How do you delegate without losing control of the quality?

You delegate without losing control by handing over the standard along with the task. A procedure lists the steps. A standard says what good looks like, what the business refuses and which calls are theirs, so they can make a good call when the steps run out. The US Army built its command doctrine on that idea.

Why does the work keep coming back to your desk?

You did the things you were told to do. You hired help, maybe several people, maybe a virtual assistant or two in another time zone because someone on a podcast made the numbers sound irresistible. You wrote the SOPs, or paid someone to write them, and they live in a shared folder with sensible names. And still, most weeks, the work lands back on your desk with a note that amounts to "can you take a look," and when you look, it's not quite right, and you fix it, and you tell yourself that nobody else has your eye.

That last sentence is where most founders stop thinking, because it feels like humility and it's also a little flattering. A thread in one of the agency owners' forums this fall made the case out loud: a design founder's taste, eye and creativity can't be written into an SOP, so design founders end up the most tangled in their own businesses. One owner in the replies described where that road ends better than I could: a founder freelancing with extra steps and a nicer website.

The thread is right that a founder's taste can't be written into a procedure. I'd push on what it concludes from that. If you're building a team, the business should look less and less like the personality of the founder as it grows. Some things do have to stay consistent, mainly the messaging, the positioning and the standard of the work, and consistency means those are written down somewhere. Everything else, the way someone explains an idea, runs a meeting or makes the small calls along the way, is allowed to stop sounding like you.

Your team won't communicate exactly the way you do, and that's the whole point of building one. The collective whole gets better than any one person in it, founder included. I think founders need to get out of their own way on this one, a little bit.

What do cheaper hands and better procedures actually fix?

Labor, mostly. An extra pair of hands, local or offshore, takes on tasks: the scheduling, the first draft, the takeoff, the data entry. A good procedure makes those tasks more consistent, and that matters. Atul Gawande's reporting on surgical and ICU checklists found that "checklists established a higher standard of baseline performance" (The New Yorker), and a service business with no written procedures at all is leaving that baseline on the table.

The trouble is that most of the work landing back on your desk came back because somebody hit a situation the procedure didn't cover (a client who wanted something slightly different, a site that didn't match the drawings, a brief with two readings) and had nothing to decide with except a guess about what you'd want. So they guessed, or they asked, and either way it ended up with you. Hands are easy to hire, even cheap ones. What's short is a shared picture of what good looks like, and no amount of procedure supplies that.

I covered the other half of this recently: why founders who care about quality end up hovering, and what it takes to mentor someone into running the business. This post is about the artifact in the middle, the thing you actually hand over.

What's the difference between a procedure, a standard and intent?

Three different tools, and most founders only ever build the first one.

| | What it tells the person | Where it works | Where it breaks | |---|---|---|---| | Procedure | The steps, in order | Repeatable tasks where the situation rarely changes | The moment the situation differs from the one it was written for | | Standard | What good looks like here, and what we refuse | Quality calls: is this finished, is this ours, would we send this | When it's so vague ("high quality") that nobody can check against it | | Intent | Why the work exists and the result it must produce | Situations nobody predicted | When it's never said out loud, so people can't use it |

The Army's definition of intent is the clearest one I've found anywhere. Its doctrine describes commander's intent as "a clear and concise expression of the purpose of the operation and the desired military end state," written so that subordinates can "act to achieve the commander's desired results without further orders, even when the operation does not unfold as planned" (ADP 6-0, 2019). The same document explains why the military bothers: "No plan can account for every possibility, and most plans must change rapidly during execution." And the payoff it claims is the one every founder wants: intent "gives subordinates the confidence to apply their judgment in ambiguous situations because they know the mission's purpose and desired end state."

Swap "operation" for "project" and that's a service business. Your procedure is the plan. It will not survive contact with a real client. What survives is whether the person on the job knows what the work is for and what good looks like, well enough to adjust without calling you.

Who has actually written judgment down?

A few famous companies have tried, and the details of how they did it are more useful than the slogans people quote from them.

Nordstrom has long handed new employees a card with one rule on it: "Use best judgment in all situations. There will be no additional rules" (Signal v. Noise). It's usually told as a story about trusting people over writing rules, and a CBS MoneyWatch writer later pointed out that the card leaves something out: there's also a policy guide of about 7,344 words that employees are bound by (CBS News). I'd call that a better story than the myth. The procedures exist. The one-line standard sits on top of them and tells people what to do when the procedures run out.

Ritz-Carlton went further and attached a number to judgment. Horst Schulze, the company's co-founder and first president, has described giving every employee, down to the dishwasher, a budget of up to $2,000 to make a guest's experience right. Franchise owners threatened to sue him when he introduced it. "So far, we have never had to max out the budget," he said. "A plate of cookies or a lunch has sufficed" (Chief Executive). The dollar figure is the standard made concrete: you may decide, up to here.

