Get Even More Visitors To Your Blog, Upgrade To A Business Listing >>

The Importance of Data Quality (Garbage In, Garbage Out)

Collecting data for analysis is not a "set it and forget it task." In order to ensure we are getting data that is useful for our business and marketing strategies, it's essential that we take time to get it right. In the age of Big Data, poor data quality is a common issue that many organizations are unaware of. In this video, I'll share some simple steps to help you collect data that matters so you can make data-driven decisions.

IMG_0430

Video Transcript:

Hey what's up, everybody? Welcome to 'Hack my Growth.' In today's episode, we're going to be talking about the importance of data quality.

Hey, thanks again for watching. If this is your first time or you've been watching a while but haven't yet subscribed, please do so right now. We would love to have you join part of our community. All you gotta do is hit the button below the video.

All right, so today we're talking about data quality. Now, that may seem like a boring topic right off the bat, especially if you're more on the creative end, you're a content writer, or you're more interested in the creative elements of digital marketing. If you want to do data backed experience, or if you want to really know if what you're doing is working, having the right data is essential. Now for me, I am naturally more of a creative; but I'm also very intrigued by data and numbers, and know what works and what doesn't work. This is kind of what's led me on this journey to learn more about BI, going back to school, and taking some classes around that. That way, I can run experiments more effectively and know if my work is actually generating real results.

Now, there's an old adage that says, 'Garbage in, garbage out.' That means if you fill your Analytics software, or you fill your business intelligence Tool with bad data, you're going to get bad answers. You're not going to get the answers you want. It doesn't matter if you have a lot of numbers. If the numbers don't line up, or the numbers were recorded wrong, or maybe there are errors in your numbers, you're not going to get the right answers.

You're really not going to know if what you're doing is working or isn't working. This is the importance, whether you're using Google Analytics, whether you're paying for an expensive business intelligence software. It doesn't matter what your solution is or how much you pay for it. If you have bad data, you're going to get the wrong answers. How do we get the right data? How do we know if we have the data we need to answer the questions we want to answer?

Well, Stephen Covey has an amazing book. If you haven't read it, I highly recommend it. It's called 'Seven Habits of Highly Effective People.' One of the habits is beginning with the end in mind. Now, this is the same advice I want to give to you. When you're doing a data-backed experiment or if you're wanting to set up your analytics for them to make sense for you, know why you want those analytics. Know what questions you're trying to solve problems for, or what answers you want to get. When you understand what you want, you can then reverse engineer the process and understand what you need.

Now, there's a little bit of a flowchart here. It's just four things and four questions that you want to ask in kind of this reversed order. Instead of setting up and testing, you want to really start with the end in mind. What do you want to figure out? Do you want to learn what your website users are doing, or your app user doing? People working on the phone. Is this for financial numbers? What are you trying to solve? What problems are you trying to find answers for?

Once you've understood that, the next step is what data will help you find those answers? A lot of times, we're collecting data we don't need or data that really has very little value to our business; but we're collecting it just because we can. Now in some cases, this is good because you can find really interesting nuggets of information you wouldn't have otherwise found. The reality is 90% of what we're collecting is trash. It's just useful in any business sense. Knowing what you actually need to collect is going to be important, because then you can put precedence on those entities, those attributes, those metrics that you really need to know something about; and make sure that you're collecting that data the right way, and collecting it cleanly.

Then you have to know how can I access this data? Is it website users? Can I use Google Analytics? Is it financial numbers? Do I need to have a financial software, like QuickBooks, or FreshBooks, or something along that line? Is it something a little bit more abstract where I need to find a specialized tool to really get those pieces of information? Do I need to use a scraper and extract that information from a website? Where is this data that you need at? Then, the next step is how do you get it? It's about setting up and testing.

