How to Build a Performance Feedback Loop in Your Business

What a business feedback loop is

A feedback loop is a cycle where what something produces gets measured, and the measurement changes what it produces next.

You already know the pattern from elsewhere. A body holds its temperature that way. An athlete's training week is built on it. Something happens, it gets measured, and the measurement decides the next adjustment.

You check the roster against last week's takings. You notice a supplier price has moved. None of that gets called a feedback loop, and all of it is one.

Where it tends to stop is at the return trip.

Why the loop breaks at the fourth stage

Because the first three feel like finishing. The report comes in, a problem gets identified, a decision gets made, and the job looks done.

It isn't done, because nothing so far has tested the decision. You have a change, and no evidence about it.

Then six months pass. The same report arrives, the numbers have moved for four reasons at once, and there is no way to separate the effect of what you did from the effect of a quieter winter, a supplier increase and a staff member leaving. The decision gets remembered as a success or a failure depending on how the year went overall, which is not the same thing as knowing.

There is also a structural reason this happens so often in Australia, and it is worth saying plainly rather than treating it as carelessness. The numbers come from the Australian Small Business and Family Enterprise Ombudsman.

Most businesses in this country have no finance team, no analyst, and nobody whose job is to go back and check. The loop does not break because owners are not paying attention. It breaks because the fourth step has no owner.

The four stages, step by step

Measure, compare, act, measure again. The order is what makes it a loop rather than a list, and the fourth stage is the one that closes it.

1. Measure what the business actually produces

   Start with what the accounting file already gives you. Gross margin, wages against revenue, stock turn, expenses against revenue, debtor days. No new system required.

   Pick three or four, and tie each one to something you are actually trying to change this year. A measure with no goal behind it becomes a number in a monthly pack that nobody reads by March.

   Three or four measures, each one tied to something you are trying to change this year.

2. Compare against something outside the business

   This is the stage that changes what a measurement means. A net profit margin of 9% is a fact about your business. A margin of 9% where comparable operators in your industry sit around 13% is a finding, and it points somewhere.

   Comparing against your own history only tells you the direction of travel. It cannot tell you whether the destination is reasonable, because your history was built on the same assumptions you are using now. Performance benchmarking sets the figure against businesses of similar size in the same industry.

   Against your own history

   What it shows

   The direction of travel. Margin up two points on last year.

   What it cannot show

   Whether two points was a reasonable amount to move.

   Where it ends

   A number that has improved, against assumptions that were never tested.

   vs

   Against the industry

   What it shows

   Where the figure sits among operators of similar size in the same industry.

   What it cannot show

   The cause. Several different situations produce the same gap.

   Where it goes

   A gap with a size, and a shortlist of places to look for the reason.

   Some of this is happening to you already. The ATO publishes small business benchmarks drawn from tax returns, currently for the 2023-24 year, and describes them as one of the tools it uses to identify businesses that may be avoiding their tax obligations. Plenty of owners meet the idea of a benchmark for the first time in that context, which is a shame, because the same comparison is far more useful pointed forwards.

   Now, a gap gives you a size and says nothing at all about a cause. Sitting below the industry middle on margin can mean a deliberate pricing position, a different cost structure, a location that costs more and sells more, a young business still investing, or an owner drawing a wage where a competitor takes profit. Any of those will produce the same number.

   The gap tells you where to look. Looking is a separate job from acting.

3. Act on the largest gap you can actually influence

   Two questions before you choose. Is the gap big enough to be worth the effort, and can this business move it?

   Rent above the industry range is real and you can do very little about it until the lease comes up. A wages line above the range might be a rostering pattern you can change this month, inside the pay and record-keeping rules the Fair Work Ombudsman sets out. Start with the second sort.

   Then write down what you expect to happen and when you will check. That prediction is the whole point of stage four, and it takes about a minute. How far to move the target in one step is its own question, and setting KPI goals covers it.

   Write the prediction and the check date down before you change anything.

4. Measure the same thing again

   Same measure, same method, on the date you set. Not a wider review, and not a different metric that happens to look kinder.

   Three things can come back, and all of them are worth having. The gap closed, so you know what worked and you can use it again. The gap held, so the cause sits somewhere the stage-two analysis did not reach. Or the gap widened, which usually means something moved against you that nobody had spotted.

   Whichever it is, you now start the next loop with a better question than you had at the beginning.

   Same measure, same method, same date you wrote down. A different metric is a different question.

What makes a measurement worth putting in the loop

Three conditions, and a measure that fails any one of them will generate reporting instead of improvement.

That last condition quietly rules out a lot of what ends up in management reports.

How often should you review performance?

Match the interval to how fast the measure moves and how long your change takes to show up.

What matters more than the interval is setting it before you start. The ABS gets a usable series out of one survey a month because it always runs. A review that only happens when somebody remembers gives you scattered observations instead of a series you can read.

A worked example

Illustrative scenario. Not a real business, and the movements described are examples rather than benchmark data.

At the quarter, wages against revenue has moved most of the way toward the industry range. Gross margin has not moved, which tells her the saving did not come out of service levels. That is a closed loop, and it is also the moment the next largest gap becomes worth opening.

Where feedback loops mislead

Naming these is what stops a loop being sold as a guarantee.

What to do this month

None of this needs a new system, and all of it fits inside a month.

The measurement is the easy part. The return trip is what makes it a loop.

Frequently asked questions