Boolean Queries For Multichannel Competitive Monitoring
The true secret of competitive monitoring is defined by the keywords being monitored. Most of us are well aware of search operators, which provide search engine users the ability to refine traditional search queries. Boolean queries, whenever possible. Think of it like this: Share of voice queries are like a horse…crap in, crap out. Look at the wrong keywords and you’re blind. Fail to filter keywords by Boolean queries when available, risk missing true focus in competitive analysis.
Building Boolean Queries
Building queries can be fun. The syntax is fairly easy to master and oh so powerful. Sysomos MAP applies Booleans to search, social, mainstream, news, blogs, forums and other channels. Let’s have a look at how segmenting data with Boolean queries can advise strategy and tactics.
The queries below marry a) bucketed brand term permutations (I.E. “American Express”) and major travel concepts like “Cruise.” In other words, we’re not just after brand mentions. We’re measuring SOV for brand terms filtered by important KWs and concepts. The first competitive metrics, brands’ share of travel or hotel mentions zoomed out is a great example. Many more blog, forum, news and Twitter users mentioned American Express as pertaining to travel and hotels.
The next level of granularity offers strategy ideas. Let’s say I’m MasterCard®. I’m getting my ass kicked most everywhere and do reasonably well in forums. Amex is rocking the blogosphere sporting big numbers of mentions associated with travel and hotels. Obviously one strategy MasterCard needs to take should involve more content distribution to bloggers of various types.
Now get tactical. Which blogs are kicking our ass? It’s interesting how MAP pulls data by mid-metrics like “Some” authority. Distributing content to these types of blogs can earn attention at reasonable authority. They could be enthusiast bloggers or lower to mid level professional.
Boolean queries are powerful tools to filter chatter by sentiment. Some tools have built in sentiment algorithms. That means there are preset bad words that trigger the sentiment analysis. The screen cap below shows the automated sentiment comparison in MAP for the AMEX/MasterCard travel and hotel query.
Don’t ever totally trust these scores. Michael Jackson was “Bad” and “Dope” means groovy. AIMCLEAR has a list of positive and negative stop words that signal sentiment. We often build them by hand.
For the various channels MAP covers there’s cool data. Here’s a breakout of blogs by age band. Cool. That tells us who we need to create content for.
Obviously the example queries are simple. Sometimes it takes lots of code, including nested Boolean elements. The power is clear.
No matter what tools we use to study share of voice, the measurement has to be tied back to original goals and KPIs. It’s no longer enough to stare at each area – SEO, PPC, social, etc. – as silos that do not impact the performance of other digital disciplines. Start by defining what share of voice means for marketing teams, and then identify which tools and metrics are most important to measuring the success of your efforts.
Understanding that competitor data isn’t evenly available and not all analytics platforms calculate true SOV, leave room for contextual analysis, and tie those numbers back to direct marketing initiatives. Thus, we establish true share of voice ROI (SOVROI for those who need an acronym to feel whole).
Living in the dark when so much competitive data is readily available is a recipe for half-ass performance. Savvy marketers-of-the-future will implement share of voice measurement to influence future digital performance in every aspect of business – be it an internal SWOT analysis, forecasting needs of their qualified communities or justifying additional budget to close a gap. They already do.
Obviously some readers will have opinions regarding tools, techniques and thinking uncovered in this post. Hopefully these writings will impact your understanding of SOV in a positive way.
