So far, one of the most important concepts we have emphasized throughout the development and setup stages has been data-driven decision-making and measurement. Even when we act based on reactions, we move forward by relying on measurement and interpreting public response. But can success be measured?
The answer has two sides. Success, in its broadest sense, can never truly be measured. We can measure whether something qualifies as successful, but the degree of success itself is not something that can be quantified. That is because every measurement will be subjective. We may be successful by the standards of our industry, yet unsuccessful compared to players in the same industry operating in a different market. Likewise, we may be performing terribly compared to our own past while still being successful relative to competitors in a market experiencing a crisis. Success can even vary considerably from person to person. For example, in our view, someone who starts a business without any capital and achieves 50% growth is more successful than someone who achieves 1,000% growth with substantial capital or investment.
But all of these judgments, as we have just established, are subjective. Success cannot be measured as a universal ratio. As we discussed in the chapter on goals, purpose, and dreams, we can recognize that we are successful when we achieve our goals. But we cannot quantify success itself.
Remember: comparison is the thief of peace. That is why, whether in business, personal ventures, or your own individual life, you should never compare yourself to others for the sake of competition. Competition is justified when your path requires it; at all other times, it is poison.
Nevertheless, we can measure our own success proportionally and mathematically. In fact, this is as much about measuring growth as it is about measuring success itself. If we have correctly defined our goals and purpose, and completed our setup in a way that allows us to see the path ahead, only one thing remains for measurement: performance metrics.
Key Performance Indicators (KPIs) are the indicators we establish to determine whether we are achieving our goals or progressing toward success at the pace we intended. They may have underlying variables. But if we genuinely want to measure our success and growth, defining our KPIs is not a choice. It is a necessity.
To illustrate this properly, let us consider two examples. Suppose we establish a B2B venture. We set a first-year revenue target of $1 million, along with two additional primary objectives: acquiring at least 20 accounts and achieving a 5% proposal conversion rate. In this case, my key performance indicators should be conversion rate, monthly and annual revenue changes, and positive or negative evaluation feedback. Supporting metrics connected to these primary indicators could include successful conversions per salesperson, A/B testing of proposal structures, and similar data. Measuring these at regular intervals will immediately reveal both our performance and efficiency on the path to success, as well as the interventions we need to make.
Or suppose I am a painter. My primary metrics would be reach and demand. How many people were exposed to my paintings, and how many wanted to purchase or exhibit them? Once again, I would establish a goal and determine the intervals at which I would measure it.
What matters here is that you define your key performance indicators not according to your preferences or subjective opinions, but according to what the business genuinely requires. An e-commerce company cannot use social media likes as its primary performance metric. A painter cannot define revenue as their primary metric, just as a corporate construction company cannot measure its primary performance through reach.
Your key metrics are your speedometer and odometer. Every venture must establish these during the maturation stage, measure them at regular intervals, and continuously monitor them. Living by data, when combined with goals, dreams and purpose, efficiency, and structural integrity, is undoubtedly one of the fundamental elements that lead us to success.
So, as always, let us look at real-world examples of this principle.
Facebook's growth team asked a question instead of simply looking at registration numbers: What did the people who stayed on the platform do during their first few days? The answer was surprisingly simple. Users who made seven friends within their first ten days tended to stay, while those who did not usually left. This observation became the team's shared benchmark. Every new feature and every experiment was tested against the same question: Does this help more people reach seven friends faster? As a result, discussions stopped being battles between competing ideas and became measurable experiments. Seven was not a universal number. It was the compass for that particular product. Their real achievement was finding that compass.
In 2002, the Oakland Athletics were operating with the third-lowest budget in Major League Baseball: approximately $40 million. The New York Yankees had a budget more than three times as large. General manager Billy Beane and his team selected players not for their spectacular hitting, but for their on-base percentages, because the market was undervaluing that particular contribution. The result was 103 wins, a 20-game winning streak, and the same win total as the team with the highest payroll. Their cost per win was approximately one-third of the Yankees'. It is important to remember that the roster also included three excellent pitchers that year; the story cannot be explained by a single metric alone. But the reason a small-budget team performed so far above average was not that it spent more money. It was that it measured the right thing and allocated resources accordingly.
