10 Data Analysis Mistakes to Avoid (#7 is most common for beginners)

10 Data Analysis Mistakes to Avoid
10 Data Analysis Mistakes to Avoid

Data about your business can be extremely powerful, when analyzed correctly. Once you get familiar with the tools, the power and understanding can become addictive. But most business decisions are based on past experiences and “gut feel”. That definitely makes sense some of the times and in some situations. Applying your wisdom and experience, rather than massive spreadsheets and dashboard certainly helps make quick decisions throughout the day. Running A/B tests to determine the perfect shade of carpet for your office is a sure fire way to make sure it never gets replaced.

10 data analysis mistakes to avoid

  1. Not cleaning up messy data. Always assume the data you are working with is inaccurate at first. Once you get familiar with it, you will start to “feel” when something is not quite right. Take a first glance using pivot tables or quick analytical tools to look for duplicate records or inconsistent spelling to clean up your data first.
  2. Not normalizing the data. Say you’re trying to build a dashboard or report to determine you best employees spread out through out the country. Well, you might think that doing some simple counts on client contacts or phone calls would do. Well, what if you have a sales rep that has been with the company for years and has exhausted all of their opportunities with their current client list. Be sure you have a level playing field (as much as possible, anyways…) or you may end up penalizing or crediting some people incorrectly.
  3. Ignoring outliers. Often times a quick glance at some trended data, like number of visits to your website each day or leads coming through various campaign channels. It’s common for most folks to ignore the extremes (either a huge spike or drop in website visitors, or lead volume) because they are most likely just flukes. But outliers in data can also indicate that something is wrong. Either a process is broken or some web form isn’t working properly. Investigate these outliers in the data to make sure nothing is seriously wrong.
  4. Fixating on outliers. Okay, so you shouldn’t ignore the outliers as I mentioned in the last point. But you can’t focus on those and ignore everything else either. For example, you probably wouldn’t want to build an entire product line around a 1 day spike in traffic from a popular guest post (unless you have other data points that confirm what your audience wants, of course!)
  5. Not adjusting for seasonality. Holidays, summer months, and other times of the year can mess up your data. Even a 3 month trend can be explained away because of busy tax season or back-to-school time. Make sure you are considering any seasonality in your data…even days of the week or times of the day!
  6. Not watching metrics in context. When you’re just getting started, it can be tempting to get focus on small wins. While it’s definitely important and a great morale booster, make sure it’s not distracting from other metrics you should be more focused on (like sales, customer satisfaction, etc.)
  7. Data overload and chart junk. Too. Many. Metrics. Sometimes less is more. Especially when building a dashboard or even just a simple analysis. Make sure that everything on the screen or the page has a clear purpose and there’s no extra stuff to distract.
  8. Using meaningless metrics. Be sure that you are measuring and responding to stuff that actually matters for your business. If you pick things that don’t matter early on (like web page response time) then you will probably end up ignoring the metric anyways along with a few others that actually do matter to the health of your business.
  9. “Where did that come from?” Most of your data will come through a collection of systems in your business. For example, your email list is managed by one provider, website visits are in your web analytics tool, and your customer list is spread across 3 other systems. When you collect and combine the data, you need to make sure you haven’t made any mistakes along the way. You may end up drawing the wrong conclusions from the end result.
  10. Missing the signals through the noise. Spotting trends is hardwired into our brains. Sometimes we see patterns emerge where this is none. Be sure you don’t have an analysis that is feeding you information based on your “need” to see signals where there’s actually just noise.

Conclusion

Data analysis doesn’t have to be a huge chore that must be performed before getting anything done. Rather, data analysis can be used to identify trends you would have otherwise never seen. It’s important to stay informed about the health of your business and tracking key metrics is a great way to do that. It’s important to keep the big picture in mind when drilling into your data but ignoring all together is not the answer either. Decisions should be be informed by your business data, not driven by it.