Myth Of Data

data analysis

Last weekend, I needed to figure out what to eat for dinner and did not know what I wanted to eat. I did what many other people do, I started scrolling through the food apps. 

I was searching for inspiration and hopefully a (delivered) solution to my hunger. 

An hour later, my discovery process had me going through more than 7 food apps that ranged from the food delivery apps, restaurant reservation apps (going out was an option), and even grocery apps (why not, cook?). 

As I went through this labyrinth, I only got more hungry and still did not have a solution. Why? Because I started the process, I discovered constraints that forced new considerations. 

  • Cuisine type
  • Time: Delivery time, cooking time
  • Opportunity cost: Time to order pickup, save on the fees, and save time waiting
  • Fees (all those fees!) 
  • Promotions: limited time no delivery fee, new order discount, spend $X, get Y% off
  • Health considerations: is fried anything a good option for the 10th time this week or should I just get a salad?
  • Socializing: Eat alone? Eat with a friend?
  • New discoveries: New restaurants or cuisines I haven’t tried but looks appetizing
  • Review: Need to know what others thought about these “platform recommended” places
  • … 

You get the picture. I had plenty of data, but further from a solution.

In the end I had to settle for second or third best choices. I ended up making something at home – ramen. Sad. 

The data rabbit hole

While the spirit of wanting more information is right, how a startup captures that benefit has a lot to be desired. 

There is a belief that having more data will provide the answers, and it potentially can but you need the right framework to get the insights you need to make actionable decisions.  

Let’s try to illustrate this with a simple example. 

Imagine someone who needs to get from point A to point B. They believe a car is the way they’ll get from A to B. Then each day, someone delivers a piece of the car. A wheel on one day. Nuts and bolts on another. Steering wheel on a future day. You get the picture. After a while you have a lot of parts. You potentially all the right parts to build the car to get you from A to B.

Except you don’t know. You were busy accumulating parts for a car or at least the parts you are aware of or can acquire. You did not figure out if you have the right parts for the right car that will get take you from A to B. 

And that’s where the promise of data leads some founders down the wrong time-consuming rabbit hole. 

Ask better questions

When a startup says, “we need more data”. I pause. 

This is a common statement I hear from founders when they need to make a decision on something and are uncertain how to proceed. 

Seeking more information is great. Seeking quality information is even better. 

Unfortunately for most startups, what this means is unclear, just short of the answer falling on their lap. Anyone who’s ever worked with data knows this just does not happen. 

Instead startups should start by asking better questions. Better questions help startups a few ways: 

  • Clarity in what the startup actually needs to know.
  • Determine what the startup doesn’t know.
  • Where urgent gaps exist. 
  • Avoid opportunity cost of waiting (worst) or collecting data that does not get you insights and solutions (still pretty bad). 
  • More importantly, consider ways to acquire better data that lead to insights necessary to build steps you can execute on. 

Use this approach to start building your strategy 

One simple way to assess how well your startup has this process down is to see what data and insight your business and pricing strategy is built on. 

Take a single piece of paper and write down in one or two lines what your business and pricing strategy is. Then underneath each strategy list the data, analysis and insight used to conclude this was the right strategy for your startup. 

Now some founders will say: “we’re building something new” or “it’s never been done before”. 

Strategy is forward-looking and it needs reason-to-believe to pursue. This is why so many startups who believe they have a strategy, don’t have one.  

Like Airbnb who discovered people were willing to sleep on air mattresses in a strangers apartment during the Democratic convention. They collected data based on specific questions. 

Like Slack who found that teams disliked email and wanted more efficient ways to communicate and collaborate. 

Even beyond product-market-fit questions and data collection, decisions by Uber to launch a  VIP program in 2016 and later a loyalty and reward program in 2018. Strategy that required hard questions, and data collection on “something new”. 

Sometimes data isn’t sitting nicely packaged for you on Google. When that happens, conducting primary research is one effective way to create your data. 

Startups who are running lean – focused on thoughtful experiments and testing to test hypotheses and iterate – will find this second nature. For others, now is the time to start. Discovery should not be confused with serendipity. 

Great decisions are waiting, otherwise you might be left with ramen that’s been in the cupboard for years.

Did you know? 

We shared this article with our email newsletter community first. If you want to get access to our articles and insights before anyone else, you can sign up here (plus, it’s free!)


Found this article helpful?

Sharing is caring. ❤️

Share this on social – super easy 1-click share buttons below 👇 – or send this article to a colleague or friend who can learn something new to empower their company or hustle.