Leading a Horse to Water

I'm doing a master's in organizational leadership, and one of the first things I've learned is that leadership is not a status only found in specific roles, it's something we all do every day. Leading projects, leading people, making decisions on next steps, risk assessments, and practicing your social skills. One such example I have been thinking much about recently is providing feedback - generally, but also when you give the same feedback multiple times to someone. You find they nod, say it was fair, thank you for being direct, and then do the thing again the following week.

It's something perplexing: you are specific in the feedback given, offer help, and mean it. You write the steps down, in order, in an email they could find again. You ask if there was anything in the way, and nothing changes.

As the saying goes: you can lead a horse to water. But that hardly solves the problem, it just describes it. How many times must you take the horse to water? Should you fill up a bowl and raise it to its mouth? A PowerPoint on "drinking water"? Should you bring in a horse expert to assist?

The Proverb

It is the oldest English proverb still in daily use, recorded as early as 1175 in the Old English Homilies:^[1]

Hwa is thet mei thet hors wettrien the him self nule drinken?

— Old English Homilies, c. 1175 (who can give water to the horse that will not drink of its own accord?)

John Heywood reshaped it into the version everyone knows by 1546. This is not a new problem among people.

The drinking is not yours to do, and no amount of leading better or putting in more effort as a leader will make a horse drink. It can be deeply irritating when the horse is now standing in the river up to its knees, you're gesturing at the water, and it just stares at you.

The version I like adds a third clause. Charlie "Tremendous" Jones, an American insurance-agency-executive-turned-motivational-speaker best known for the bestseller Life Is Tremendous, used to put it this way:

You may not be able to make a horse drink, but you can put salt in his oats and make him thirsty.

— Charlie "Tremendous" Jones

You cannot make the horse drink, but you CAN make it thirsty, you can check the water is fit to drink, and you can find out whether the animal in front of you is a horse that isn't thirsty or a horse that is ill.

First, check the water

Before anything else: was the feedback actually constructive? In Kluger and DeNisi's 1996 meta-analysis, which pooled 607 effect sizes from feedback studies, feedback improved performance on average, but in over a third of cases it made performance worse.^[2] The feedback that hurt was the kind that pulled attention towards the self (who you are, how you compare) rather than the task (what to do differently, next time, at this step). No one wants to hear what is wrong with their character, but many are happy to hear how they can improve the technique, the strategy or the steps.

It's always satisfying when a scientific study rings true. I know that I have this problem when I receive feedback: I feel attacked, if it's about my underlying drive or passion, but when I am given feedback on specific actions or tasks, I feel empowered. So, now I try to do the following when giving other people feedback: can they repeat back, in their own words, what they are meant to do differently, by when, and how they will know they've done it? If the answer to any of the three is a shrug, then I did not lead a horse to water, I led a horse to a description of water. "Be more proactive" is a high-level descriptor and nowhere near specific enough.

Second, check whether the horse is thirsty

The second question is whether the person thinks there is anything to fix. Prochaska and DiClemente's stages of change model, built on decades of watching people quit smoking and fail to, puts a stage before "thinking about it": precontemplation, where the person does not see a problem at all.^[3] Advice delivered to someone in precontemplation is like telling someone it will eventually rain in a city fifty miles away, at least to them. The smoking example also rings true for me: I remember begging my parents to stop smoking, even threatening to start myself if they didn't stop. Ultimately, a person needs to want to change a habit, otherwise any well-meaning, well-informed, kind and considered advice is utterly for naught.

"Yeah, fair, thanks for saying" is what a person says when they agree and intend to act, and it is also what they say to make the meeting end while being polite. This means we can't really settle on that as the finale to the feedback conversation. Klein and colleagues' meta-analysis of goal commitment says: goals predict performance when people are committed to them, and barely at all when they are not, and commitment is not the same as having heard the goal read aloud.^[4]

Worse, pushing harder at this stage makes it worse. Miller and Rollnick, the motivational interviewing people, call the helper's urge to fix the "righting reflex," and their central finding, replicated across addiction, health and now management, is that when you argue for change, the other person starts arguing against it, and hearing themselves argue against it makes them believe it.^[5] Brehm named the underlying push-back in 1966: reactance, the instinct to reclaim a freedom someone is trying to take.^[6] You can't tell a horse it's thirsty and expect it to believe you.

Third, check whether the horse can drink

Someone agreeing and then not doing is what inspired me to write this, and it is usually read as a will problem when it is often a capability or environment problem. Landsberg's skill/will matrix is a two-by-two from a coaching book.^[7] Low skill needs teaching while low will needs reasoning, and low both needs direction. Getting these wrong is like trying to inspire a fish to walk.

How to know if it's a "will" problem:

• Have I seen them do it once? If I have never watched the person do the thing, even badly, I do not know whether they can. "Write clearer tickets" assumes they know what a clear ticket looks like. Show them one. Then watch them write one.
• What is paying them to do the opposite? If they are incentivized or otherwise follow a protocol established by a higher authority or by tradition, they may ignore your feedback.
• Is there room? Asking someone drowning in tickets to also improve how they write tickets is asking for the improvement to be ignored.

Salting the oats

Given a horse that is capable, not obviously ill, and not thirsty, here are some of my thoughts and learnings compiled together:

1. Listen to them, don't assume. Ask them to say what they'll do, and write down what they said, not what you wanted them to say. A plan in someone else's words is often very different from what you think they know. Locke and Latham's goal-setting work has been saying this for forty years: participation in setting the goal raises commitment to it.^[8]
2. One behaviour, one fortnight, one checkpoint. "Be a better communicator" is useless. "Send the summary before the meeting, for the next three meetings, and we'll look at it on the 20th" is tangible, measurable and clear - the outcome is "becoming a better communicator" but the steps are clearly stated.
3. If they don't see shortcomings they don't see a problem to fix. If the report is late and you fix it for them at eleven at night, the report is never late from their standpoint.
4. Changing a person's behaviour is difficult. Changing a template, a checklist, a pairing session, a definition of done, is straightforward and easy.
5. "This is the third time we've had this conversation. What's going on?" and let silence reign until an answer comes. Listening is important.
6. "We agreed X by the 20th." A dated, agreed expectation turns the fourth conversation from an argument about memory into a paper trail.

