How to Make Ethical Business Decisions: A 7-Step Framework

Most workplace ethical dilemmas do not arrive labelled. They show up as an awkward request, a discount that feels off, a complaint about a colleague, or a deadline that tempts you to cut a corner. The manager who handles these well is rarely the one with the best instincts. It is the one who runs the same clear process every time, so the decision is consistent, explainable and defensible later. This guide gives you that process: a seven-step ethical decision making framework you can use in a meeting, with a concrete example, the key question to ask, and the common mistake to avoid at each step.

A quick word on where this sits. This is not the same as applying an ethical theory to a single choice. If you want to weigh one option against utilitarian, duty-based or virtue reasoning, our explainer on the main ethical theories in business covers that in detail. The framework below is the broader organisational process a manager runs end to end. The theories are one input inside it, used at the testing step, not the whole model. If you are new to the subject, the pillar guide on what business ethics is sets the foundations.

The 7-step framework at a glance

Here is the whole model in one view, so you can keep it in front of you during a decision. Each step is expanded below.

• 1. Define the dilemma and stakeholders. Name the actual decision and everyone it touches.
• 2. Gather the facts. Separate what you know from what you assume.
• 3. Identify duties, rights and obligations. Find the rules, promises and legal limits in play.
• 4. Generate options. Produce more than two, including ones you do not like.
• 5. Test each option. Run them against ethics tests and the theories.
• 6. Decide and document. Choose, then write down what and why.
• 7. Act, review and learn. Carry it out, check the result, feed it back.

Step 1: Define the dilemma and stakeholders

Start by stating, in one sentence, the decision you actually have to make, and list everyone affected by it. Vague problems produce vague answers. Be precise about the choice in front of you and who has a stake in it.

Workplace example. A sales manager finds that a long-standing client has been overcharged for six months because of a billing error in the team's favour. The dilemma is not "the client is upset". It is "do we proactively disclose and refund, stay silent unless asked, or wait and see". Stakeholders include the client, the finance team, the sales rep involved, the wider customer base who would judge the company if this came out, and the business itself.

Key question: what exactly am I deciding, and who is affected by each option?

Common failure mode: framing the problem too narrowly, usually around your own convenience or the loudest voice in the room, so that affected people are missed and the real choice never gets named.

Step 2: Gather the facts

Ethical mistakes are often factual mistakes in disguise. Before reasoning about right and wrong, find out what is actually true. Separate confirmed facts from assumptions, rumours and gaps, and go and fill the important gaps.

Workplace example. In the overcharging case, the manager checks the contract terms, the exact amount, when the error started, whether it was a one-off or systemic, and whether the client has a clause covering billing corrections. They discover it was a pricing-table error affecting only this account, and that the contract is silent on refunds.

Key question: what do I actually know, what am I assuming, and what do I still need to find out before deciding?

Common failure mode: acting on the first version of events, or on what you wish were true, without checking. A decision built on a shaky fact base is impossible to defend, however good the intentions.

Step 3: Identify duties, rights and obligations

Now map the constraints. What duties do you owe, to whom? What rights do the stakeholders have? What legal, contractual, professional or policy obligations apply? This is where the law and your own code set the floor for the decision.

Workplace example. The manager identifies a duty of honesty to the client, a legal obligation not to retain money taken in error, the company's own code of conduct on fair dealing, and a duty to the finance team to fix the underlying process. The client has a right to accurate billing. These obligations point clearly toward disclosure, before any wider ethical weighing begins.

Key question: what am I obliged to do here under the law, our contracts, our code and our profession, regardless of what would be convenient?

Common failure mode: skipping straight to "what feels reasonable" and missing a hard obligation. If an option breaks a law or a clear rule, it is off the table before you weigh anything else. Your code of ethics exists to make these duties explicit.

Step 4: Generate options

With the constraints clear, produce a real set of options, not a forced choice between two. The aim is to widen the field before you narrow it, so you do not anchor on the first idea or a false either-or. Include options you instinctively dislike, because they sometimes contain part of the answer.

