Does displaying revenue increase trust in online-business advice?

The question is narrower than it looks

A founder posts a piece of business advice.

Under the post sits a revenue card: $50,000 in the last 30 days.

Does the advice now feel more trustworthy?

That is easy to assume and surprisingly difficult to answer from existing evidence. Revenue is doing several jobs at once. It can imply experience, competence, status and access to information. It can also look like marketing, invite suspicion about what the number leaves out, or make the creator seem more interested in selling an identity than explaining a useful idea.

The evidence reviewed for this protocol does not provide a clean causal test of the exact question. There are surveys showing that some people value visible financial success when choosing whom to listen to. There are experiments showing that source and social cues can change perceived credibility. There is also evidence that a status signal can raise perceived wealth while lowering perceived trustworthiness.

What is missing is the boring but decisive comparison:

same person, same advice, same page — revenue visible versus revenue absent.

This protocol proposes that test. It has not been run. No participants have been recruited, no data have been collected and no result should be inferred from the design.

Existing evidence says success cues matter — but not cleanly

The closest large-scale evidence comes from personal finance rather than online-business advice.

In a 2025 national survey, the Federal Reserve Bank of Philadelphia asked more than 5,000 U.S. adults about financial advice and social media. Among people who used social media for financial advice, 58.4% rated the perceived wealth or financial success of the advice provider as important. Among all other respondents, the figure was 47.4%, an 11-point gap. The same group was also much more receptive to the idea that an unqualified social-media adviser could still be credible if that person appeared financially successful: 48% agreed and 24% disagreed. The authors explicitly treat these results as associations, not causal effects.

That is relevant, but it does not prove that adding a revenue number makes advice more trusted. People who already seek advice on social platforms may simply care more about visible success to begin with.

Experimental research gives the mechanism more support without answering the revenue question directly. A 2026 meta-analysis of 18 social-media experiments, covering 14,188 participants, found that manipulated source and social credibility cues increased perceived credibility on average, with an overall effect size of g = 0.307. The size of the effect varied by the type of cue and the people seeing it. A cue can influence credibility, but cues are not interchangeable.

Performance itself is also an imperfect credential. In a series of studies with 8,693 participants, David Levari, Daniel Gilbert and Timothy Wilson found that advice from the best performers was no more beneficial than advice from other performers. Participants nevertheless believed the top performers’ advice had been more helpful. Doing a task well and giving good advice about it were not the same skill.

That is exactly the tension around a revenue screenshot. A number can be relevant evidence that somebody has operated a business successfully. It is not direct evidence that the advice underneath it is accurate, transferable or useful.

And status can backfire.

In a PLOS ONE replication study, a luxury-brand label made the person wearing it appear wealthier and higher-status, but the luxury condition also produced lower ratings on trustworthiness and prosociality. The signal increased one desirable inference without improving every other judgment.

The plausible answer, before collecting any new data, is therefore not simply “yes.”

A revenue display might make a creator look more successful and more expert while making them look no more honest — or even less honest.

The first experiment should change one thing

The strongest first test is a two-condition randomized experiment.

Each participant would see a fictional creator giving one short piece of online-business advice. The creator name, portrait, biography, platform layout, follower count, wording, typography and advice would be identical across conditions.

Only the revenue cue would change.

Condition	What the participant sees
No-revenue control	Creator profile + business advice
Revenue display	The identical profile + identical advice + a neutral card showing a specified recent revenue figure

The revenue card should be original study artwork, not a copied Stripe, Shopify or payment-platform interface. Using a recognizable commercial dashboard would add another variable: the participant’s familiarity with and trust in that brand.

The number should also be explicitly framed as a self-reported business result. A label such as “Last 30 days — revenue” is enough. The study should not invent a verification badge or imply independent auditing that did not happen.

The exact amount needs to be fixed before recruitment. A pilot can check that it reads as meaningfully successful without becoming cartoonish. Once selected, the amount stays constant throughout the primary study.

Use more than one piece of advice

One business tip is too fragile.

A claim about pricing may interact with revenue differently from a claim about advertising or retention. The experiment should therefore use several prewritten advice stimuli across ordinary online-business topics, with participants randomly assigned to one topic as well as one revenue condition.

The advice itself should be plausible, specific and low-risk. It should not depend on the creator’s displayed revenue to make sense.

For example, the study could use advice about:

• running one landing-page test at a time instead of changing five variables;
• separating gross revenue from payment fees and refunds when evaluating an offer;
• interviewing recent customers before rewriting a product’s positioning.

The exact stimuli would be finalized and archived before data collection.

“Trust” needs to be split into separate judgments

Asking only “Do you trust this person?” would waste the experiment.

Source-credibility research regularly separates judgments that ordinary conversation bundles together. A successful-looking creator may seem more competent without seeming more honest. That distinction matters because either route could change behaviour.

A 2025 study comparing social-media peer advice with professional financial-adviser advice found that information quality, transparency and personalisation were associated with trust and perceived usefulness, and professional advisers were perceived as more credible, trustworthy and transparent than the social-media sources. Trust depends on more than a single prestige cue.

The proposed study would therefore keep the primary outcome narrow and measure the rest separately.

Measure	What it is trying to capture	Role
Trust in the advice	“I trust this advice” / “This advice seems reliable”	Primary
Perceived expertise	How capable or knowledgeable the creator seems	Secondary
Perceived honesty	Whether the creator seems candid rather than manipulative	Secondary
Perceived business success	Whether the revenue cue actually changed the intended impression	Manipulation check
Willingness to follow the advice	Behavioural intention toward the recommendation	Secondary
Willingness to buy	Interest in a hypothetical paid product from the creator	Exploratory
Confidence the advice would work	Expected effectiveness of the recommendation	Secondary

The wording and scale should be fixed before looking at condition differences.

