How to Evaluate Advice Before You Follow It
A useful way to evaluate advice is to ask five questions: What exactly is the claim? What conditions made it work for the person offering it? What evidence supports it beyond that person's story? Who carries the cost if it fails? What is the smallest informative test you can make without committing to the whole decision? The answer may still be “follow it.” You will at least know what you are following.
Before using the questions: They are editorial prompts, not a validated decision tool or a substitute for qualified medical, legal, financial, employment or safety advice. Do not run a personal “test” when experimenting could put your health, housing, rights or another person's well-being at risk.
A proverb can give you courage and tell you almost nothing. “The squeaky wheel gets the grease” recommends speaking up; “the nail that sticks out gets hammered down” suggests caution. In “Advice and Its Evil Twin”, the book's author uses those opposite sayings to expose what a portable rule leaves behind: who, when and under what conditions. The task is not to find a proverb that always wins. It is to recover the conditions the short sentence threw away.
That published essay describes a scene from the book in which a saver and a spender each defend an opposing maxim. Collecting more evidence does not produce a winning slogan, because each maxim can be right under different circumstances. The scene offers a sharper question for this guide: what would you need to learn about a particular situation before either short rule helped? The five checks below are an editorial exercise in asking that question; they are not the book's own five-step system.
1. Turn the sentence into a claim you can check
“Be persistent” is not specific enough. Persistent at what, until when and for what result? Write down an action, a situation and an expected outcome. For example: “If I keep asking for a meeting with this editor over the next two weeks, I am more likely to get a useful reply than if I stop after one unanswered email.” That claim may be wrong, but it is clearer than a slogan. It suggests what you would observe and when to stop.
Watch for words such as always, never and anyone. They do not automatically make advice bad. They do make its unstated exceptions important. “Do what you love” may mean “try more work you find interesting,” or it may mean “quit your job tomorrow.” Those are different claims with different costs. Do not evaluate one while silently acting on the other.
2. Ask which circumstances made the story possible
Imagine two people who hear “take the leap” about leaving a job. One can test a new project for three months, has a second income at home and can return to a similar role. The other has a dependent, a contract they cannot easily leave and no reliable route back. These are illustrations, not real case histories. Neither person is less ambitious. The same line of advice just fails to describe both choices.
List the conditions that made the adviser's success possible: starting resources, obligations, timing, available alternatives and what failure would have cost. Ask whether the conditions match yours. If you cannot see the other person's starting point, the story offers a possibility—not a forecast for you.
The Critical Appraisal Skills Programme makes a useful distinction when judging research: whether a study is trustworthy, what it found and whether the findings are relevant in a particular context are different questions. Its framework concerns research appraisal in health and social care; it does not validate this guide's questions as a general decision method. The narrower point is not to mistake “this happened” for “this applies here.”
3. Look for the missing cases
A success story tells you about one outcome, not the full set of outcomes. It need not be dishonest to leave the other attempts out. Ask what happened to people who tried the same action and stopped, found a different route or did not get the advertised result. If all the evidence offered is a memorable example, label it an example, not a rate of success; even a full count would not, by itself, establish that the action caused the result.
A practical three-line check helps:
- Experience: What exactly happened to this person?
- Cause: What suggests this action, rather than something else that changed, produced the result?
- Transfer: What suggests it would work under my conditions?
If the answer to line three is “I don't know,” you may still learn from the experience. You simply should not borrow its certainty.
4. Compare the cost of being wrong with the cost of waiting
“Who pays if it fails?” is not a rhetorical question. An adviser may care deeply and still be unable to bear the other person's lost income, strained relationship or missed chance. Equally, delay has a cost. Refusing every uncertain choice is not a magic alternative to following every exciting rule.
This small grid is an editorial worksheet, not a tested diagnostic tool:
| What you know | Lower-cost, reversible step | Higher-cost, hard-to-reverse decision |
|---|---|---|
| Mostly one person's story | Run a modest test and record what happens. | Do not let the story decide alone; look for independent evidence and your actual downside. |
| Stronger relevant evidence | Check whether your circumstances still differ. | Test applicability, alternatives, downside and qualified input before committing. |
The grid cannot calculate a correct answer. Its value is making two things visible: how much you know and how much it would cost to be wrong. For health, legal, debt, safety or other high-stakes decisions, use appropriate qualified help rather than treating a website exercise as personal guidance.
5. Decide what would change your mind
Before trying a safe, low-stakes idea, write one observation that would count for it and one that would count against it. “If I can make a small version of this project and learn whether people use it, I will know more than I do now.” Also write a stopping point: a date, a resource limit or a clear result that tells you to reconsider. A test with no boundary can quietly become the irreversible decision you meant to avoid.
Some choices cannot be made small. An important conversation may still need to happen; an opportunity may expire. If the decision cannot be tested safely, do not experiment just to satisfy this worksheet. Say so honestly, examine the evidence and trade-offs more carefully, and seek qualified help where the stakes require it. The absence of a small test does not prove that a big leap is either brave or foolish.
Worked example: “Speak up more”
The instruction sounds sound in a team that invites criticism. Now imagine a manager who punishes disagreement, or a meeting where the relevant decision-maker is not present. The advice may still contain a valuable aim: make your views heard. The next question is how—in that meeting, in a private conversation, with documentation, or after getting advice from someone who knows the workplace. This is an example of making the rule conditional, not a universal recommendation to speak or stay silent.
Write your version as: “If I can raise this concern through a channel that is likely to be heard without creating an unacceptable risk, I will do it by [method] before [date].” If you cannot fill in the method or risk, the missing words are the next things to investigate.
A one-minute version you can reuse
When a confident recommendation comes your way, write:
- The claim: “If I do __ in situation __, I expect __.”
- Its conditions: “It worked for someone who had __; I have/do not have __.”
- Its evidence: “Beyond the successful example, I know __; I still do not know __.”
- Its downside: “If it fails, the cost falls on __; waiting would cost __.”
- Its test: “A smaller or safer way to learn is __; I will reconsider if __.”
You do not need to become cynical about advice. You need the confidence to ask what it would mean in this case. For a deeper exercise, use Unearned Wisdom's free Assumption Audit. For the larger argument about why apparently wise rules often survive without enough scrutiny, see the The End of Wisdom book page. There you can read a free related piece before choosing a format. The book asks questions; neither it nor this guide can make a consequential decision on your behalf.
Editorial note: This original guide was prepared with AI assistance for Unearned Wisdom. Its decision scenarios are illustrative, not reported case studies; the linked proverb scene comes from the author's published book-related essay. Read our editorial approach.