When Alan Mulally joined Ford from Boeing in 2006, the company was on course to finish the year with a loss of approximately $12.7 billion, and its market share had been declining for a decade. Mulally began bringing senior executives together every Thursday morning for a weekly business plan review. Every item would be marked green, yellow, or red. During the first few weeks, every item was green, even as the company was losing billions. Eventually, Mark Fields, who was responsible for North America, presented a technical problem that was preventing the launch of the new Edge model and marked it red. Tension filled the room. Mulally, however, applauded and asked how he could help. Seeing red was the first condition for solving the problem. By the beginning of 2011, Ford had become the world's most profitable automobile manufacturer without going bankrupt or accepting a government bailout.
In 2009, Airbnb's weekly revenue was approximately $200, and growth had stalled. The numbers revealed that growth had stopped, but they did not explain why. When Joe Gebbia explored the product as though he were a user, he saw the answer: the photographs in New York listings were poor. The founders chose a simple solution that could not be scaled. They rented a $5,000 camera and went from home to home, taking the photographs themselves. Revenue in the city doubled within a month. The starting point was small, but the lesson was significant: a number revealed the problem, another number validated the solution, and the same improvement worked in Paris, London, Vancouver, and Miami.
Now let us examine the opposite: those who measured the wrong things, manipulated their measurements, or failed to measure at all.
For many years, Wells Fargo publicly reported a figure that its competitors envied: the number of accounts and products per customer. When that figure became a target for branches, it created pressure to sell volume regardless of what customers actually needed. Employees opened millions of accounts without customers' knowledge. According to government records, senior executives at the bank had known about the pressure created by these targets as early as 2002. The scandal erupted in 2016: a $185 million penalty and the dismissal of approximately 5,300 employees. The matter was not settled until 2020, with a $3 billion resolution. The green indicators on the dashboard did not show that value was being created for customers. They showed only that accounts were being opened.
When WeWork disclosed its initial public offering documents in 2019, it highlighted a profitability measure of its own invention: "community-adjusted EBITDA." This measure excluded major expense categories such as marketing and general administration. When investors examined the real numbers behind the metric, they discovered that the company was recording almost a dollar in losses for every dollar it earned. Its $47 billion valuation collapsed within weeks, the IPO was withdrawn, the founding CEO stepped down, and the company filed for bankruptcy protection in 2023. When you manipulate a metric in your own favor, you turn the speedometer that should guide your journey into a comforting decoration.
On the morning of August 1, 2012, during the first 45 minutes after the market opened, Knight Capital's system sent more than four million orders in response to just 212 customer orders. It was a faulty software deployment, but what brought the company to the brink was not the error itself. It was the absence of a control capable of stopping it. The SEC later found that the company had not connected its overall risk exposure to automated controls and lacked adequate procedures for code deployment and testing. Approximately $440 million disappeared in less than an hour. That amount exceeded the company's market value at the time, bringing Knight to the edge of bankruptcy. An indicator capable of detecting the deviation, combined with a limit capable of stopping it, could have contained the disaster within minutes. Neither existed.
Target entered Canada in 2013 and closed all 133 of its stores less than two years later. Some 17,600 people lost their jobs, and pre-tax losses reached $5.4 billion. At the heart of the problem was product data within the system managing the supply chain. According to reported accounts, only 30% of the data was accurate, and the system lacked a mechanism to detect input errors and issue warnings. Product dimensions had been entered using incorrect units, shelves were empty, and warehouses were overflowing. The data was wrong, and because no one measured the extent of those inaccuracies, the problem multiplied with every new store opening.
What these eight stories have in common is this: the winners chose indicators that reflected the real value of their businesses, monitored them regularly, and intervened when they saw red. Among the losers, the indicator was either wrong, manipulated, or entirely absent. Luck is part of every success, but measurement allows us to see the road ahead instead of waiting for luck to return to the average.