Keeping these outcomes separate creates a more useful result. If revenue increases expertise ratings but not trust, the headline should say exactly that. If it raises willingness to buy while lowering perceived honesty, that is a different story again.

The primary comparison should remain boring

The analysis does not need to be clever.

The primary test is the difference in the preregistered trust measure between the no-revenue group and the revenue-display group.

Report the size of that difference and its uncertainty, not only whether it crosses a statistical-significance threshold.

Then examine the secondary outcomes.

The study should also preregister a small number of moderators that have a real reason to matter. The Philadelphia Fed survey suggests one obvious candidate: whether a participant already uses social media for financial or business advice. Prior experience running an online business is another. These can show whether the same cue lands differently for people who already live inside this culture.

Those moderator tests should stay secondary. A weak overall effect should not be turned into a dramatic story by slicing the sample until one subgroup produces an exciting result.

Do not choose a sample size because it sounds substantial

This protocol does not specify a participant count.

The final number should come from a prospective power analysis based on the smallest effect worth detecting, the preregistered analysis and the expected exclusions. Picking “1,000 people” because it looks serious would be the research equivalent of a vanity metric.

The recruitment source also needs to be reported clearly. A convenience panel of U.S. adults can test a causal comparison inside that sample. It cannot automatically establish how every entrepreneur, creator or social-media user behaves.

Do not remove participants because the revenue cue failed to impress them

The study should include a manipulation check: did participants notice the number, and did the creator appear more financially successful?

That check should not become a way to clean the treatment group after the fact.

The primary analysis should compare people according to the condition they were randomly assigned to. If the revenue display barely changes perceived success, that is part of the result. Excluding everyone who was unconvinced would answer a different question: what happens among people who accepted the signal.

That can be explored later, clearly labeled as such.

A second experiment can test the screenshot effect

If the first study finds that displaying revenue changes trust, the next question is why the format matters.

A follow-up could compare:

1. no revenue claim;
2. the same revenue amount written as plain text;
3. the same revenue amount presented in a neutral dashboard-style card.

That would separate the informational content — “this person says the business made X” — from the visual language of a dashboard.

This distinction matters because screenshots carry their own authority. Numbers arranged inside a familiar interface can feel like records rather than claims even when the underlying evidence is no stronger. SIGNULL’s guide to screenshots and context makes the broader point: a screen can accurately show what appeared on it without proving the larger story attached to it.

A later study could also manipulate whether the figure is described as self-reported or independently verified. That is a different question and should not be smuggled into the first experiment.

Revenue also leaves out most of the business

Even a perfectly authentic revenue number is incomplete.

Revenue is not profit. It does not show refunds, ad spend, cost of goods, payroll, payment-processing fees, taxes, debt or how long the result lasted. A business can show impressive top-line sales and still have poor economics.

We separate those layers in our breakdown of where the money goes after an online sale and in the explainer on what payment processing actually costs.

That is not a reason to treat revenue as meaningless. It is a reason to test exactly what people infer from seeing it.

If identical advice is trusted more because a revenue figure appears above it, the experiment would document a credibility effect. It would not validate the creator’s margins, prove the advice caused the revenue, or show that the audience could reproduce the result.

The commercial context changes the stakes

A revenue display beside free advice is one thing.

A revenue display used to sell a course, coaching program, community or money-making opportunity can become part of an earnings claim.

In June 2026, the U.S. Federal Trade Commission warned marketers of money-making opportunities to substantiate earnings and lifestyle claims and noted that unusually successful outcomes from a small number of people do not by themselves support claims about what a typical participant is likely to earn. The FTC’s guidance is about the claims consumers are being asked to infer, not merely whether a screenshot contains a real number.

The exact legal obligations depend on the product, claim and jurisdiction. This protocol is not a legal test.

It does explain why the psychology is worth measuring. If a visible revenue number increases trust in advice or willingness to buy, then the display is not just decoration. It can change how the surrounding message is received.

What would count as an answer?

The cleanest possible result is not necessarily the most dramatic one.

If the revenue condition produces a clear increase in the primary trust measure, the study would support the claim that displaying revenue increased trust in this controlled online-business-advice setting.

If perceived success and expertise rise but trust does not, the better conclusion would be:

Revenue made the creator look more successful and capable, but did not make the advice more trusted.

If trustworthiness falls while expertise rises, the display created a trade-off.

If nothing meaningful changes, that is useful too. It would suggest that the revenue cue, at least at the tested amount and in the tested format, does less persuasive work than internet-business culture often assumes.

None of those outcomes should be generalized beyond the actual stimuli and sample without another test.

Our prediction before seeing any data

The directional prediction should be recorded now, not written after the chart exists.

Based on the evidence above, the most defensible prediction is that a visible revenue number will increase perceived financial success and probably perceived expertise.

The effect on trust is genuinely less certain.

The Philadelphia Fed survey shows that people who already consume financial advice on social media place more weight on perceived financial success. Experimental credibility research shows that source cues can matter. But the luxury-signaling study is a useful warning: looking richer and looking more trustworthy are not the same thing.

So the protocol should not preregister “revenue increases trust” as though the answer is already known.

It should preregister the question.

Status

This study has not been run.

Before recruitment, SIGNULL would need to finalize the advice stimuli and revenue display, pilot the manipulation, choose the sampling frame, define exclusion rules, conduct the sample-size analysis and preregister the primary outcome and analysis.

Only then should the experiment collect responses.

Until that happens, the most accurate answer to the headline is:

Displaying revenue is a plausible credibility cue, but we do not yet have a clean causal answer for online-business advice.