Why Good Evidence Still Fails to Move a Decision

In 2024 behavioural scientists ran 11 climate-communication interventions on 59,440 people across 63 countries. Take the place of one participant. You read a message designed to change how you think about climate change. The researchers then measure several things separately, what you believe, which policies you support, whether you would share climate information, and how much effort you will put into a tedious task that funds real tree planting. Across the study, some interventions moved beliefs or policy support a little, and one increased stated willingness to share information. None of the 11 increased effort on the tree-planting task, and several reduced it.

The experiment did not test investment or procurement, and it does not show that communication cannot change behaviour. Its more modest lesson is useful enough. Movement in one outcome does not guarantee movement in another, and the outcome that cost the participant something was the one no intervention raised.

There are many situations in which strengthening a case is exactly the right work. Weak evidence needs better evidence. Bad economics need fixing. A claim that cannot survive scrutiny should be narrowed or abandoned, since presenting it more persuasively leaves the weakness in place. The harder cases begin once the proposition is credible enough to deserve a decision and the decision still does not move.

Information and action are different outcomes

There is an intuitive model of persuasion in which action follows information. Someone does not know enough, so the case explains more. Once the person understands, they believe, and once they believe, they act. Real decisions run less neatly. A person can understand more without agreeing, believe more without acting, or intend to act and then hesitate once the cost becomes real. Why Climate Messaging Fails follows the same gap between belief and action through the wider research.

Commercial decisions make the distinction hard to ignore because action carries consequences. Saying yes can commit money, create implementation work, expose a reputation, consume political capital or make somebody answerable for the decision later. At that point another true fact may add very little. The evidence has done its job and something else now determines whether the decision can move.

Roee Nir, CEO and co-founder of the cultivated seafood company Forsea, described meeting that limit earlier in his career in his Brighter Future interview, published in December 2022. “I led successful clinical development plans that did not reach the market, although their results were very promising.” Among the lessons he took from those failures was “how to make an effective sales pitch both inside and outside the organization”. He does not say what stopped those plans, and one founder's recollection cannot show it. What it records is a biotech engineer whose clinical results were promising and whose products still did not reach the market, and who drew from it a lesson about pitching inside the organisation as well as outside it.

Different failures can sound the same

The Belief Gap uses four states as an order for checking a stalled case. They are Evidence, Clarity, Conviction and Commitment. Clarity concerns whether the proposition can be understood. Conviction concerns whether the reader accepts it and sees it as relevant to their own situation. Commitment concerns whether somebody can justify acting on it.

The sequence is an order for checking, and people rarely move through four neat psychological stages. Concerns about cost can reopen questions about evidence. A new stakeholder can challenge something that seemed settled. Different people in the same organisation can be in different places at the same time. The value of the sequence is diagnostic, because similar language can conceal different problems. A reader who does not understand a mechanism can sound sceptical. So can a reader who understands it and distrusts the source. A buyer who accepts both and still cannot get a purchase through the organisation may also ask for more evidence, and each of those three needs different work.

Clarity problems often start inside the company. Everybody understands the technology so well that they have forgotten which parts an outsider cannot reconstruct. Research on the illusion of explanatory depth shows how easily people overestimate how well they understand familiar mechanisms until they try to explain them step by step. Authors face the reverse problem, because familiarity makes missing explanation hard to see. That calls for a better explanation, and more evidence would leave the gap where it is.

Conviction brings a different set of problems. A reader can follow the mechanism and still reject the claim because the source is not trusted, the evidence feels too distant from their own case, or the implications of accepting it are unattractive. Research on solution aversion is relevant here. In experiments, acceptance of an underlying proposition shifted depending on how acceptable people found the proposed response to it. The study did not examine climate-tech procurement, so the analogy has limits, and it still warns against assuming that every objection framed as doubt about evidence begins with evidence.

Belief does not remove the decision

The final crossing is different again. A person can understand the proposition, accept it and still be unable to act. The sentence at the centre of the Belief Gap report marks that crossing. “I believe you. I cannot sign this.” The essay of the same name, I Believe You. I Cannot Sign This., follows what changes when a private belief has to become a visible commitment.

The same crossing appears in capital and governance. An investment partner may understand the company and still need a case that survives committee. A board member may support a strategy while the organisation remains unable to implement it. A funder may accept the evidence while the structure of the grant or programme makes the commitment impossible. In each case the quality of the original argument still matters, and the case now has to survive beyond the meeting, which is where a communications problem starts to become an organisational one.

Find where the questions stopped

Before examining any of these crossings, the Belief Gap checks six underlying conditions, Evidence, Economics, Incentives, Authority, Capacity and timing, and Regulation, because a person can reject a case for a completely rational reason. When Better Communication Is the Wrong Answer sets out what each check rules out. Once the decision underneath is viable, the work is to read the material in the order the decision-maker receives it and find the First Break, the earliest point where the case stops carrying them. Start With the Decision walks through that method.

This approach often reduces the amount of work. A team that expected to rebuild a deck may discover that one number needs a proper comparison. A founder who expected to produce more evidence may discover that the investor does not understand what the next round actually funds. A commercial team may find that the customer understands the proposition and has never been offered a first commitment it can realistically make.

Brighter Future will not add evidence to a credible case before it has seen where the questions stopped. Evidence, explanation, commercial structure and organisational decision-making each have different jobs, and a case becomes weaker when one of them is asked to do the work of all the others. Adding proof to a case that stalled at commitment asks evidence to do the work of structure.

Overwork

For the past five years I have read the material behind cases like these, investment decks, commercial proposals, programme cases, board papers and campaigns from climate companies, frontier technologies and mission-led organisations. One pattern recurs. Teams often continue strengthening the case after the weakness has moved somewhere else, and the result is a strange kind of overwork. The material improves while the decision remains in the same place.

That is the tree-planting result at the scale of a single company. Each new round of work moves the outcomes the material can reach, understanding and belief, and measures its success there. The outcome that costs the reader something, the signature or the transfer or the purchase order, sits where it was, and the team reads its stillness as a sign that the case needs to be stronger still.

This essay opens a series drawn from The Belief Gap Report 2026. The next two pieces look at the moment belief becomes a visible commitment, in I Believe You. I Cannot Sign This., and at the method for finding the First Break, in Start With the Decision. Brighter Future applies the method through the Belief Gap Diagnosis.

Sources and further reading

• Vlasceanu, M. et al. (2024). Addressing climate change with behavioral science: a global intervention tournament in 63 countries. Science Advances, 10(6).
• Rozenblit, L. and Keil, F. (2002). The misunderstood limits of folk science: an illusion of explanatory depth. Cognitive Science, 26(5).
• Campbell, T. and Kay, A. (2014). Solution aversion: on the relation between ideology and motivated disbelief. Journal of Personality and Social Psychology, 107(5).
• Roee Nir, Forsea, founder interview
• Why Climate Messaging Fails and What Story, Psychology and Neuroscience Actually Tell Us