Challenging A Leader You Advise

To challenge a senior leader you advise, first separate the disagreement into facts, judgement and authority.

Test the facts with three questions, and turn any disagreement of judgement into a dated prediction the leader can check.

Then state your position once, in four parts, put it in writing before the decision, and say what you will do if the decision stands.

The adviser's position differs from an employee's disagreement with a line manager.

The leader is a client who owns the decision, can ignore the advice and has no obligation to explain why.

That leaves two poor responses.

An adviser who implements in silence becomes complicit in an outcome the adviser predicted, and one who argues at every meeting spends the relationship on a single decision.

Three Kinds Of Disagreement

Most disagreements with a sponsor contain three disagreements at once, and mixing them is why the conversation goes badly.

• Facts. One party holds information the other lacks

• Judgement. The same facts support different conclusions

• Authority. The decision belongs to the leader

A disagreement about facts is cheap to resolve.

A disagreement about judgement is legitimate and should be recorded, because neither side can win it by argument.

A disagreement about authority is settled before the conversation starts, and continuing to argue it damages the adviser's standing.

The first step is to say, to yourself, which of the three you are having.

Testing The Facts

Advisers skip this step because they assume the leader holds the same information.

Three questions test that assumption.

1. What have you seen that makes this the right answer

2. What would you expect to be true if it were the wrong answer

3. What has already been decided that I do not know about

The third question resolves a large share of these conversations, because a leader is often working to a constraint that has not been shared.

When the adviser does hold different facts, the way they are put matters.

Kerry Patterson and colleagues advise starting with what was seen and heard, described outside your own head, before offering any conclusion about why.

Conclusions come second and are offered tentatively, because leading with a conclusion tends to produce defensiveness.

Turning Judgement Into A Prediction

Where the facts are shared and the conclusions differ, the argument is about weighting, and no further evidence will settle it today.

The move that works is to make the disagreement testable.

State what you expect to happen if the decision proceeds, in specifics and with a date.

Ask what the leader expects instead.

Agree what evidence would settle the question and when both of you will look at it.

A prediction with a date lets both parties keep their positions without either being wrong yet.

Roger Schwarz and colleagues describe this as jointly designing ways to test disagreements, and they use two scientists designing one experiment to test rival hypotheses as the analogy.

Their prompts are "How might it be that we are both correct?" and "How could we each be seeing different parts of the same problem?"

The Four Part Statement

The adviser gets one clear statement of the position, and it should be prepared.

1. What you agree with, said first and meant

2. What you expect to happen, in specifics

3. What you recommend instead, with its cost

4. What you will do if the decision stands

The fourth part separates advice from obstruction, because it tells the leader the adviser will carry out the decision, and it is the part most often left out.

Peter Block notes that support and confrontation can come in the same sentence.

The support is an acknowledgement that you have heard the client, which does not commit you to agree.

The confrontation then names the difference between how you see the situation and how the client sees it.

Schwarz adds a closing move, which is to explain your reasoning and then invite questions about it.

Patterson also recommends stating in advance what you do not mean and what you do mean, when the other person is likely to hear disrespect.

A Worked Example

A customer operations director has decided to merge two regional service centres by 1 June and asks the HR business partner to plan the move.

The business partner expects the merger to cost the smaller centre its most experienced advisers.

The partner asks the three fact questions and learns that the lease on the smaller site ends in July, which was not in the papers.

The facts now support the merger, and the remaining disagreement concerns judgement about the effect on people.

The statement at the next one to one runs like this.

"The lease date makes a merger this year sensible, and I agree with the timing. I expect at least three of the eight senior advisers at the smaller centre to resign within four months. I recommend a retention payment of £4,000 each to those eight, paid at six months, which is £32,000 and less than the recruitment cost of replacing three of them. If you decide to go ahead without it, I will run the move as planned, and I would like us both to look at the resignation numbers on 30 September."

The figures are an illustration, not a case.

The statement can be checked by both parties on a named date, and it commits the adviser to the implementation.