Ask a founder what’s holding the business back and you’ll usually get a confident answer. Ask the rest of the leadership table and you’ll often get a different answer from each person, with a different fix attached to each.
That’s normal in founder-led scale-ups, typically somewhere between 15 and 75 people, where the business has proven itself but the old ways of operating have started to creak. Below you’ll find why symptoms mislead, a quick test for how big a problem really is, a diagnostic you can run yourself and what a good diagnosis should leave you with.
Why is it so hard to see what’s actually wrong?
Because the problems you notice are rarely the problems you have.
When founders describe what’s going wrong, the list tends to sound similar. Work gets duplicated. Things fall through the cracks. People seem reluctant to make decisions. Everyone communicates constantly and still nobody knows what’s going on.
Those feel like four separate problems, but often each is a symptom of the same underlying condition, typically a business that has outgrown the way it was organised. The informal set-up where everyone pitches in on everything stops coping long before anyone decides to change it. Because growth happens one hire and one client at a time, nobody notices the moment the threshold is crossed.
There’s a second reason too. You’re inside the system you’re trying to see. The business has adapted around you over years, and many of its workarounds are invisible precisely because they work well enough. You don’t notice the founder-shaped gap in a process when you’re the one filling it every day.
What’s the difference between a symptom and a root cause?
A symptom is what you can see. A root cause is what’s producing it.
Missed deadlines are a symptom. The root cause might be that nobody owns the handover between sales and delivery. Slow decisions are a symptom. The root cause might be that decision rights were never moved away from the founder. A leadership meeting that’s all updates and no decisions is a symptom. The root cause might be that the team has titles without authority.
The test is simple to state and hard to apply. If you fixed this thing completely, would the pain go away? Or would it reappear somewhere else in a different form?
That second outcome is common. Fix one function in isolation and the pressure often just moves to the next weakest part of the business. We look at that pattern in more detail in you fixed sales, the bottleneck just moved.
A quick fix on a symptom can still buy you breathing space. But if you keep treating the same symptoms quarter after quarter, that’s a strong sign you haven’t found the cause.
Why does my leadership team argue about what’s wrong?
Because each of them is describing the part of the problem they can see from their seat.
Ask your sales lead why growth has stalled and you’ll hear about delivery capacity. Ask delivery and you’ll hear about sales overpromising. Ask the founder and you might hear that the leadership team isn’t stepping up. Ask the leadership team and they might say, more carefully, that decisions still wait for the founder.
None of them is wrong. None of them has the whole picture.
Leadership teams often think they disagree about the problem when they actually disagree about the order. Most founders and senior teams agree on where the business is heading. They split on what to do first. That’s a sequencing argument, and it’s much easier to resolve once everyone can see the same evidence.
It’s also worth watching for false agreement. A team can nod along to a neat summary of “the problem” in a meeting and still hold quite different views privately. A useful check is to ask people to explain the agreed problem in their own words, without the document in front of them. If the explanations wander, the agreement was thinner than it looked.
How do you tell a narrow problem from a wide one?
Some problems are exactly what they appear to be. An invoice is stuck. A proposal template is out of date. Nobody chairs the weekly ops call. You have the information you need, a decision will settle it and you can fix it from where you’re sitting. These are narrow problems.
Others look different from every seat in the business. Their shape depends on who’s describing them. A manager who “won’t delegate” looks very different to the founder, to the manager and to the person who supposedly isn’t being given anything. Fixing it means understanding the history, who trusts whom and the unwritten rules. These are wide problems.
The trap is that wide problems are good at passing for narrow ones. They turn up as a single, nameable issue. So founders, who are wired for action, give them the narrow treatment. They write a new process, tighten a job description or have a stern word in a one-to-one. The fix holds for a few weeks, then the problem comes back somewhere else.
Narrow problems reward speed. Wide problems punish it.
The “ask five people” test
The simplest way to tell them apart is to ask five people in the business to describe the problem.
- If the descriptions broadly match, it’s probably narrow. Fix it quickly and move on.
- If their descriptions start to diverge, it’s wide. Slow down before you fix anything.
A few other clues help:
- Is this a repeat? If it’s turned up before in another form and you keep hearing “we tried that already”, assume it’s wide. Genuinely new problems are more often narrow.
- Does a fix in one team fail in another? A fix that works well in one team and does nothing in a similar one probably only touched the surface.
- Is there something nobody is saying out loud? It might be an old tension, a gap between who has the title and who’s really trusted, or a memory of what went wrong last time. Wide problems usually carry something like that just below the surface.
What is the problem standing in for?
Some of the hardest problems to diagnose are the ones dressed up as system problems.
Founders often open with lines like “nobody knows who owns what”, “we need a better meeting rhythm” or “our structure doesn’t work any more”.
Sometimes those statements are exactly right, and plenty of founder-led businesses are held back by the lack of clear roles and a steady rhythm.
