Discerning Signals in the Noise

Startup leaders must become adept at figuring out what is happening in their worlds. What are the signals, and what is the noise? This is one of the profoundly complex leadership tasks because figuring out the direction to take the organization depends on how accurately you can perceive and then translate that perception into positive forward movement.

Many years ago, when we were deep in the development of a sophisticated electronic instrument, I remember peering over my co-founder and CTO’s shoulder as he gazed intently at the screen of an oscilloscope (a type of electronic test instrument that graphically displays varying voltages of one or more signals as a function of time. Its primary purpose is capturing information on electrical signals for debugging, analysis, or characterization.) On the screen were many colored lines buzzing by in what looks like chaos. I could not discern a pattern in all the dynamically moving and overlapping waveforms. My CTO was collecting very sensitive measurements. So sensitive that we were mainly looking at electronic noise and hoping to tease out some meaningful patterns from all that noise. By applying a whole series of sophisticated signal processing techniques, our CTO was able to filter out the noise and discern the meaningful patterns that mattered in ultimately providing valid information to users. Years later, in another startup, we used sophisticated machine learning to establish a reliable algorithm that would once again tease out the meaningful features of a fast-moving waveform to provide new insights to our users.

Electronics signal processing is the domain of brilliant technologists who know how to clean, filter, and translate noisy signals into something meaningful. As an observer, I have always been profoundly impressed by the process of identifying the real signals from amidst all the noise. There is an analogous process for the business leaders of a startup who must figure out how to filter the signals from the noise in the environment surrounding the startup. Unfortunately, we do not have the advantage of fancy tools with adjustable knobs or even necessarily best practice techniques for doing so! Nonetheless, I do think it is a valuable image analogy to reflect on.

A Startup’s Environment is Inherently Noisy

One of the great challenges of building an innovative startup is that you are trying to do something new. That means that there are not many others out there who have figured out how to solve the problems of building the novel product you envision and selling it to your target customers. You cannot just copy what already works. Instead, you must gather the right information, sift out the signals from the noise, and figure out how to solve this multi-variate problem to deliver enough value to your customers to persuade them to change whatever they are currently doing and try your new solution.

Just think for a moment about all the sources of noisy confusion inherent in this scenario:

• Often, the nature of the customers’ problems you plan to solve in a new way is poorly understood. More than half the battle in creatively solving a problem is correctly understanding the problem in the first place! This means that grasping what the critical problem dimensions are requires detailed investigations and parsing out what the important factors are (signal) while recognizing what the extraneous information (noise) is.
  
  
• Designing a minimally viable product is often iterative. Figuring out an effective and efficient way to solve customers’ problems takes sifting through the potential solution paths and determining the relative prioritization of various solution features. What potential customers say is often only part of the picture because they will frequently assume that your solution will, of course, also do x, y, or z and fail to mention that during your customer discovery interviews. You may discover critical new signals and weed out less important noisy ones as you test your solution with real customers. There truly is no substitute for real customer testing with a sufficiently large sample set to distinguish signal from noise. Do not wait too long to begin running experiments with real customers so that you do not over or underbuild your product because of a lack of clear signal versus noise feedback.
  
  
• Who exactly your target customer decision-makers will be is often hard to nail down. Your customers will be those who feel enough pain from your targeted problem to be willing to spend money (a business, after all, has to sell something!) and endure the pain of change. Finding the right product/market fit is often about narrowing down precisely who experiences enough pain to do something about it. That is discerning and confirming the signal amongst the noise.
  

This list could go on and on because almost every unique element of your startup’s strategies will involve discerning the real signals from the chaotic noise. What marketing channels work, what sales messaging works, how to recruit the right team members, what service providers have the right stuff to fit at this stage, the fundraising environment trends, and so on. Generic information about startups often adds noise and rarely helps reveal the relevant signals you should base your decisions on.

Discerning the Signals in the Noise

The signal is the meaningful information that you are trying to detect. The noise is the random, unwanted variation, fluctuation, or distraction that interferes with discerning the signal. Your goal as a startup leader is to base your decisions on the genuine insights, trends, and patterns that provide an accurate understanding of the world. Finding such insights is essential but tricky, so here are a few techniques that should help you tune your perceptions and ultimately result in better decisions:

• Gather Enough Data Points to Identify Patterns Accurately. The human brain is a pattern-seeking machine, so engaging in the direct gathering of information yourself can help bring that powerful tool into play. This means you want to constantly gather, sift, and explore the world around you. Reading. Talking to people. Observing. Gathering enough data points and systematically reflecting on them is essential to enable accurate pattern identification. One way I like to do this is by personally doing customer discovery interviews where I listen carefully to open-ended questions intended to surface the perspective of my potential customers. I take notes and then type them up to allow me to synthesize what that individual was saying. Then, I repeat this with as many people as I can. One interview is not enough. Ten is not enough. As you approach multiples of tens, you will likely begin to experience repetition. That repetition indicates a signal pattern as long as you are not asking leading questions. Summarizing the results of interview notes and other sources of information forces me to seek synthesis and identify themes. In addition, I also pay attention to the counterpoints that undermine a particular pattern, so I do not assume something is always true and try to capture some of the variation into my thinking.
  
  
• Know What You Are Looking For to Match Patterns. This is the benefit and risk of experience. As an experienced entrepreneur, I know what positive and negative signals look like. Since I know what they look like, I can often more readily identify signals from within the noise because I know what I am looking for. This technique works well for those with domain knowledge, as it helps identify what is important and what is extraneous. For example, I was recently exploring getting involved with a startup that had partial overlaps with four different previous startup experiences I have had. Those past experiences helped give me perspective on this present opportunity as I sought to see how this new startup’s situation mapped or did not map to my previous experiences. It is a delicate balance of not imposing the past onto the present while still leveraging the learnings from before to accelerate understanding now. One way to accumulate such useful experience is to learn others’ stories to gather more and more data points that help identify the critical questions and what the signals might look like to make it easier to discern them from the noise.
  
  
• Run Experiments to Test Your Understanding. When you think you have teased out the crucial signals, test your understanding by running experiments to confirm that what you think is happening actually is. I call it an experiment because you want to allow for the possibility that the signal pattern you believe you have seen is possibly inaccurate. For example, I once thought I knew what messaging would work for telling our startup’s story to a particular audience, while one of my colleagues saw the world differently. We both drafted a version, and then we shared it with five target audience members for feedback without any indication of where the drafts had come from. This was a version of the classic marketing A/B testing technique. The feedback was consistent and told us which version worked and, therefore, whose perceptions of the world were most accurate. The bottom line is to look for ways to confirm your understanding by testing in the real world and being open to what the testing results tell you about your understanding of the signals.

Ultimately, an entrepreneurial team’s success will hinge on whether they can sift through all the noise around them and figure out what really matters – and how to engage that understanding in innovative ways to build a success story.