Delegation in leadership, unique ability, and hiring accountable people: Cameron Herold returns

Asked for his best advice on delegation in leadership, Cameron Herold does not qualify it. Delegate everything. Your to-do list does not have your name at the top of it. It is a list of things that need to get done, and your job is working out who can do them.

Herold is the founder of COO Alliance, the former COO of 1-800-GOT-JUNK, and the author of The Second in Command. This is his return visit, and Michael structures it as a rapid pass through his best material.

The book that seemed inevitable to everyone but him

Herold had already written five books and says he never wanted to be a writer. But six years running COO Alliance and five years hosting a podcast on the second-in-command role left him with a large amount of intellectual property around recruiting, interviewing, onboarding and everything else in the COO space, and not much good published material to point people at. He credits Rocket Fuel by Gino Wickman and Mark Winters as one of the few things doing the job well.

He nearly published it as a two-sided book, printed so that one cover gave the CEO's version, how to hire a COO, and flipping it over gave the COO's version, how to find the right company. Production complexity killed it, but the content was designed that way, so every page had to be valuable to both sides of the relationship.

Two weeks after launch, the book brought in a COO Alliance member. His reaction: he should have written it six years earlier.

What he wishes he had known at twenty

At twenty, Herold had twelve full-time employees and six houses being painted simultaneously.

What he wishes he had understood is how to block his time. His memory of that summer is his friends, several of whom had moved to work for him, playing basketball in his driveway at the end of the workday while he stayed inside running the business.

His diagnosis is worth hearing from someone who built what he built. A lot of entrepreneurs and CEOs tell themselves the business is their hobby and they enjoy it, and at the end of the day they are no longer fun people. Looking back at that summer, he was not a fun person.

What he knows now that he did not then: he will never get it all done and never catch up. He can delegate more. He can say no more, because no is a yes and yes is a no. He can work on the critical few instead of the important many. And perfect does not scale.

Unique ability, and delegating everything except genius

The delegation instruction is absolute: keep delegating and growing people. If nobody has the skill or the confidence for something, delegate it to someone anyway and help them build both.

The filter for what stays with you comes from Dan Sullivan of Strategic Coach, and Herold's version of the exercise is concrete.

Imagine someone follows you with a camera for a month and records everything you do. At the end, you write down every activity, and you will have perhaps eighty. Then you sort them into four categories.

I for incompetent, meaning you are bad at it. C for competent, meaning you are adequate. E for excellent, meaning you are genuinely good at it but get no energy from it. And U for unique ability, meaning you are good at it and it fires you up, and you would do it for free except your children need to eat.

The instruction that follows: delegate everything except genius. Get everything off your plate except the unique ability work.

His own day at the time of recording was two media interviews, coaching a CEO, and lunch with a CEO and COO who are clients. Everything else on his list is busy work that other people can do.

And crucially, this is not a founder's privilege. He is explicit that the head of finance can do it, the head of marketing can do it, and a frontline manager can find ways to say that they are bad at certain things and would rather work on the projects they are good at.

How he prioritizes

Two filters.

The first he calls the low PITA factor, meaning the things with a low pain-in-the-neck cost to do that will pay dividends over a long period. His analogy: how much energy does a satellite need to orbit the earth? Once it is in orbit, none. It goes around forever for free. So what can you put in orbit?

The second is anything that generates revenue or gross margin, on the grounds that there is not a single problem that writing a check cannot solve. Better systems, better playbooks, better technology, all of it costs money that revenue pays for. So revenue-generating projects come first.

Hire accountable people

Twenty years ago Fortune magazine asked Herold how he holds employees accountable. His answer was that he does not. He hires accountable people.

The recent version: a CEO told him they were looking to hire a project manager to manage their four main people. His response was to ask why they did not teach those four people how to manage projects.

The consistent thread through his career is growing the skills of the people already there. Two years before the recording he launched a course putting every manager and leader through twelve core modules, specifically to grow both skill set and capacity.