But sometimes the system is standing in for something harder. “We need clearer roles” can mean “someone needs to change how they work and I haven’t said so yet.” Redesigning the operating model feels productive. It’s visible and structured and nobody has to be uncomfortable. But putting off a hard conversation with one person by redesigning the system around them is still putting it off.
Two questions help here. First, if this were already fixed tomorrow, would something still need to change? If the answer involves a particular person, that’s worth noticing. Second, when you or your team describe the problem, is there a name or a role that comes up once and then never again? That’s often where the real issue lives.
How do you run your own diagnostic across people, practices and technology?
A business operating system is how your people, practices and technology work together as an integrated whole. Looking at all three, and at how they connect, is the most reliable way to move from symptoms to causes. (There’s more on the model in what is a business operating system.)
Here’s a simple version you can run yourself over a few weeks.
1. Write down the symptoms, without explaining them
List everything that’s frustrating you about how the business runs. Don’t diagnose yet. “Projects overrun.” “I’m in every pricing decision.” “Onboarding takes too long.” Aim for ten to fifteen.
2. Ask five people
Pick five people from different parts and levels of the business. Ask each the same open questions. What gets in the way of doing your best work? Where does work wait? What would you change first? Listen for where their stories match and where they diverge.
3. Look through the people lens
- Who owns each important outcome, not just each activity?
- Which decisions can people make without asking you, and do they know that?
- Where does the business depend on one person’s knowledge?
- Do people have the competence to take ownership of what you’re asking them to own?
4. Look through the practices lens
- How does work actually flow from first conversation to happy client? Map it, not the org chart.
- Where are the handoffs, and what gets lost at each one?
- What’s the regular rhythm of meetings and reviews, and does it survive a busy month?
- How are decisions recorded, so people know why something was decided as well as what?
5. Look through the technology lens
- How many tools hold versions of the same information, and which one is trusted?
- Can people see the numbers that matter for their own work without asking someone?
- Which tools exist because nobody ever switched them off?
6. Group the symptoms by cause
Go back to your list from step one. For each symptom, ask which of the findings above would explain it. You’ll usually find that most symptoms cluster around two or three causes. Those are your candidates.
7. Put them in order
You can’t work on everything at the same time, so sequence matters. For each cause, ask: is this narrow or wide? How much of the rest depends on it? Is it reversible if we get it wrong? Fix the thing that unblocks the most first.
A word of caution. Resist the urge to reach for a ready-made solution at this point. Choosing a framework before you understand the problem is how many businesses end up running a system that doesn’t fit them. We explore that in why off-the-peg operating systems don’t fit every business.
What should a good diagnosis give you?
A good diagnosis leaves you with three things you can act on.
- An honest account of how work really gets done. Who actually decides, where work actually waits and which workarounds hold things together. It rarely matches the org chart.
- A prioritised roadmap. The root causes, in the order you should tackle them, with the reasoning visible so your leadership team can own it.
- A baseline. Enough measurement of where you are now that, six months from now, you can tell whether the work actually got better rather than just different.
That last point is easy to skip and expensive to miss. Without a baseline, every change gets judged on how it feels, and how it feels depends on who you ask.
What can you try in the next fortnight?
Before you commit to a full diagnostic, five small experiments will tell you a lot:
- Run the five people test on your biggest frustration. Ask five people to describe it in a sentence. Compare the answers.
- List the problems that keep coming back. Anything you’ve “fixed” more than once is a candidate for a wide problem or a root cause you haven’t found.
- Ask your team what they’d fix first. Not what’s wrong – what they’d change first if it were their call. Where the answers cluster, you’ve probably found something real.
- Ask what you’re avoiding. Is there a conversation you know you need to have that a new process would let you put off?
- Hold the next big decision for a fortnight. Before you hire, buy a tool or adopt a framework, check it’s aimed at a cause rather than a symptom.
When should you bring in outside eyes?
You can get a long way on your own. Many narrow problems will fall out of the steps above. There are some signs that an outside view will save you time, though:
- The same problems keep returning under different names, however many times you fix them.
- Your leadership team can’t agree on what’s wrong, or keeps agreeing in the room and diverging outside it.
- You suspect you’re hearing a tidier version of the truth. People are understandably careful about what they tell the founder.
- You’re too close to it. If the business still runs through you, you’re part of the system you’re trying to see.
There are two ways Mallard helps at this point. If you want your leadership team to build its own shared view, the Reflect workshop is a half day for the founder and leadership team, up to eight people in all. Guided by a Mallard fractional COO, you work out together what’s holding the business back.
If you want a deeper, independent picture, the Find diagnostic takes 14 days. It looks at 262 operational data points across people, practices and technology, scores 12 operational capabilities and ends with a prioritised roadmap, not a slide deck.
Whichever route you take, the principle is the same. Diagnose before you fix, and make sure the first thing you fix is the thing that’s really holding the business back.
Contact us