The sustainable-investing intention–action gap

Surveys often find substantial interest in sustainable investing. Product holdings rarely mirror that interest perfectly. It is tempting to label the difference an “attitude–behaviour gap” and move on. For a financial-services team, that label is only useful if it opens a more specific investigation.

A person may care about sustainability but not know which product is credible. They may be unable to compare impact claims, expect a financial trade-off, distrust marketing language, face an overwhelming fund list, postpone a complex decision or simply lack a relevant opportunity to act.

The gap is therefore an outcome to explain, not a single behavioural cause.

What research tells us about sustainable preferences

Several studies provide unusually strong evidence that sustainable preferences are not merely survey politeness.

Riedl and Smeets linked administrative holdings, survey responses and incentivised experiments. They found that social preferences and social signalling helped explain socially responsible fund holdings, while financial motives played a smaller role in their sample.

Bauer, Ruof and Smeets studied members of a Dutch pension fund who received a real vote over its sustainable-investment policy. A majority supported increased sustainability, including among some participants who anticipated lower returns. Follow-up research found continuing support after implementation.

But preference for a sustainable option is not the same as sensitivity to how much impact it creates. In a framed field experiment, Heeb and colleagues found that experienced investors were willing to pay for a sustainable investment, yet willingness to pay did not significantly increase with a tenfold increase in the stated impact. Their findings point toward emotional as well as calculative value.

Four kinds of barrier

u impact’s original theory-building study combined 106 questionnaire responses with 53 investor interviews and 12 investment-adviser interviews. It found recurring factual and behavioural barriers, including jargon, too many options, difficulty quantifying impact and concerns about trust and greenwashing.

For practical diagnosis, barriers can be organised into four overlapping groups:

1. Capability

The investor does not yet understand the product, relevant financial concepts or the sustainability information needed to compare options. The appropriate response may be explanation, but only at the level needed for the decision.

2. Confidence

The investor may have a reasonable foundation but still feel that investing is not for them. More technical content can accidentally confirm that belief.

3. Ambiguity and trust

The person cannot judge how a claim was produced, whether outcomes are measurable, or whose interest the recommendation serves. Transparent evidence and limitations matter more than reassurance.

4. Journey and choice architecture

The path from interest to action is too long, too abstract or overloaded with undifferentiated options. A motivated investor can still leave when the next step is unclear.

Why averages hide the work

A single conversion rate combines people with different barriers. Sending the same sustainability explainer to all of them may help a low-knowledge group, bore a capable group and do nothing for people whose real concern is trust.

Segmentation becomes useful when it changes a decision:

• Which group receives a confidence-building explanation rather than another lesson?
• Who needs a concrete impact comparison and provenance?
• Where is a smaller choice set more helpful than a bigger catalogue?
• Which cohort needs human reassurance or the ability to ask a question?

The purpose is not to give everyone a catchy label. It is to identify a sufficiently specific hypothesis that a team can test.

From hypothesis to transparent intervention

A responsible intervention should support agency. That means the intended action is clear, relevant information remains available and opting out is not made punitive. Hansen and Jespersen’s nudge framework distinguishes transparent from non-transparent approaches, while the Behavioural Insights Team’s applied method emphasises defining the outcome, understanding context, designing the intervention and then testing and adapting.

A good sequence is:

1. Define the behaviour. “Invest more sustainably” is vague. “Complete a comparison and choose whether to save one fund” is observable.
2. Identify the cohort and stage. Find where a particular group stalls rather than applying a global explanation.
3. State the hypothesis. For example: unfamiliar impact language may be preventing otherwise interested, capable investors from comparing products.
4. Change one meaningful part of the journey. Replace jargon with a comparable impact explanation and visible source.
5. Measure the agreed outcome. Look for movement and unintended effects. If nothing changes, revise the hypothesis.

What to measure

Measurement should reach beyond clicks without pretending every action proves long-term impact. Depending on the journey, useful outcomes may include comprehension, comparison completion, shortlist creation, first contribution, retention or the quality of an adviser conversation.

Where possible, compare the survey profile with later observed actions, with appropriate consent and governance. That is how a scenario-based signal can acquire predictive evidence over time. Until then, describe it as a survey response, not behaviour observed in a market.