Why are business professionals more inclined to believe in something that took a lot of effort to create, or was difficult to model, over something that was quick/easy? Ideas are ideas, right?
Unfortunately, most business strategy meetings follow a predictable script.
A business team walks into a meeting room with a 127-slide deck filled with regression models, sensitivity analysis, or Monte Carlo simulations. The only goal seems to be to justify decisions that can be tested in a few weeks using real people, people who represent our consumers. Somewhere we became convinced that “difficulty” equals “strength”. We concluded that the sheer complexity of the situation makes decisions credible, and that anything achieved without apparent struggle has no real value.
Behavioral scientists call this The Effort Heuristic. If we cannot measure quality directly, we use effort as its proxy. What started as a mental shortcut has turned into an organizational doctrine. We have built all of the company’s structures to reward performance over results, systematically punishing iterative thinking that succeeds in the marketplace.
In the real world, consumers don’t care how we arrived at our pricing strategy. As far as they are concerned, a two-year regression analysis is like a quick sketch drawn on a napkin. They actually want to know if the price feels on the right. Shareholders don’t pay for efficiency, they pay for results. Yet within most businesses, we behave very differently.
Why create complexity when simplicity works
An 18-month digital transformation involving multiple vendors and endless stakeholder meetings looks safer than it is. When something goes wrong, responsibility starts spreading faster than a fart in an elevator. The seller’s problem. Even the consultant was strong. Or a cross-functional team that can’t align. At the end of the day, a person’s work is not directly affected, as we all look to blame elsewhere.
Compare that to three-week price tests conducted in the real world. If it fails, there is nowhere to hide. Failure is always there as clear as day for all to see, immediately traceable to certain people who have made certain bets.
So what do we do? We avoid doing those kinds of tests!
We avoid small, reversible decisions that teach us faster than our competitors. Instead we create detailed models and justify them through conventions. Customer observations such as “people abandon carts because they can’t find the exit button” rises to heat maps and journey funnels and engagement matrices. Instead of adding clarity, we add bloat.
What happens when an intelligent model meets reality
While their competitors poured millions into predictive modeling, Airbnb decided to buckle down when everyone else was shaking. Instead of running one big test with a gazillion variables, they ran thousands of smaller, faster tests. They look at what customers actually do instead of modeling what they might do. When testing revealed something important, they would fix it. When they don’t, they leave things alone and move on. The judging panel was incredibly low – there were no long committee meetings full of people with little else to do than model validation, or six-month forecasting cycles. What they had was real-world user learning that was aggregated quarter after quarter. The rest is history.
Starbucks didn’t measure up due to using endless customer surveys. Amazon didn’t build customer obsession with statistical models. These businesses won because they made quick, payback bets (Amazon’s famous two-door theory) and learned from actual user behavior instead of expected behavior.
The tyranny of pluralism
Too many businesses continue to treat quantitative data as evidence and qualitative data as anecdote. Interviews, observations, and direct customer feedback are all classified under “soft”, because they are difficult to measure and quantify. Finance requires five-year models before approving tests, even if those models rely on “waving a wet finger in the air” assumptions that are more tenuous than the actual efforts they were intended to test. Meanwhile, the real competitive advantages that drive markets often cannot be put into a spreadsheet.
Let’s say a marketing team discovers through three customer interviews that a certain phrase increases the likelihood of a conversion result. They bring this up at a leadership meeting and watch as everyone nods politely. Then they present a dashboard with engagement metrics from 100,000 users and watch as everyone in the room leans forward. The dataset is noisier, and the detail is weaker. But because it has lots of big numbers and is done in a pseudo-scientific way (look! it has pivot tables and everything!) it looks legitimate and is taken seriously.
What does it mean to be strong
Of course, there are many decisions that warrant complexity. I don’t want my pilot to take responsibility for doing something different this time”…just to see what happens.“Security systems, regulatory environments, or bets we can’t reverse are situations where ‘trying’ is probably not the best thing. But the thing is, most business decisions aren’t. We use nuclear-reactor-level rigor and consider complexity in banner headlines and price adjustments. We consider every choice to be irreversible if most of them are irreversible.
True robustness is not about data volume or method complexity. It’s about clear predictions and quick learning. A hypothesis like, “This banner increases signups by 15% because it removes confusion about the next step” is more robust than the 100-slide model built on stacked reasoning. The first checks in days, while the other locks in the reasoning in circles for months.
The actual cost of modeling instead of testing
Six months to model a decision not only costs budget, it costs learning cycles. A competitor who does a dozen small tests in that same window learns more than we can. Their judgment improves, as does their competitive nature. The more repetitions we can do in a predetermined time, the more we can learn. Fail fast, fail often, right?
Markets don’t reward thorough analysis or an impressive approach. Speed and timing are of the essence. Every dollar we spend justifying decisions with “complex theory” is a dollar spent preventing us from learning what works. The business that calculates things quickly, wins.
Not because they are smarter, but because they have practiced more.
