A mature firm needs to move beyond informal rules

12 min readPennine Pacific

How a business is run turns out to be measurable. It predicts performance better than most of the things owners spend their time worrying about. That is the finding of the largest body of research on the subject, built over two decades across 35 countries.

An abstract image in two halves. On the left, tangled scribbles and scattered fragments in grey. On the right, the same forms resolved into an ordered grid of interlocking cubes, lit in teal.

What the research actually measures

The World Management Survey has been run since 2004 from the London School of Economics and Stanford. It scores organisations on eighteen everyday management practices, each rated from 1 for worst practice to 5 for best, then averaged into a single score.

The questions behind those ratings are ordinary ones. Is performance tracked continually and shared with staff rather than handled ad hoc? Are process improvements sought out as a matter of routine rather than made only when something breaks? Are performance measures well-defined and known to everybody they apply to rather than vague and private?

The method is what makes it worth reading. Interviews are double-blind, in both directions.

The manager is not told they are being scored, so there is nothing to perform to. The interviewer is not told how the company is actually doing, so a firm already known to be successful cannot be marked generously on the assumption that it must be well run. That second half matters more than it sounds, because it is the usual way this kind of research goes wrong.

The 2021 review of the project, published in the Oxford Review of Economic Policy, covers more than 13,000 firms alongside 4,000 schools and hospitals.

What a higher score actually goes with

Sort the firms into ten bands by their management score, then compare how each band performs. The higher the band, the better the performance on six separate measures: productivity, operating profit, output growth, exports, research spending and patents. That holds at every step of the scale. Not one of the six reverses anywhere along it.

How one management practice is scored from one to five One of the eighteen survey questions asks how performance is tracked. A score of 1 means no real tracking, where problems surface when somebody complains. 2 means some tracking, informal, held in one person's head. 3 means tracked regularly and reviewed by managers. 4 means tracked and visible to the teams doing the work. 5 means tracked continually, visible to all staff and acted on. Eighteen questions are scored this way and a firm's overall score is the average of all eighteen. HOW ONE OF THE EIGHTEEN QUESTIONS IS SCORED The question: how is performance tracked? 1 No real tracking. Problems surface when somebody complains 2 Some tracking, informal, held in one person's head 3 Tracked regularly. Reviewed by managers 4 Tracked and visible to the teams doing the work 5 Tracked continually, visible to all, acted on Eighteen questions like this one. A firm's score is the average of all eighteen, so it lands between 1 and 5.
Figure 1The survey publishes the two ends of each scale, from little or no structure through to well structured. The steps between are our own plain-English reading of the progression. Most firms land between 2 and 4, which is why one point is a large move.

The size of the effect is substantial. The earlier summary of the project, published in the Journal of Economic Perspectives in 2010, reports that firms scoring one point higher on management practice have about 57 per cent higher labour productivity.

The obvious objection is that this might run the other way round: successful firms can afford to manage themselves properly. So the researchers tested it directly. They took textile manufacturers in India, gave one group high quality management consulting to help them adopt these practices, gave another group nothing but data collection visits, then measured what happened. Productivity in the treated firms rose by around 10 per cent. A follow-up years later found the practices had stuck. They had spread to other plants inside the same firms.

Which settles the direction. It is not simply that well-run firms do better. Changing the practices changed the output.

The finding a founder should sit with

The same research separates firms by who owns and runs them. Family-owned firms with a family chief executive score significantly lower on management practice in every country surveyed, without exception. Where a chief executive role passes to a family member rather than to a professional manager, management scores fall markedly.

Founder-run firms score badly too, which the authors describe as a surprising result. Their explanation is not that founders are poor managers. It is that the job changes:

the entrepreneurial skills required of a start up, like creativity and risk taking, are not the primary skills required when a firm grows large enough

They follow it with the sentence worth pinning above a desk. A mature firm, they write, needs to move beyond informal rules.

That is the whole thing in nine words. It is also the argument in scaling is the achievement, maturing is the follow-through. The informal arrangements that made a small business quick are the same arrangements that hold a larger one back. Nobody decides to keep them. They were simply never replaced, because replacing them is nobody's job and nothing visibly breaks on the day they stop being enough.

Where Australia sits

The Australian arm of the project was commissioned by the Department of Innovation, Industry, Science and Research and published in November 2009, covering 439 manufacturers of between 50 and 5,000 people.

Australian firms averaged 2.98 out of 5. That placed the country sixth of the sixteen surveyed at the time, level with France, Great Britain and Italy, behind the United States, Japan, Germany, Canada and Sweden.

