Tuesday 7 January 2014

Digging through the Data: How to Separate Quality from Quantity


With current technology, we could amass thousands of data points on millions of people almost instantly. The Internet allows us access to countless surveys and metrics to measure nearly anything imaginable. With the ability to reach almost any type of audience and collect nearly limitless data, what’s the end result?
Usually about a billion bits of information that gather dust while the next pile of data is prepared.
Data is only as valuable as what you do with it. The volume of data we collect means nothing if it doesn’t help where it hurts. Where to begin?
1. Know Your Destination before Reading the Map
Planning for a trip is always easier once you know where you’re going. Data-driven analysis always comes back to the scientific method. The first step of the method: Form a hypothesis. Too many experts overlook this basic premise and jump headlong into the metrics. But if you have no idea what your prediction or variable is, then how can you measure its success?
That first step might be the toughest, so take a breath. Your belief in science now points you to your destination. You have an end state in mind. On this journey, consider data to be your fuel and the reporting engine your vehicle. Analyze the data to see if your hypothesis turns out to be valid. Notice I didn’t recommend discovering as many shortcuts as possible. One tactic isn’t going to totally revolutionize your industry. Instead, trust the science and your own ideas to change it.
2. Let Your Changes Drive Your Research
The analysis and future use of your data is based upon your expectations. These expectations determine the time, energy, and money you’re willing to spend on relevant research. If you’re contemplating changing your website background color to a seasonal print, you have no need for expensive consultants. However, if you’re developing a total redesign of your online home, you’ll want to invest in more than just a Magic 8 Ball to guide the way.
Imagine a small business that immediately opts for a total website overhaul after a short period of profit decline. Did that business measure what was working or not working with the site before? Did it assess its marketing and customer service strategies? Is the outlook “very doubtful”? Instead, that business hopes a stab in the dark will guarantee success. Most likely, it was snap decisions based on flimsy data that led to declining profits in the first place. Target your variables, and base your changes on valid data.
3. Measure around Risk
Now that you have a destination and recognize the resources needed to reach it, you can focus on your variables. Assess your risks, and choose which element you think needs to change to better your business. Is it a higher conversion rate, more downloads, or lower monthly costs? Specifically, how is your “perfect world” version of the business different from its current state?
You can collect reliable data once you’ve determined what you’re measuring. Try to isolate your variables as much as you can. This can be difficult in a large company, but it’s necessary to interpret your data accurately.
4. Use or Lose Your Data
Data is not wine — it doesn’t get better with age. Methods of collecting, storing, and reporting data constantly change. If you’ve spent hours gathering information, put it to work. Get as much value out of it now as you can. Don’t collect data just to store it. Unfortunately, many businesses do use this approach. No clear hypothesis or purpose for the data was established from the beginning, and it winds up collecting digital dust on a virtual warehouse shelf.
Quick and convenient data-gathering devices and metrics allow for obscene amounts of information to be gathered. However, data for data’s sake doesn’t help move you or your business forward. By clearly defining your goals, risks, and methods at the beginning of your collection, you can forge a practical path to positive change. Data can be an incredible tool. Use it wisely.

Follow Me On Twitter: https://twitter.com/arjunsantosh

No comments:

Post a Comment