Before You Decide

Improving Decisions Is Easier Than You Think

In February 2018, Brian Niccol was named chief executive of Chipotle. The company was still working its way back from a series of food safety incidents that had begun in late 2015. The stock was well off its highs. The conventional diagnosis of what needed to change was entirely internal: fix the supply chain, tighten operations, rebuild trust with a customer that had been burned.

However, some very important early decisions Niccol made resulted from looking outside the company and proved important to the future of Chipotle.

Mobile ordering at coffee chains had moved from novelty to habit. Third-party delivery had become a national infrastructure. The demographic Chipotle depended on most, the under 35, was now spending more time choosing meals on a phone than choosing between restaurants on a sidewalk. These were observable to anyone paying attention. Niccol’s team chose to act on them.

Within the first year, Chipotle accelerated a digital ordering system the organization had previously treated with ambivalence and committed meaningful capital to a new category of physical asset, namely, pickup-only drive-thru lanes branded as Chipotlanes. Neither investment was a response to an internal problem. Both were a result of adapting consumer trends to their business.

Two years later, the pandemic forced restaurants to close their dining rooms almost overnight. Chipotle, nearly alone among its direct peers, had the digital and pickup infrastructure waiting. Digital sales, roughly a tenth of revenue the year Niccol arrived, rose to approximately half of revenue at the height of the disruption. The company’s market capitalization, about eight billion dollars when he took over, would multiply several times in the years that followed (Chipotle Mexican Grill, Form 10-K, 2020 and 2021).

Niccol did not predict the pandemic. No one did. What he did was focus on what was coming from outside his company accurately enough to recognize that the odds of a particular kind of future had shifted, and he built the business to perform well across a wider range of those futures.

There are two things that stand out about Niccol's choice. First, he was not guessing. He had run the same digital playbook at Taco Bell that included mobile ordering, kiosks, and delivery. He recognized the same thing was coming to Chipotle. He understood the importance of fixing the operational shortcomings which he did. However, what was game changing was the commitment to initiatives that would drive transaction volume, recognizing the need for digital platforms, menu innovation, and brand to solve what he recognized as the key need, transaction volume. His second decision was that he tested before he committed. The first Chipotlanes and digital pickup shelves were piloted in single stores before any meaningful capital followed.

The discipline that made the difference was not bold prediction. It was reading where the outside world was already moving, allocating attention and capital toward those signals, and testing the bets at small scale before scaling them.

The point is not Chipotle. The point is the discipline to consider external factors in decision making.

What Big Decisions Actually Depend On

Every company eventually reaches a decision that doesn't have a clear internal precedent and that can meaningfully change the trajectory of the business. Pricing, hiring, and debt come up every year and that's not the kind of decision I'm talking about. I'm talking about the moments when last year's answer is no longer enough. For example, what if the annual price review turns into a question about whether to reposition pricing altogether, or what if the hiring plan turns into a question about whether to hire ahead of demand that hasn’t happened. A leader makes the routine versions of these decisions many times. The strategic version of these decisions, occur less often. And the effect of these decisions may impact the company for years.

These decisions share a quality: the information that should most influence them tends to sit outside the business. If we use pricing as an example, this is not exclusively about what you charge, rather, it is about what your customer is willing to accept given what is happening to their own costs, their own consumer confidence, and their own alternatives and how a product or service makes them feel.

To make a decision, leaders not only need to understand their own internal capabilities to execute but also the external signals that can help determine if this is a good time or the company is likely to face significant headwinds. And often leaders at smaller companies do not have the time or the infrastructure to compile the information required to produce those signals.

Why Outside Data Can Present Challenges

There are reasons external signals are hard to absorb, such as:

•    The signal-to-noise ratio is punishing. News, forecasts, and commentary all crowd into the thin margin of time a busy business executive has left over for the world beyond their own business.

•    Forecasts are often wrong. And leaders know it. Years of miscalled recessions, inflation surprises, and expert misreads have justifiably made executives cautious of any single view or time.

•    Traditional risk frameworks don’t help much. They catalog what has already happened. They do not tell a leader whether something new needs to be incorporated.

•    Precision is mistaken for reliability. Decision cultures reward forecasts that sound exact and dismiss probability-weighted forecasts; However, to better understand a course of action, a leader must be able to make decisions that factor in potential outcomes likely enough to plan around and large enough to matter.

