One client trend we’ve seen at Engage during the past few years is a growing desire for a Share of Voice report that measures a company’s brand name and keywords as a percentage of the overall industry news coverage of its market sector. When starting a PR program, immediately executing a Share of Voice report can be a great way to establish the baseline level of press attention an organization has received to date. Later follow-up reports can allow you to properly measure the growth and success of your media relations program and thought leadership initiatives as you move forward over time.
Engage has developed several unique approaches to Share of Voice reporting that vary based on the amount of time needed to conduct a report and the client's budget. Regardless of your approach, these four rules will keep you focused on the right metrics instead of getting caught up in the minutiae.
1. Establish the right source selection and stick to it
With Share of Voice reporting, it’s easy to compound your work by opening up your source selection too broadly. You may be interested in every instance when the media mentions your company, but today the Internet is full of websites that may pick up your news story but have no value whatsoever to you or your company. In our experience, news aggregator websites (Topix being one example) can often distort the number of press hits by acting as an echo chamber, reposting the picked-up news article but adding no material value to it. A Share of Voice search using a too-broad selection of sources could skew your results and reflect a larger - but inaccurate - share of voice for your organization. Furthermore, it’s extremely important once you determine source selection that you continue to pull from the same sources in the future, so each report provides an apples-to-apples comparison with previous reports.
2. Formulate the right search terms: The Apple dilemma
Finding out how many press hits a given keyword generates is easy - for example, if I wanted to gauge press coverage for Microsoft, I could just search “Microsoft.” But other terms, such as Wi-Fi, aren’t so clear cut. Some journalists spell Wi-Fi differently (e.g., WiFi, wifi, and Wi-Fi), so it’s important that you not accidentally exclude such variants in your search processes. Searches are even more frustrating when a company has a generic name, like Apple; press searches may turn up instances of news about the actual fruit instead of about the tech giant. Many search services have Boolean operators as one option to empower the user and ultimately prevent this problem, but you still have to know the right way to formulate your search string. Studying up on your Boolean terms can certainly pay off when executing your Share of Voice report.
3. Decide when to count your own press releases in a report
One debate in Share of Voice reporting is whether or not your own press releases should count toward your own metrics. For example, if Microsoft issues a press release, should the mention of the word “Microsoft” in the announcement count toward Microsoft’s Share of Voice in the media? There is no right or wrong answer - it just depends on the purposes for which you’re running a Share of Voice report. Are you aiming to measure media uptake of your new messaging or just trying to gauge what percentage of media content as a whole (including press releases and press release pickups) your organization comprises? Based on your needs and the purpose of the report, we might advise either excluding or including your own press releases. However, we generally advise against including self-generated data because of the natural propensity to skew the results in one’s own favor. Regardless of how you decide to measure your own content, beware of how your own press releases (and the keywords in your boilerplate, for example) could be returning biased data that results in an inaccurate Share of Voice.
4. Determine if automated media monitoring services will accelerate your report process
Producing Share of Voice reports can be very time consuming, depending on the number of keywords used and competitors included in the search. It’s only natural to try to save time and money by going the automated route and using any one of the many media monitoring services hawked by companies like Meltwater. Meltwater is a fine service, but many services have not kept up with the technical capabilities such as natural language processing and highly efficient automation that are needed for accurate Share of Voice reporting. Too often, we find automated solutions have not been properly configured to return the most accurate results, and in some cases, the automated solution is so flawed from a methodological standpoint that we need to restart from square one with a fresh approach. In addition, services that promise sentiment analysis are still in their infancy—more often than not, you’ll find 99 percent of your coverage was rated ‘neutral’ when in fact it wasn’t; the system just isn’t smart enough to tell the difference yet. So while automation can be a good thing, for Share of Voice reporting we caution clients not to leave out the human element. And when you’re presenting data to key stakeholders, be sure that an educated human being has curated the results first.
In sum, it’s important to make sure not only that you’re gathering metrics about your PR program’s success but also that you’re gathering the right metrics.
Are you interested in Share of Voice for your organization? Contact us at firstname.lastname@example.org and we can help. If you use Share of Voice reporting for your organization, we’d love to hear your tips—leave us a comment or send us a Tweet at @EngagePR!