The business area review

The mechanism Herold credits with raising the game at a 900-person internet company is a quarterly meeting he calls the business area review.

The head of each of nine business areas presented to the board and the leadership team: the plan for the quarter, how the previous quarter went, the metrics, and any pain points. Then they were grilled for thirty to sixty minutes.

The reason it worked is that everyone stayed in the room. Over eight hours, you watched every business area go through it, and everyone got their turn on both sides of the table.

Michael asks how you keep that intense without making people miserable. Herold's answer is that they were glad to do it, because it visibly raised everyone's performance and it was not something being done to them. His comparison: high-performing athletes want their coach to encourage them, and they also want the coach to say they are getting something wrong. It has to be both.

He brought the same rhythm to 1-800-GOT-JUNK once head office reached a couple of hundred people.

How the feedback actually ran

The structure is the transferable part.

The person presenting goes first. They are asked to name three things they should continue doing next quarter and three areas they should improve. The coaching principle underneath it is that the person being coached has to do the introspection themselves.

Then the leadership team offers their own three and three. Always starting with what is going well.

Two things made it safe. First, that everyone had areas to grow, including him. His framing is an athlete's or a performer's mindset, where the best in the world are always trying to improve, and once that is the culture, feedback feels normal.

Second, and more structurally, they ran a no-blame environment built on Michael Gerber's idea from The E-Myth that people do not fail, systems fail. If something went wrong, it was a broken or missing system rather than someone's fault, so feedback was always tied back to the underlying system.

And the reassurance that makes candor possible: nobody spends an hour giving you feedback and then fires you. You are only getting feedback because they want you here.

Promote from within, or hire out

Herold's test for whether an internal candidate can become COO has three parts.

First, the COO needs to be good at everything the CEO is bad at. So look internally and ask whether you have someone strong in exactly those areas, which you will know from what they have already been doing.

Second, do they have the leadership skills? Situational leadership, coaching, delegation, interviewing and hiring, growing people, managing conflict, communication. The soft skills.

Third, and more strategic: are they the person you cannot afford to lose? Promoting them into the role can be a way of keeping them for the next five years. Because the alternative calculation matters. You could bring in a better person externally and in doing so upset seven people internally and lose two or three of them. Sometimes the upside is smaller than the downside.

His one structural rule: if your business carries a lot of technical depth, engineering, architecture, software, anything requiring deep understanding, you are usually better promoting from within. It takes an outsider twelve to eighteen months to learn the whole encyclopedia of your business.

Bob and weave

Asked how you hold to an annual plan while staying flexible, Herold describes what he calls a bob-and-weave culture. You cannot be so rigid that you break.

His image is water flowing down a mountain. It follows the path of least resistance and still arrives at the bottom, which is exactly where it was supposed to go.

The constraint is not the plan, it is the boundaries. He describes core values on one side and core purpose on the other, like the walls of a bobsled track, with the vivid vision at the end. You will get there, and the path will meander.

The failure mode he names is the alternative: enforcing this is how we have always done it, which mostly frustrates people.

Sell them, don't tell them

For introducing change into an organization with entrenched ways of working, Herold draws on building the group now known as Gerber Collision, where he was second in command through a run of acquisitions.

The CEO's phrase was sell them, don't tell them. They were acquiring auto body chains and keeping the existing owners in place, but those owners now had to operate inside the acquirer's systems. So rather than mandate it, they showed the benefits and the opportunities and kept nudging.

His explanation for why the mandate fails is the best line in the episode. We are all sixteen-year-olds trapped in older bodies. Because I am your mother stops working when a child is about seven, and the business version, because I am the CEO or because we have always done it this way, produces exactly the pushback you would expect from a teenager.

Systemized gratitude

Herold's example of praise done properly is Howard Behar, a longtime Starbucks leader, and Herold heard it firsthand because he was being mentored by Starbucks' COO at the time.

