The Domino Effect of Bad Data: 5 Reasons Why Your Data Is Wrong

  |  February 1, 2012   |  Comments

When the foundation of your data is unstable, all interpretations and assumptions made using the data begin to crumble.

Many marketers are so hungry for data to prove a campaign is working, a segment exists, or an investment is effective, they don't invest enough upfront time in ensuring proper data collection and integration. But when the foundation is unstable, all interpretations and assumptions made using the data begin to crumble, and the accuracy of your insights and projections often falls faster than a row of dominoes.

There are five primary drivers of bad data:

1. Lacking a coherent integration strategy. Far too often, data integration starts with a casual conversation about how it'd be nice if that offline sales data could be integrated into the web analytics platform. Of course, everyone thinks that it is a fine idea, and soon after, something is slapped together and your data is "integrated." As time goes on, more often than not problems start to emerge. You don't actually have all the data you need to make decisions. The data you're importing isn't actually tied to any of the data in the web analytics platform. The overall setup doesn't allow for meaningful analysis leading to clear action. All it does is allow you to overlay a chart of your offline sales with your online traffic. #fail

Taking a step back, any data integration platform needs to start with a solid plan. Talk to all of the stakeholders who will be using the platform to define the business questions it should answer, the comprehensive list of data sources to integrate, the plan for how that integration will take place, and what views will be needed of the data. Start with a robust, solid foundation that plans for the future and anticipates growth.

2. Assuming data integration is a web analytics issue. Web analytics software is great, but it is simply another data collection tool and another component of your integration. Do not approach integration as web analytics plus a couple of other things. You need to think about all the other data sources that are just as critical to true measurement success and weight them appropriately. More clients than I care to admit try to force all their data into a web analytics-based construct rather than pull that data out and put it into a more ideal and robust data structure.

3. Garbage in, garbage out. This one should be clear to everyone, but too often it is overlooked. Your data is only as good as the component inputs. Partners, vendors, customers - they all have different standards for collecting and tracking what should be the same attribute. If they're measuring things wrong or have other measurement inconsistencies present, then your integration is going to have little value.

4. Critical data sources are missing. While the quality of the data is critical, what data sources you incorporate is equally important. Many integrations stop once the convenient data sources are completed or reach the boundaries of what data sources the digital marketing team owns. Integration projects need to be enterprise-wide to truly add enough value to achieve the desired ROI and understand the entire the impact of your marketing. If your digital channel is driving offline sales, then those sales need to be included in your integration. If your online sales are driven by print or direct mail campaigns, that information needs to be included as well.

5. Does not evolve with the business. Targeting and measurement technologies are evolving at a furious pace and your business is changing as well. Data strategies need to be based on a flexible schema to allow for growth and change. New marketing tactics will be added to the media mix. Two years ago, Facebook Advertising and Insights data wasn't on the radar for many companies. Now it is critical marketing intelligence. Don't look at data integration as a point-in-time project but instead as a continuously evolving campaign. Every quarter, upgrades should be made, data sources changed, and overall functionality enhanced to stay current with business needs and industry capabilities.

While it may be less glamorous than ideating the next viral campaign, having a solid foundation will ensure you are focusing on the right audiences, with the right media.


Andrea Fishman

Andrea Fishman, VP of strategy and a partner at BGT Partners, leads BGT's Chicago office and has extensive experience in marketing and management consulting. She and her team drive value to BGT's clients through the development of behavioral marketing programs, web analytics, measurement programs, industry benchmarking, competitive assessments, and the design of integrated marketing programs.

Andrea has been with BGT since 2003 and is credited with strengthening partnerships with such clients as ADT, Sony, ADP, and Avaya. Prior to joining BGT, she served as global vice president at divine, inc. She's also held strategic positions within marchFIRST, The Lewin Group, and the office of U.S. Sen. Edward Kennedy.

A graduate of Brandeis University, Fishman was awarded the Wasserman Scholarship for academic achievement and was named a 2010 Stevie Awards Finalist as Best Executive in a Service Business. She is a frequent judge for the eHealthcare Leadership Awards and is involved with the Special Olympics and Chicago Cares, a community service organization.

COMMENTSCommenting policy

comments powered by Disqus

Get the ClickZ Marketing newsletter delivered to you. Subscribe today!



Featured White Papers

2015 Holiday Email Guide

2015 Holiday Email Guide
The holidays are just around the corner. Download this whitepaper to find out how to create successful holiday email campaigns that drive engagement and revenue.

Three Ways to Make Your Big Data More Valuable

Three Ways to Make Your Big Data More Valuable
Big data holds a lot of promise for marketers, but are marketers ready to make the most of it to drive better business decisions and improve ROI? This study looks at the hidden challenges modern marketers face when trying to put big data to use.