Horizon Scanning Examples: The Signal Was Never Missing

Everyone in merchandising wanted Asia.

This was 2009. I had just come back to run the whole factory network. Asia was the lowest-cost producer and nobody was wrong about that. Our plants sat in Latin America, inside NAFTA and CAFTA, and we were never going to win a price comparison. I knew it. They knew it.

The argument I was losing was not about cost. It was about which cost.

Quality was not the differentiator either, and not for the reason people assume. You cannot inspect quality into a garment. You build it in or you do not have it, and any competent factory anywhere can be set up to build it in. Controllable does not mean differentiating.

What we had was time. We cut the cycle from order to delivery under fifteen days. Ocean freight does not care how good your factory is.

Under fifteen days did something that never appeared in the comparison anyone was running. In aggregate we ran low to mid nineties on forecast accuracy. Not because we forecasted well. Because we were only ever guessing two weeks out. Push that same team to a ninety to a hundred twenty day commitment and the number falls apart, and every point it falls costs you inventory you did not need or a sale you could not make.

Inventory turn. Inventory investment. Service level. Obsolescence. Every one of those was better on our side. Add them up and total cost of ownership was lower in the Western Hemisphere.

FOB was lower in Asia.

We won that argument, eventually. What I did not understand at the time was why it took as long as it did, or why I would have it again seven years later, in 2016, at a medical products distributor, with different people who were equally competent.

Last week I said the third instrument answers the only question the first two cannot. What is coming that has not arrived yet.

This is that issue. Foresight Compass. The Strategic Leadership Constellation and Horizon Scanning, Frameworks 15 and 16 in The Resilience Engine, chapter five.

And the thing I had wrong about foresight for most of my career is that I thought it was about looking further out.

It is not. It is about what happens to a signal after it arrives.

Inside the instrument

The first part is how you divide your attention. The Strategic Leadership Constellation splits it three ways. Down into the work, Out at the world, In at yourself. The framework is about proportion, not presence. Most operations run heavy Down, thin Out, and empty In. That is what happens when only one of the three has a meeting attached to it.

The second is how far the Out view reaches, and how wide.

Horizon Scanning takes the Out view and pulls it forward. It also widens it, and the widening is what most people miss. Out is not your market. Your market is the part of Out you already watch.

The book asks you to define three to five domains of strategic uncertainty and gives its own examples. Customer preferences shifting. Regulatory change. Platform companies entering your space. Supply chain and nearshoring. Talent availability.

The fourth one on that list is the entire story I just told you, and I did not recognize it as a scanning domain at the time. I thought it was a sourcing decision. Two of them, seven years apart.

What I would add to the book's list now, from my own operations, is macroeconomics and geopolitics. All of them reach your P&L eventually. None of them is on your competitive set.

It is also a practice, not a project. Defined sources, a standing block on the calendar, written records. That is the whole operational footprint and it is smaller than most people expect, which is exactly why the absence of it is a fixable gap rather than a resourcing problem.

The third part is what happens to a signal once you have one, and the book is explicit. The last stage of the scanning cycle is integration into strategic decision-making.

That is the connector, and it is where the strategic work has to marry the tactical. A signal that does not change what you watch Down, or how much of your week goes Out, or what you are checking In, has not been integrated. It has been noticed.

The In view does one specific job inside that, and here I have to correct my own book.

Chapter five writes the In view as a check on cognitive bias. Your assumptions, your history, the blind spots that come from how you are built. That is what it says and I would write it differently now. Not because bias is not real. Because both times I lost this argument, the reporting line was what dropped the signal. I will get to that.

Assessing a signal

Assessment is what stands between noticing something and doing anything about it, and it is where most scanning stops.

I use six checks. I called them SPICED in the book and never explained them there, so here they are.

Scale. How big is this if it lands.

Probability. How likely is it to land at all.

Impact. What it hits in our operation specifically, not in the industry generally.

Corroboration. Is anyone outside our walls seeing the same thing, or is this one source and one reading.