How far to go

A failure of leadership is waiting too long. When behaviours become patterns that have been that way for years, you are fighting both the problem and established tradition. You can always try one more thing, one more framing, one more month. So the leader who doesn't decide in advance ends up deciding by exhaustion, which is late, and usually unkind to the team, to yourself, and to your colleague.

Set the budget before the third conversation. What decides it is not how much you like the person (something I know we all struggle with). It's the cost of letting the issue stand:

• You. The hours, obviously, but also the attention. A stuck person takes up a share of a leader's head that is wildly out of proportion to their share of the team.
• The others. Every hour spent on the horse that won't is an hour taken from the horses that would.
• The standard. What you tolerate for long enough becomes the definition of acceptable. And that can also spread: "Well, Jon takes 3 days on these tasks, why should I deprioritize it too?"
• The person. Every month they spend in a role they are not going to grow into is a month they are not spending somewhere they might. Endless patience inhibits their growth, prevents them from achieving their goals (and possibly stymieing their future income, prospects, benefits, etc.).

Then, how do you read which kind of stuck you're looking at?

1. They try, and fail. They attempt the thing, it comes out wrong, they attempt it again. This is a skill gap and it is - believe it or not - the good kind of stuck. Keep going, teach more concretely, lengthen the horizon. At least the horse is trying to drink!
2. They agree, and don't. Nod, thanks, nothing. This is will or environment, and the response is a single change of approach, followed, if nothing shifts, by the written expectation with a date on it.
3. They dispute the premise. "I don't think that's a problem." Either they're in precontemplation, or you are wrong. Stop advising and focus on listening. Then go and check with somebody who has no stake in it whether the water you're offering is actually water.

When to give up

Giving up on an approach is not giving up on a person. Most of what I've described above is giving up: on explanation, on the plan in your words, on absorbing the consequence. Surrendering is often surrendering the current strategy for a better one.

Giving up on the person is very different. Informal feedback, then explicit written expectation, then a formal process, then parting. The leader's obligation at each rung is the same: the person must never be surprised. If someone learns at a formal meeting that their job is at risk and this is news to them, the leader failed. Additionally, if you're not their leader but are providing feedback, sometimes giving up is giving up your part in the process to someone else - their leader, their mentor, the person themself.

I hate this part.

I'd rather run the sixth conversation than the formal one, and I think it's something many people avoid out of feelings of not wanting to be the bad guy. Ultimately, you can only control that which is your own action, not the outcome. Some wise words by an idol of mine, a physically crippled man who escaped slavery and learned much about our control over our personal destinies:

The things in our control are opinion, pursuit, desire, aversion, and, in a word, whatever are our own actions. The things not in our control are body, property, reputation, command, and, in one word, whatever are not our own actions.

— Epictetus, Enchiridion

You can lead the horse to water, you can make sure the water is clean, and you can show it how to drink, but no matter how much you want to, out of compassion, kindness or care, there is a point where you have to accept that you are not the horse and you cannot force it to drink. The servant leadership test is whether the people you lead grow. If they are not growing with you after a fair run, the kind act may be to help them find where they will, perhaps a sister team, or through a friend you know from college. I personally believe leaders should help colleagues beyond their current role and company, and encourage growth, because paths can cross many times, so help them even if not from inside your own team or organization. And remember, this cannot be a surprise, this is only after a long period of trying and goal setting.

A Flow Chart

References

1. Old English Homilies, c. 1175. John Heywood, A Dialogue Conteinyng the Nomber in Effect of All the Prouerbes in the Englishe Tongue (1546). Discussed at phrases.org.uk
2. Kluger, A. N., & DeNisi, A. (1996). The effects of feedback interventions on performance: A historical review, a meta-analysis, and a preliminary feedback intervention theory. Psychological Bulletin, 119(2), 254–284. 607 effect sizes; mean d = 0.41; over one third of interventions reduced performance.
3. Prochaska, J. O., & DiClemente, C. C. (1983). Stages and processes of self-change of smoking: Toward an integrative model of change. Journal of Consulting and Clinical Psychology, 51(3), 390–395.
4. Klein, H. J., Wesson, M. J., Hollenbeck, J. R., & Alge, B. J. (1999). Goal commitment and the goal-setting process: Conceptual clarification and empirical synthesis. Journal of Applied Psychology, 84(6), 885–896. Meta-analysis of 83 samples. pubmed.ncbi.nlm.nih.gov
5. Miller, W. R., & Rollnick, S. (2013). Motivational Interviewing: Helping People Change (3rd ed.). Guilford Press. The righting reflex, ambivalence, and why arguing for change produces arguments against it. Summary at ncbi.nlm.nih.gov
6. Brehm, J. W. (1966). A Theory of Psychological Reactance. Academic Press. Reviewed fifty years on in Steindl et al. (2015), Zeitschrift für Psychologie, 223(4), 205–214.
7. Landsberg, M. (1996). The Tao of Coaching. HarperCollins. Source of the skill/will matrix (direct, guide, excite, delegate).
8. Locke, E. A., & Latham, G. P. (2002). Building a practically useful theory of goal setting and task motivation: A 35-year odyssey. American Psychologist, 57(9), 705–717.