Workplace example. The manager lists: disclose in full and refund immediately with an apology; disclose, refund, and offer a goodwill credit; correct future billing quietly and refund only if asked; or do nothing. Laying them out side by side makes the weak options visible and surfaces a stronger one, full disclosure plus a process fix, that a yes or no framing would have hidden.

Key question: what are at least three or four genuinely different things I could do, including the uncomfortable ones?

Common failure mode: the false dilemma. Treating the decision as "do the right thing or protect the business" when a third option usually does both. Narrow choices produce narrow outcomes.

Step 5: Test each option

This is where ethical reasoning does its work. Run each surviving option through a set of tests. Two layers help. First, a quick set of plain questions. The Institute of Business Ethics promotes a simple three-part test built around transparency, effect and fairness: would I mind others knowing what I decided, who does the decision harm, and would those affected consider it fair. Second, the formal theories: the outcome (consequentialist) view, the duty (deontological) view, and the character (virtue) view. Where they agree, you have a strong option. Where they disagree, you have found the real tension to resolve.

Workplace example. "Do nothing" fails the transparency test at once, because the manager would not want the client or the public to know. Full disclosure and refund passes all three plain tests and is supported by the duty view and the outcome view alike, since the short-term cost is outweighed by retained trust. The theories converge, which is a good sign. For how to apply each theory to an option, use our guide to ethical theories in business, and check options against your core ethical principles too.

Key question: would I be comfortable if this option were reported on the front page, who does it harm, and do the outcome, duty and character views point the same way?

Common failure mode: reverse-engineering a test to justify the answer you already wanted. The tests only work if you are willing to let them rule out your preferred option.

Step 6: Decide and document

Make the call, then write it down at the time. Documentation is not bureaucracy. It is what turns a private judgement into a defensible decision, and it forces clarity about your own reasoning. For anything beyond a trivial, reversible choice, a short written record protects everyone.

Workplace example. The manager chooses full disclosure, a refund and a process fix, and records it in a few lines: the date, the dilemma, the key facts, the stakeholders, the options considered, the decision, the main reasons, who approved it, and a date to review the billing fix. If the client later disputes the timeline, or an auditor asks, the record speaks for itself.

Key question: if a regulator, auditor or new manager read my note in a year, would they understand what I decided and why, without needing me to explain it?

Common failure mode: deciding well but recording nothing, or writing the note after the fact to fit the outcome. Contemporaneous notes are credible; reconstructed ones are not.

Step 7: Act, review and learn

A decision is not finished when it is made. Carry it out properly, then come back to check what happened and feed the lesson into how the organisation works. This is the step most people skip, and it is the one that stops the same dilemma recurring.

Workplace example. The manager makes the refund, apologises to the client, and asks finance to correct the pricing table and add a check so the error cannot repeat. A month later they confirm the fix held and that no other accounts were affected. The episode becomes a short anonymised example in the next team ethics session.

Key question: did the decision work as intended, and what needs to change in our process, training or controls so we handle this better next time?

Common failure mode: treating the resolved case as closed and never fixing the system that produced it, which guarantees the dilemma comes back.

Make it a one-page checklist

The framework is most useful when it stops being a guide and becomes a habit. Turn it into a one-page checklist that lives in your decision log or on the meeting room wall. A workable version asks, in order: What is the decision, and who is affected? What do I know versus assume? What duties, rights and laws apply? What are my three or four options? Which option passes the transparency, effect and fairness tests and the theory views? What did I decide, why, and who approved it? When will I review it? Run every non-trivial dilemma down that list and the quality of your decisions stops depending on the mood of the day.

For an external reference point, the Institute of Business Ethics knowledge hub sets out its own simple ethical test and its "Stop and Think" model, which sit comfortably inside the testing step above.

Frequently Asked Questions