How dirty is your data?
Good data from media metrics can help industry pros uncover valuable insights and demonstrate PR’s importance to their organization.
Inaccurate or corrupted “dirty data” can lead to wildly incorrect conclusions, tainting presentations and even ruining entire campaigns.
Follow these expert recommendations for collecting, maintaining and measuring data:
1. Use Excel rather than Word. Excel enables you to divide data by category into rows and columns that can be cross-referenced and dissected with pivot tables, says freelance media analyst Steph Bridgman. List each new media item in a new row, and divide coverage up into categories. Organizing data into rows and files makes it easier to produce valid analytics about media coverage.
2. Use Boolean search terms. A fast-food chain once collected 4,000 extraneous references to underground transportation systems that had no relevance to the business, producing a ridiculously large reach figure, recalls PR measurement expert Katie Paine, CEO of Paine Publishing. Most all media monitoring services, including free tools such as Google Alerts, offer Boolean search queries that can eliminate extraneous results. Simply enter “not” before a word to exclude the word from results. You can write “and” between search terms to include both words in any order.
3. Eliminate duplicates. Take advantage of Excel’s Remove Duplicates option. Go to the Data tab and click “Remove Duplicates.” Select a column with a unique number, like “Clip ID,” or “Item ID.” If that doesn’t exist, use the unique URL of the article and click OK.
4. Periodically prune metrics. An organization might accumulate a glut of metrics over time as team members add their opinions. Some metrics may be obsolete, irrelevant and even dangerous. In addition, a vast supply of metrics becomes time-consuming to track, report and analyze. Once a year, or after a change in leadership, review the metrics. Eliminate useless figures; keep only those linked to business objectives. “It all comes down to one thing: Does the metric help you make decisions? When you see the metric, do you know what you need to do?” says Lars Lofgren, KISSmetrics marketing analyst.
5. Establish standard terminology. Different data providers and vendors might apply different definitions for the same metrics, causing confusion. Gather your team to agree on specific definitions for key terms, relying on industry benchmarks such as the Media Rating Council standards for social media or the Institute for Public Relations standards for PR, Paine advises. Then post the lexicon where everyone can access it.
6. Audit data. Audits check entries to ensure they are accurate and relevant. They’re especially important if you combine multiple sources. The most frequent error in combined data sets is misaligned dates. Make sure date formats are consistent.
7. Use an established media monitoring service. Doing so can be more effective overall, in addition to saving your organization time and money.
A version of this post first appeared on the Glean.info blog.