A lot of times, what we'll do is we'll put a tracking code on our site. Maybe we've got a new product, maybe we're using HubSpot or Google Analytics, or maybe we're using a specialized BI program where we're tracking and setting up information that's dumping into our analytic software or to our database. We don't really take time to test to make sure that the data's right. This is where a lot of the problems happen. This is where the problems have happened for me before where I'm collecting all this data, but maybe we've set something up wrong or we're getting the wrong data, and now it's really useless. I see this all the time when we implement Google Analytics and Tag Manager. Well, then you have two Google Analytics scripts running, one in Tag Manager, one in Analytics. Now your bounce rate is all off, or you're not collecting hits the right way and your traffic's just wrong because you've got two different scripts that are contradicting each other; and now you're getting the numbers you need, the information you need.

This happens because we don't test. We should put things into test environments and see how they're responding, and see what information we're getting back, and then execute. The worst thing that can happen is you can install a tracking code on a website, think you're collecting information; come back 30 days later, realize it was wrong and the tracking information was off. Unfortunately, that's happened to me more than once. That's where often you learn these things; by the mistakes, where you set something up and you go to look at the information. It's not there, or it's not the information you thought you were getting, because you didn't take the step to say, 'What do I need? What's the data that I need? Where is it at?' Then set up and test it to make sure that it works.

Another thing that I see that happens, especially today with so many different tools, so many different programs out there. These are really great tools, and most of the BI tools out there are front-end tools. They're tools that are going to help you visualize your data, or display your data in a visually appealing way. We end up making really cool graphs that people still don't understand. We make really cool graphs that we don't understand. We assume that these all-in-one tools are going to solve all of our problems; but a lot of times, an all-in-one tool isn't really an all-in-one tool. We assume that if we buy the subscription, automatically our data is going to show up in the right place; and it's going to do what we want it to do.

I've been looking to automate this process myself, and I'm sure there are ways to automate it and I know there are ways to automate it. It's typically a very siloed into different stages, where you're extracting the data, then you're loading it into a data platform or a data warehouse. Then, you're pulling that data into your different front-end tools. That's a very, very narrow look, not really getting into the details of what's going on; but each of those processes is extremely important. We assume that if we have a quote-unquote, 'all-in-one tool.' It's just going to do everything for us. Our data is going to be clean. It's going to go into the right place, and we're going to easily be able to find it.

Well, that's not really the case all of the time, or most of the time; because BI can be pretty complex, but you can simplify it by giving the numbers that you need instead of overloading yourself with stuff you don't need, overloading yourself with tools that you don't need. Often times, we're using a jackhammer to solve a problem that we could have easily fixed with just a small hammer, and tapping on that nail a little bit. Before you invest in these big tools and these big all-in-one solutions, make sure you know what you really need. Make sure you know where that data is, make sure that you're pulling that data out in a way that it's going to be quality data. Then, you can actually find the answers that you're looking for.

Data quality gets way harder the more and more you add to your business intelligence process, to your analytics process. If you can slow it down and just really start with what you need, you're going to find that keeping that data at a quality level is going to be much easier. There is no such thing as set it and forget it when it comes to BI.

Things are moving too fast. Data changes, user behavior changes. You need to make a tweak to your study. Really, if you're doing this more from an agile standpoint where you're running A/B tests, you're testing different aspects of your marketing campaign or your business, and you want to be able to test them quickly, get information quickly, and pivot quickly, you can't set it and forget it. You need to be able to access this data and look at these decisions and be able to make decisions fast. The only way you can do that is if you have good data. That means you have to set up the process in the right way in order for you to extract the value you need out.

If you have any more questions, maybe you have questions on Google Analytics, that's one of my favorite tools. The Google Cloud environment is a very powerful environment. It's free. It's not set it and forget it. There are some tools that can automate it and help you work a lot more efficiently inside there. If you have questions and want help setting this up, please let us know. We would love to help you get on the right foot. Or, how does this actually impact a marketing campaign? Again, please comment below. Give us some questions. We would love to help you in this process and make sure that you get the most information that you can to make intelligent decisions to grow your marketing campaigns and your business.

Until next time, Happy Marketing and thanks for watching.

marketing metrics



This post first appeared on Https://www.smamarketing.net/, please read the originial post: here

Share the post

The Importance of Data Quality (Garbage In, Garbage Out)

×

Subscribe to Https://www.smamarketing.net/

Get updates delivered right to your inbox!

Thank you for your subscription

×