Ethical Decision Making Framework: 7 Steps That Work

An ethical decision making framework is not a way of arriving at a comfortable answer. It is a way of making sure that when the answer is uncomfortable, you still see it, name it and can explain why you chose what you chose. Most bad corporate decisions were not made by bad people. They were made quickly, by one person under a target, with nobody in the room whose job it was to ask the awkward question.

Why a framework, and not just judgement

Judgement is not neutral. Under commercial pressure people reliably discount harms that land on strangers, overweight the concerns of whoever is present, and construct a reason for the option that helps them hit a number. None of that feels like dishonesty from the inside; it feels like being pragmatic.

A framework does three specific jobs. It slows the decision at the moments where bias bites hardest. It brings in the people the decision affects, or at least their interests. And it leaves a record, so that the next person facing the same question inherits reasoning instead of folklore.

The seven steps

1. Establish the facts, separately from the interpretation

Write down what is actually known, what is assumed, and what nobody has checked. A surprising number of ethical dilemmas dissolve at this step because the central "fact" turns out to be a rumour, or survive it in a different shape because a detail everyone assumed was benign is not. Note the source of each fact and the date. If a claim cannot be attributed, mark it as an assumption and decide whether it is worth the cost of verifying.

2. Name the ethical question precisely

"What should we do about the supplier?" is not a question a framework can answer. "Should we continue to buy from a supplier whose subcontractor we have credible reason to believe uses underage labour, while an audit is arranged?" is. The discipline is to state the question in a form that contains the conflict, because a vague question invites a vague answer that everyone can sign off and nobody has to own.

3. Identify who is affected, including those with no voice

List the parties: employees, customers, shareholders, suppliers, the communities the operation sits in, regulators, and future employees or customers who will inherit the precedent. For each, note what they stand to gain or lose and whether anyone in the decision is representing them. The parties with no representative in the room are the ones whose interests get quietly discounted, which is why writing them down changes outcomes.

4. Generate genuine options, then apply three lenses

Two options is usually a sign that the thinking has stopped early. Push for at least four, including one that is more costly and more conservative than anyone wants to propose, and one that involves telling someone the truth earlier than is comfortable.

Then test the shortlist through three lenses drawn from the main core ethical principles in business:

Consequences. Which option produces the best overall result, and for whom? Be specific about the distribution, not just the total.
Duties and rights. What have we promised, contractually and otherwise? What rights are in play that a good outcome does not license us to override?
Character. What would an organisation we would want to work for do here? What does this decision make us, if we repeat it?

When the lenses agree, the decision is easy and you did not need the framework. When they disagree, that disagreement is the actual dilemma, and naming it is the useful output of this step.

5. Run the three tests

Before choosing, put each surviving option through three quick checks.

The publicity test. If this decision and the real reasoning behind it were reported accurately, would we defend it? Not "would we survive it", but "would we defend it".
The reversibility test. Would we accept this if we were the customer, the employee or the supplier on the other side?
The generality test. Are we content for this to be the standing rule, applied every time, including when it costs us?

An option that fails a test does not need a better communications plan. It needs to be changed or dropped.

6. Decide, and record the reasoning

Write down the decision, who made it, who was consulted, which option was chosen, and, crucially, what was given up. The record of what you decided not to do is more valuable than the record of what you did, because it is the only evidence that the trade-off was seen at the time rather than reconstructed afterwards.

Name the accountable owner. A decision with no owner is a decision that will drift when the pressure that produced it changes.

7. Implement, then review against what actually happened

Set a date to look back. Did the harms you expected materialise? Did the mitigations get delivered, or did they quietly fall off someone's list once the immediate problem passed? Was the precedent applied consistently when a similar case arrived? Frameworks that stop at the decision produce good minutes and no change. The review is where a one-off judgement becomes an organisational standard.

When every option is bad

Plenty of real dilemmas have no clean answer: a redundancy programme, a product recall that will not fully compensate the people affected, withdrawing from a market and leaving local staff behind. The framework does not manufacture a good option in these cases, and it should not pretend to.

What it does is force the choice to be made explicitly. State which harm you are accepting, who bears it, why it is the lesser harm, what mitigation is being funded, and what would have to be true for a different answer. That is what distinguishes a hard decision taken responsibly from a bad one taken conveniently, and it is what a regulator, a tribunal or a journalist will look for later.

Making it usable rather than ceremonial

A framework that lives in a policy document does nothing. Three things make it operational.

First, a threshold. Define which decisions require the full process, by value, by number of people affected, or by category, so that people are not expected to run seven steps on routine calls and can recognise the ones that matter.

Second, a second pair of eyes that is structurally independent of the pressure. The most common failure mode is that the person with the target is also the only person deciding.

Third, a safe route for someone junior to trigger the process. If the only people who can call a decision ethically difficult are the people who benefit from it not being called that, the framework will never be used on the cases it exists for. Our guide to setting up a whistleblowing policy covers building that route, and writing a code of conduct covers where the standards themselves get written down.

Where to read further

The Institute of Business Ethics publishes practical decision-making tools and survey data on what employees actually observe at work, and its material is written for practitioners rather than academics. For the wider context on why any of this pays, see our pillar on the importance of business ethics and ethical leadership, or start from the E-Business Ethics homepage.

Frequently Asked Questions

What is an ethical decision making framework?

It is a repeatable sequence of questions that forces a decision out of instinct and into the open: what are the facts, who is affected, what options exist, what do our obligations and values require, what would each option look like if it were public, and what will we do to make it stick. The point is not that a framework produces a single right answer. It is that it produces a decision you can explain, defend and review.

Why not just rely on good judgement?

Because judgement under pressure is systematically biased. People discount harms that fall on strangers, overweight the interests of whoever is in the room, and rationalise decisions that serve their own targets. A framework slows the decision down at exactly the points where those biases bite, and it creates a record, which matters when the same question comes back a year later.

What are the three quick tests for an ethical decision?

The publicity test: would you be comfortable if this decision, and your reasoning for it, were reported accurately? The reversibility test: would you accept this decision if you were the person on the receiving end? The generality test: are you content for this to become the rule your organisation follows every time, not just this once? A decision that fails any one of the three needs rethinking, not better wording.

How do the ethical theories fit into a practical framework?

They are lenses, used at step four. A consequentialist lens asks which option produces the best overall outcome. A duty-based lens asks what obligations, promises and rights are in play regardless of outcome. A virtue lens asks what a person of good character would do. Real dilemmas are hard precisely because the lenses disagree, and naming the disagreement is more useful than pretending one lens settles it.

What if every available option is bad?

That is common, and it changes the task. When all options carry harm, you are choosing which harm to accept and who bears it, so the framework's job is to make that choice explicit rather than to find a clean answer. Document why the chosen harm is the lesser one, who is affected, what you did to mitigate it, and what would have to change for the answer to be different. That record is the defensible part.

Who should be involved in an ethical decision?

At minimum, someone who does not benefit from the outcome. The most common structural failure in ethical decision making is that the person under the commercial pressure is also the only person deciding. For anything material, involve a second function, legal, compliance, HR or a non-executive, and record who was consulted.