Every Friday, for two hours, Behar handwrote thank you notes to people and stores across the company's thousands of locations.

The operational detail is what makes it work. Behar did not have to know who deserved one. Every Friday morning there was a list on his desk of about a hundred stores and people, with the notes stacked and addressed. All he did was write. Congratulations on the record week. Congratulations on selling more of something than anyone else in the system.

The chief executive of one of the largest companies in the world spending 5% of his week saying thank you. Herold says it is the only genuinely systemized gratitude he has seen.

His broader point, drawing on The One Minute Manager and Ken Blanchard's work with Paul Hersey on situational leadership: he has never seen a leader praise their people too much, or say thank you too often, or celebrate the values too much. And the reason to look for things to praise is that it earns you the standing to give feedback and areas to improve.

Playing with AI, and the Monday book report

Herold's position on AI is unambiguous: not a single employee's job is at risk because of AI, unless they refuse to use it. Embrace it and it becomes a tool.

His analogy is typing. Thirty years ago you had to learn where the keys were. Children today never learn that, because they have been using keyboards from the start. It became infrastructure.

The mechanism he now recommends to COO Alliance members is simple and repeatable. Tell every employee to spend one to two hours a week playing with AI tools. Then, at the Monday team meeting or the business area meeting, each person gives a five-minute book report: here is a tool I tried, here is what I used it for, here is how it is going, or here is why it flopped.

He points people at directories cataloguing thousands of AI tools across thousands of tasks, on the grounds that most people talking about ChatGPT are talking about one of them.

Michael's version at Tucows is a Slack channel where people show how they are using it. Herold's team records short screen-capture videos and shares those, on the principle that the way to educate everyone about a new tool is to teach each other.

The specific tools they trade in the episode are worth noting. Herold on Descript, which can remove coughing and filler words from a finished video and, by editing the transcript, change a spoken word so the voice says something else. Michael on the tooling that generates his show notes, summaries, episode titles and social posts, which he says has cut hours and saved real money.

The COOs he admires

Sheryl Sandberg, whom he calls the queen bee of COOs, for coming in as the experienced operator alongside a twenty-three-year-old founder and staying fifteen years. His view is that she does not get enough credit for what Facebook became under her leadership.

Rippling's COO, for publishing an operating manual to himself: a two or three page document given to every employee on day one describing how to work with him, what irritates him, how to interrupt him in a meeting, how to give him feedback, how to ask for help. Herold's point is that this compresses something that normally takes a year or more.

And Eric Church, who succeeded him as second in command at 1-800-GOT-JUNK. Herold took the company from $2 million to $106 million. It then fell to around $70 million through the financial crisis and, in his telling, the wrong second in command. Church came in at $70 million and took it to $450 million.

What Herold finds interesting is how different Church is from him. Church never talks to the media and never speaks at events. Herold was the outward-facing sales, marketing and culture COO. Church is inward-facing, focused on process and people, and indifferent to accolades or building a personal brand.

Herold had to ask him personally to contribute anything to the book, which Church did only because they have been friends for thirty-five years, having founded a fraternity together where Herold was president in year one and Church in year two. Which means Church has now succeeded him twice.

Hire someone you are willing to listen to

Herold's best piece of advice came out of nearly bankrupting a company at around $60 million in revenue.

Their head of finance was quiet and amiable. He kept coming to them asking whether they were sure they were not growing too quickly, whether they were spending too much. And they kept telling him they had it.

When they nearly went under, they realized they had not been listening to him. The CEO had to borrow several hundred thousand dollars from his mother to make payroll.

The lesson Herold took: if you are not willing to listen to somebody on your team, hire someone you are willing to listen to. Because you do not have all the answers, and if the person raising the alarm cannot get your attention, the problem is the arrangement rather than the alarm.

How he learned to listen

His method is physical, and slightly odd, and he thinks that is the point.

He sits on his hands.

Herold describes himself as a high quick-start personality who talks with his hands and generates a lot of nervous energy. If the energy has nowhere physical to go, something changes. And, more prosaically, it reminds him to stop talking.

