The Harvard Business Review reported that data users waste 50% of their time finding and correcting errors and attempting to confirm data sources they don’t trust*. In addition to the risk implications of using bad quality data, such as missed opportunities and loss of reputation, the financial costs can also be great. Last year, bad data quality cost US companies $3 trillion in wasted expenditures*. Attribution, DMPs, algorithms and predictive modelling bring new and exciting opportunities, however bad data quality can undermine the success of not only these projects, but also the business and its bottom line.
We believe that data governance and data quality are the foundations of analytics, and where businesses will get the most value for money. Patch-work fixes are a drain on resources and overtime become very costly.
One of the largest external data sources feeding into analytics comes from campaign and marketing activity. Data from channels such as display, email and social are constantly feeding data into analytics, and need to follow a consistent format that matches the rules set-up in Adobe Analytics. Commonly, marketing activities are managed by different silos within an organisation or through multiple vendors servicing the different teams. Each team may have their own process for rolling out campaign activity, which doesn’t follow an organisation-wide structure for analytics, resulting in lost data or inaccurate reporting.
Companies need an organisation-wide campaign governance process, that takes into consideration the core needs of the whole organisation. Without organisation-wide campaign governance, unnecessary hours are wasted analysing dirty data, incorrect decisions are made, return on investment drops, and trust in analytics is lost. Companies often times want to jump straight into the newest analytics trend, but the best investment an organisation can make is in the quality of their data.
Want to talk to us about data quality? Get in touch!
Sources:
https://hbr.org/2013/12/datas-credibility-problem
http://www.ibmbigdatahub.com/infographic/four-vs-big-data