Evidence. Is there data underneath this, or is it commentary on top of data somebody else already interpreted.

Direction. Which way is it moving.

The last two carry most of the weight and they are the ones that get skipped.

Evidence separates a signal from a story about a signal. Most of what reaches an executive is the second one, already shaped by somebody with a stake in how you read it.

Direction is the difference between a snapshot and a trend. A signal that is decaying and a signal that is accelerating look identical the day you first notice them. Exactly one of them should change what you watch, and telling them apart requires looking twice. That is what makes this a practice.

Each one alone

The Constellation without Horizon Scanning gives you an Out view that reports the present. You look at your market on a schedule and you see current customers, current competitors, current conditions. The practice is real and the finding is always slightly historical. You arrive on time at a market that already moved.

Horizon Scanning on its own produces a collection. Signals get gathered, filed, circulated, and nobody's job description contains the word. There is no shortage of interesting weak signals in most companies. There is a shortage of anyone whose work changes because of one.

Put both in place with no connector between them and you have the failure that cost me two years. Merchandising was doing the job merchandising exists to do. The FOB number was accurate and anyone who quoted it could show their work.

For years I told this story in the direction that flattered me. The true version is duller and more useful.

Every cost of going long was already in our financials. Markdowns on the styles that did not sell. Air freight on the ones that did, when we tried to chase a season we could not chase. Inventory sitting in a size and a color nobody wanted. Reorders that would have landed after the season closed, so we never placed them.

All of it was reported, on schedule, accurately. Merchandising owned markdowns. Logistics owned freight. Planning owned inventory. Sales owned the misses.

What nobody owned was the addition.

That is what a filter looks like inside a real company. Not bias in anyone's head. Structure. The number that gets acted on is the number somebody owns, and the signal that gets discarded is the one that belongs to everybody.

The evidence

McKinsey published a piece in February on how competitive edge wears down. The finding that stopped me was not about strategy.

Most companies are not monitoring how their industry standing and their edges might be shifting underneath them. Not doing it badly. Not doing it.

The same work found that the shuffle rate, meaning how quickly leaders and laggards trade places, has sped up in more than sixty percent of industries over the past decade, with the median up eleven percent.

Hold both correctly. The shuffle figure is an observed industry measure and it is solid. The monitoring finding is what companies report about themselves, so the true number could be better or worse than stated, and it certainly includes people who believe they are watching because someone forwards them a newsletter.

That finding does not prove scanning causes durability. It says the ground is moving faster and fewer people have anybody assigned to watch it move.

Which raises a question I would rather you sat with than agreed with. Do you have a practice that would have surfaced any of this before somebody else told you about it.

One move this week

Not a scanning practice. You will not stand one up this week and I will not pretend otherwise.

Do this instead.

Pick one decision your organization keeps making the same way. A sourcing call, a vendor renewal, a staffing model, whatever yours is.

Write down the number that decision gets judged on. Then write down the name of the person who owns that number. That part takes ten seconds, which is the point.

Now write down everything else that decision costs. Not estimates. The categories. Then try to put a name next to each one, and then try to put a single name next to the total.

If the total has no name, you have found your filter. It is in the reporting line. And you found it in an afternoon, without building anything.

One instrument out of three

Act III has been building one argument. Where you stand, where you are looking, and how far ahead you can see are three different instruments, and most operations own one.

Two weeks ago we named the thing worth defending. Last week we looked at where a leader's attention actually goes. This week it extends forward, and the surprise is that extending it is the easy half. The hard half is whether anything you find has somewhere to land.

Merchandising was looking hard, at the only number that existed, and they were right about it.

I was not arguing for a better number. I was arguing that the one that mattered had to be assembled, and assembling it was on nobody's calendar.

Next issue closes the trio and asks the question underneath all three. What you have that is worth defending, and whether it still survives the terrain in front of you.

Oscar

Want to see whether the signals reaching your operation have somewhere to land? The Resilience Engine Maturity Assessment maps it: myresilienceengine.com/maturity-assessment