None of these are reasons to conclude that outside information is unusable, rather they are reasons leaders develop a defensible habit to wait until the external environment is obvious before acting on it. The trouble is that by the time it's obvious, the leader's options and timelines have already narrowed. Competitors have taken action. Customer expectations have reset. The question stops being "what should we do?" and becomes "what can we still do?" The information was available earlier. The discipline to use it wasn't.

What Has Actually Changed

The case for paying attention to external signals is not new. What is new is how much more practical and reliable it has become.

The ability to understand the collective thinking used to be impractical at best. Recently, the practice of aggregating independent forecasts has matured into something leaders can actually use. When enough people make independent judgments on the same question and those judgments are aggregated well, the group tends to outperform individual experts. Two decades of economic and behavioral research (Snowberg, Wolfers, & Zitzewitz, 2013; Tetlock & Gardner, 2015; Atanasov et al., 2017; Augenblick & Rabin, 2021) have formalized when and why. What has emerged is a new venue. Regulated prediction markets including CFTC-supervised exchanges, where participants put real money on questions like Federal Reserve decisions, GDP releases, election outcomes, and a growing list of business-relevant events, now operate in the United States. With growing use, the markets have enough liquidity that the prices function as usable probability estimates. That is what turns a research finding into a tool a leader can pull from when they need to make a decision.

Add to that the fact that artificial intelligence (AI) has collapsed the time and cost of synthesizing any of this information. What previously required a research team can now be done by a disciplined leader with the right framework and an afternoon.

The combination matters. Market-implied probabilities give a raw read on where the odds are moving. Structured judgment aggregation sharpens the reading on specific nuances of the questions. AI collapses the time required to turn those inputs into something usable. No single piece of this is sufficient. Together, they give leaders better tools for reading the environment than they have ever had.

Context, Not Prediction

The instinct is to treat anything probabilistic as a forecast to be judged against the outcome. That is the wrong test.

The value of external signals is not telling a leader what will happen. It is telling them when the distribution of possible outcomes has shifted and when the odds of a scenario happening have moved enough to credibly inform their most critical decisions.

That is the difference between waiting for proof and choosing while you still have choices.

The Series, and the Dashboard Behind It

Before You Decide takes one concrete decision at a time such as pricing, segment entry, debt, hiring ahead of demand and walks through the external signals that should inform it before it's made. It is written for leaders whose next big decision should be more on what is happening outside the business than on what is already understood inside it. There is no fixed formula. The decisions leaders face are too varied, and the inputs that matter get pulled together differently for each question, industry, and leader. None of this is meant as advice. The series exists to model the discipline: what to look for, where to find it, and how to weigh it before a decision gets made.

The point is simple — when a decision comes up, the outside picture should be a minute away, not a project away.

Brian Niccol had the time, the expertise, and a team to build his own read on the outside world. Most leaders don’t. The series is built to model the discipline so reading the outside world stops being something you hope to do when time allows and becomes something you actually do, before the decision is made.

The information exists. The question is whether you are using it before you decide.

‍Sources

• Atanasov, P., Rescober, P., Stone, E., Swift, S. A., Servan-Schreiber, E., Tetlock, P., Ungar, L., & Mellers, B. (2017). Distilling the wisdom of crowds: Prediction markets vs. prediction polls. Management Science, 63(3), 691–706.
• Augenblick, N., & Rabin, M. (2021). Belief movement, uncertainty reduction, and rational updating. Quarterly Journal of Economics, 136(2), 933–985.
• Chipotle Mexican Grill, Inc. (2020). Annual Report on Form 10-K for Fiscal Year Ended December 31, 2020. U.S. Securities and Exchange Commission.
• Chipotle Mexican Grill, Inc. (2021). Annual Report on Form 10-K for Fiscal Year Ended December 31, 2021. U.S. Securities and Exchange Commission.
• Hayek, F. A. (1945). The use of knowledge in society. American Economic Review, 35(4), 519–530.
• National Federation of Independent Business. Small Business Economic Trends (Optimism Index). https://www.nfib.com/surveys/small-business-economic-trends/
• Snowberg, E., Wolfers, J., & Zitzewitz, E. (2013). Prediction markets for economic forecasting. In Handbook of Economic Forecasting(Vol. 2, pp. 657–687). Elsevier.
• Tetlock, P. E., & Gardner, D. (2015). Superforecasting: The Art and Science of Prediction. Crown.

‍