Netflix calls its version "context not control," and asks managers to give their teams "the context and clarity needed to make good decisions instead of trying to control everything themselves." Then it adds the sentence people tend to skip: context not control "should not be confused with hands-off management," and managers still need to be involved and coaching (Netflix). The standard changes what you spend your attention on, and you're still in it.

And Jeff Bezos gave the most practical sorting rule for which decisions to hand over. Some decisions are one-way doors, "consequential and irreversible or nearly irreversible," and deserve care and consultation. Most are two-way doors, "changeable, reversible," and "can and should be made quickly by high judgment individuals or small groups" (Amazon 2015 shareholder letter, SEC filing). The founder who reviews everything is treating every two-way door as a one-way door, and that's what the backlog on your desk is made of.

A landscaping crew, an accounting firm and a four-person design studio will write very different pages, so take the principle from these examples and leave the card where you found it.

Why is a quality standard really a positioning document?

Look at what goes on the page: what this business is for, who it serves, what good looks like for them, what we refuse to do. That's the same set of decisions that sits under your positioning, aimed at your team instead of your market. When a founder can't delegate the judgment calls, the usual cause underneath is a business that has never said what it's for, not even to itself. The only place that answer lives is in the founder's head, so the founder is the only person who can check the work, and so the founder checks all of it.

That's an architecture problem, and it doesn't get fixed by hiring. A new person, however talented, can only make good calls on the business's behalf if the reasons behind those calls have been put somewhere they can reach. It's also why the page tends to pay twice. The standard you write so your team can make decisions is the same one a good client wants to hear before they hire you, and the same one that lets the wrong client see early that you're not for them.

I learned a version of this in youth ministry. Most of our volunteers were young adults working with teenagers, and their default, understandably, was to act like a peer. I couldn't write a rule for every situation a volunteer would walk into, so I leaned on one question: are you thinking like a parent? That question gave them a standard to check themselves against and left the call with them, which a rulebook never could. A good standard for your team can be that short.

Is delegation a talent or a skill?

Some of both, and the research is humbling for most of us. Gallup studied 143 CEOs of Inc. 500 companies and found that those with high "Delegator" talent averaged 1,751 percent three-year growth and generated 33 percent more revenue in 2013, $8 million versus $6 million. It also found that only about one in four employer entrepreneurs have high levels of that talent (Gallup). That's a correlation from a small sample, and I'd read it that way. The part that matters for you is the one-in-four figure: most founders don't delegate well by instinct.

If it doesn't come naturally, it has to come from somewhere else. A written standard is the cheapest somewhere else I know of. All it asks is that you write down, once, what the person you already are would decide, so you don't have to be in the room for every decision.

Write the one-page standard

Do this for one kind of work first, the one that comes back to you most often. A kitchen install, a brand identity, a year-end file, a proposal.

Write the intent in one sentence. Who is this work for, and what must it do for them? If you can't say it in a sentence, stop here and work on that, because nothing below will hold without it.
List three to five marks of "good" in plain words someone on your team could check against. "The client can find the shut-off without calling us." "Every page reads like our best client wrote the brief." Next to each, point to one real past job that met it.
Write what you refuse. Two or three things this business won't send out the door, however tight the deadline.
Sort the decisions. Which calls are two-way doors your team makes alone? Which are one-way doors that still come to you? Put a dollar or time limit on the two-way doors if it helps, the way Ritz-Carlton did.
For the next two weeks, don't fix returned work yourself. When something comes back, point to the line on the page that covers it. If no line covers it, write the missing line together, then hand the work back.

Sources

Frequently asked

How do I delegate without losing control?

Hand over the standard along with the task. Write down what the work is for, what good looks like, what you refuse, and which decisions your team can make alone. Then review against that page. When the standard is clear enough to check against, you keep control without checking every piece of work yourself.

What is the difference between an SOP and a standard?

An SOP, or standard operating procedure, lists the steps for a task. A standard describes the result: what good looks like, what the business refuses, and why the work exists. Procedures work for repeatable tasks. Standards are what people need when the situation doesn't match the procedure, which is where most returned work comes from.

Can you delegate creative work or work that depends on taste?

Yes, if the taste is written down as a standard. Taste resists step-by-step procedures, but it can be described in marks of good, real examples of past work that met them, and a short list of things you'd never send out. Someone else can check their work against that page without ever matching your eye, and the work may end up better for it.

How do I stop being the bottleneck in my business?

Find the decisions that keep routing through you and sort them into reversible and irreversible calls. Hand the reversible ones to your team with a written standard and, if it helps, a spending or time limit. Keep the irreversible ones. Most founders who review everything are treating reversible decisions as if they were permanent.

Ready to look at the architecture honestly?

If you've hired the help and written the procedures and the work still ends up on your desk, the gap is usually a decision the business never made about what it's for. We'll tell you what we see, and whether the work we do fits where you are.

Book the conversation →