliance problem. The person across the table is being invited.

When you predict what another person will do, your brain does not look up a file on them. It starts with you and adjusts. Diana Tamir and Jason Mitchell found a region of the medial prefrontal cortex whose activity tracked the distance between the self and the person being judged (PNAS, 2010). Self first, correction second, and the correction is never large enough. So when a gift officer says “she won’t go for it,” much of what is being reported is the officer’s own discomfort, wearing the donor’s name.

What that simulation leaves out is the entire return side of the donor’s ledger. In Neurogiving I spend a chapter on warm glow and the helper’s high, the private reward that arrives when someone gives. It is real, and it is invisible from the outside. You cannot see it on a donor’s face while you are rehearsing the ask in your car. So you model the cost of giving and not the payoff, and your forecast runs gloomy. Then one more thing happens, and this part is about you rather than the donor: a “no” lands on the asker as a loss, and losses loom larger than equivalent gains. So we protect the relationship by refusing to risk it, which is a polite way of saying we starve it.

We make the choice on behalf of the donor.

Open your sent folder. Search four phrases over the last ninety days: “no pressure,” “just wondering,” “if you’re comfortable,” and “I know you’ve already.” For each hit, count the words between the start of that paragraph and the actual question. That number is your runway. Anything longer than a sentence is not politeness. It is three or four small announcements that what follows is a burden.

Then add one more column, and this is the one that will cost you something: was this ask ever made out loud? Not followed up by email. Not mentioned in passing at a lunch. Asked, in a voice, with a pause after it. If your most valuable open ask is sitting in a sent folder, the channel research says you have put your most important question in your weakest medium, and that you do not believe the gap is as large as it is.

So the move this week is narrow. Take the highest-value ask, double the odds you perceive of a yes, and move forward. One sentence, then stop talking. Put the exit after the question rather than braided through it: “Some people decide now, some want to talk to family first, some want to revisit. All three are fine with me.” That sentence does not cost you the yes. It is what makes the yes worth having. And notice what changed between the two versions. A softener is about you. An invitation is about them.

Research Sources

In Lab Notes Pro

The version of this research that will get repeated badly at every conference next spring is “just ask more.” Pro has the evidence that keeps you out of that ditch: a Salvation Army field experiment where asking out loud raised donations by up to 69% and at the same time sent roughly one in six shoppers to a different door, and a door-to-door study that identifies exactly which slice of your revenue depends on donors being unable to avoid you. It is not the slice most people guess, and knowing which one it is should change what you do with your acquisition budget. Pro also carries what this means across six specific contexts, three failure modes including one called The Cornered Yes, two hundred words from inside the head of a donor who has been waiting two years to be asked, and The Sealed Prediction, a test that scores your team’s forecasting against what donors actually do. It costs nothing to run, and most shops discover their qualification process is filtering on staff comfort rather than donor capacity.