Michael's version is a piece of art he generated using an image model, a stop sign floating in space splattered with paint, which he had framed with the words "and think" added, and keeps in his sight line.

The leadership question

Herold closes with the story his Starbucks mentor told him, and Michael calls it a mic drop.

Howard Schultz called and asked why the letter B on the sign at a particular intersection in Seattle was not working.

The mentor's answer was that he did not care why that letter on that sign was not working. Schultz said he had better care. The mentor said it was not a leadership question.

Asked what the leadership question was, he said: what system do we have in place, or what system is missing, to ensure that every letter on every sign at all of our locations is always working? That, he said, is a question I will answer.

Herold's conclusion is the one to keep. Leaders should look for the systemic question rather than the symptomatic one. Do not answer the symptom, answer the root cause.

And the payoff is cultural rather than procedural. Once people know that a failure is not going to be pinned on an individual, they will tell you when letters are out.

The 5 things I took away from this conversation

1. Your to-do list is not a list of your work. Cameron's reframe is the sharpest statement of delegation I have heard. It is a list of things that need doing, and your job is figuring out who does each one, including handing work to people who are not ready and helping them get there.

2. The four-letter sort is a practical exercise, not a philosophy. Incompetent, competent, excellent, unique ability. The distinction that does the work is between excellent and unique: the things you are genuinely good at but that drain you are the ones you never think to delegate, and they are exactly what should go.

3. Have the person grade themselves first. Three things to continue, three to improve, from them, before the leadership team says anything. It changes the whole character of the review and, as Cameron says, opens people up to hear more.

4. Ask the systemic question, not the symptomatic one. Not why is that light out, but what system would make sure lights are never out. I keep catching myself asking the first version, and it produces a fixed light and nothing else.

5. Give people two hours a week and a five-minute book report. The cheapest AI adoption programme I have heard. No strategy document, no vendor selection. People try things, report what worked and what flopped on Monday, and the whole company learns at once.

FAQ

What is delegation in leadership? Assigning work to others in a way that develops them rather than simply offloading tasks. Herold's version is absolute: delegate everything, and where nobody currently has the skill or confidence, delegate it anyway and help them build both, on the basis that this is how organizations scale.

How do you decide what not to delegate? Use the unique ability filter from Dan Sullivan's Strategic Coach. Sort everything you do into four categories, and keep only the work you are both genuinely good at and energized by. Notably, this means delegating things you are excellent at but that drain you.

How do you hold employees accountable? Herold's answer is that you do not, you hire accountable people. His related move is to invest in growing existing people's leadership skills rather than layering additional managers above them to enforce accountability.

Should you promote a COO from within or hire externally? Look for someone strong in exactly the areas the CEO is weak, who has the leadership soft skills, and whose loss would damage the company. Herold also notes that businesses with deep technical intellectual property usually favor internal promotion, because an outsider needs twelve to eighteen months to learn the business.

How do you introduce change to a team set in its ways? Sell them, don't tell them. Herold's experience across a run of acquisitions was that showing people the benefits and nudging them works, while asserting authority produces the same pushback you would get from a teenager, regardless of how senior everyone involved is.

Also mentioned

• COO Alliance and The Second in Command, Herold's book on the CEO and COO relationship
• Rocket Fuel by Gino Wickman and Mark Winters, one of the few prior books on the role
• Strategic Coach and Dan Sullivan's unique ability framework
• The E-Myth by Michael Gerber, source of the people do not fail, systems fail principle
• The One Minute Manager, and Ken Blanchard and Paul Hersey's situational leadership
• Howard Behar's Friday morning thank you notes at Starbucks
• Descript, and the transcript-level video editing Herold demonstrates

Listen to the full episode

Cameron Herold on Between Two COO's

Between Two COO's is hosted by Michael Koenig. Subscribe on Apple Podcasts, Spotify, or wherever you listen.