The average is the least useful number of the lot. Split into the three areas the survey scores, it shows Australia strong on running things and weak on managing people.

Australian management scores by area, 2009 Australian manufacturers scored 3.22 out of 5 on operations management, 2.96 on performance management and 2.74 on people management, against an overall average of 2.98. People management was the weakest of the three areas. About one firm in ten scored below 2 on it. AUSTRALIA BY AREA, OUT OF 5 Operations 3.22 Performance 2.96 People 2.74 1 2 3 4 5 Overall average 2.98. About one firm in ten scored below 2 on people management.
Figure 2Australian manufacturers, 2009. The country ranking is the least useful figure in the report. The gap between operations at 3.22 and people at 2.74 is the part a business can act on.

People management is where the gap sat. The bottom end is where it shows. About one firm in ten scored below 2 on it. On the scale in Figure 1, below 2 means little or no structure at all: nobody formally reviewed, objectives that exist only in somebody's head, keeping a good person left to luck rather than to practice.

The ranking is beside the point. Being sixth of sixteen tells an individual business nothing. Knowing that the weakest area nationally is the one dealing with people tells you where to look first. Knowing that a tenth of firms are barely doing it at all tells you how low the bar currently is.

What this evidence does not say

The Australian fieldwork is from 2009 and has not been repeated at that scale, so read it as establishing a pattern rather than as a current score. The international project has continued since, which is why the figures above come from the 2021 review.

It is about larger organisations. The survey covers firms with at least 100 employees. The Australian study looked at medium and large manufacturers. Reading it across to a twenty-person services business is a reasonable inference rather than a finding.

Most of it is association rather than proof. The Indian experiment is the exception, one industry in one country. The honest position is that the direction is well evidenced while the exact size of the effect in your business is not.

How to test it in your own business

The question is the one the survey asks: are the practices formal enough for the size the business is now? Pull the numbers to answer it, because the same study is unkind about the alternative.

Its final question asks managers to rate their own organisation out of 10, excluding themselves, after an hour discussing the practices. Firms average 7.1 against a scale where 5 is stated to be average. That self-score is uncorrelated with the score the interviewer gives them as well as with their own productivity. Nobody holds a comparison from the inside, because a score out of 10 needs a sense of what a 4 looks like in somebody else's business.

Closing that gap is what our diagnostic is built to do. It scores six dimensions on evidence pulled from your own systems rather than on anybody's impression of them, weighting the people closest to the work above those furthest from it, then testing both against the artefacts behind them. Each dimension carries a confidence rating, so you can see how far the evidence actually supports it. Eight weeks, ending in a ranked view of where the value sits.

Where to lookWhat to pullThe calculationWhat it tells you
Accounting in Xero or MYOBGross profit against average headcount, three yearsGross profit ÷ average FTE = profit per headThe nearest proxy for what the research measures
Your documented processesRoutine tasks with a written route, against all routine tasksDocumented ÷ total routine tasks = coverageFormality is what the survey scores. Most firms overstate it
Approval history in your finance systemDecisions that still require the owner, by valueOwner-approved ÷ all approvals = concentrationWhether the business has moved beyond informal rules
Performance recordsPeople with a written objective reviewed this yearReviewed ÷ headcount = review coverageThe area Australia scored weakest

The one number to hold

  • Profit per head, across three years. Gross profit ÷ average FTE. Both figures already close every month
  • Rising means added people are adding output. Flat or falling while headcount grows is the pattern the research describes

The control worth introducing

  • A written authority limit for each manager. What they may decide, to what value, without asking. Of everything the survey scores, this is the practice that most directly replaces an informal rule
  • A standing review date for the practices themselves. Most are set once, at a size the business has long outgrown

Where your number will mislead you

  • Profit per head flatters a business that has just moved work to contractors, since the cost leaves payroll for cost of sale
  • On process coverage, count only what somebody other than the author could follow
  • A single year tells you nothing. Direction over time is the finding, so three years is the minimum honest read

Which of your practices were set when you were smaller?

That is the question the evidence supports. It is more useful than any score. Not whether your people are working hard enough. Whether the arrangements around them were built for the business you run today.

Most owners can answer it in about a minute. Most are uncomfortable with the answer, because the informal rules are usually the ones nobody has looked at since the business started.

The Free Health Check covers the six areas these practices sit in, in about three minutes. If you would rather see your own position evidenced from your own systems than inferred from a survey of somebody else's business, that is what a diagnostic produces.

Where this applies to you

The Free Health Check scores your business across six operating areas in about three minutes, or read how the independent diagnostic works.

Start the Free Health Check →