The Power of High-Value Target Analysis on Optimization Efforts

From a military perspective, a high-value target (HVT) is an enemy target that is considered an essential capture to completing operations against a significant foe. Likewise, from a technology perspective, a HVT represents an essential targetable task that is necessary to achieve an objective that may provide crucial competitive advantage and leverage. In the realm of search performance, HVTs come in the form of influential factors that enable an organization to optimize ranking, traffic, conversions, and engagement through a focus on key signals such as optimization, quality, freshness, and relevance.

While a number of tools on the open market are available for organizations to help identify which variables might potentially influence their ranking position and onsite engagement, the reality is that few tools or analytics platforms provide definition or context resulting in vague interpretations and observations on high-value target opportunities. Few, if any, are capable of determining, based on rank position, how much traffic, conversions, or engagement are actually apportioned to specific HVT influencers. They tend to lack in the ability to accurately gauge and forecast, when HVTs are implemented, how much new traffic may be earned through a search engine such as Google with a statistically significant confidence level.

Often times the justification for such a lack of critical insight is blamed on the inability to peer inside PageRank, the enigmatic black box that is Google’s Ranking Algorithm. Perhaps fear of the unknown has acted as a barrier at identifying innovative ways to penetrate these perceived obstacles. However, there is always a way to get beyond such fears and overcome the seemingly impossible. There is a wonderful proverb, “Fear knocked at the door. Faith answered. Nobody was there.” Through faith in research and analytics, we can overcome this fear of the unknown and establish a process which enables transparent line-of-sight visibility into a range of highly influential HVT factors that, when implemented, may provide significant lift in traffic, conversions, and engagement.

Google logic isn’t completely hidden in a proverbial black box. Did you know that Google’s PageRank algorithm is patented? Patents are freely available to read and evaluate. One may glean tremendous insights from a read through any of Google’s Patents on PageRank or other related tools. For example:

Don’t just settle for reviewing patents explicitly on PageRank, though. You may find associated and relevant insights by evaluating other Google patents as well such as for Google Panda, Google Maps, Google’s Query-Based Circles, or any new patents they may continuously be added. One such example is with Google’s Panda patent release on March 25, 2014.

Before the release of this patent, analysts could only speculate how brand authority was measured by Google. Wording in the patent revealed that Google would identify which terms users searched for and clicked on. This would enable Google to codify and map brand mentions in queries to identify a prioritized index of authority for each mention. Page quality could then be evaluated and measured based on the commercialized intent of the content regardless of whether the mentions are called out as hyperlinks or non-links.

From an HVT ranking influencer perspective, another example is the emerging prominence of content readability using Flesch reading ease algorithm. The equation for Flesch Kincaid readability is used as a weighted ranking signal, and is derived through an evaluation of syllables and sentence structures in on-page content. While a few rank optimizer tools have started to take notice of this algorithm over the last couple years, one may note that Google included the use of the Flesch algorithm in their patents several years ago.

Coupled to contextual evaluations of technology patents is the significant value of staying current on Searchmetrics Ranking Factors. Every year Searchmetrics releases a list of high-value targets that influence search engine results page ranking and their associated spearman correlation coefficients. Their analysis provides a statistically significant view of how particular HVT ranking factors are trending, as well as how they are correlated by ranking position. While this study tends to provide general industry influencers, there is a way to tie the results directly to your own organization’s needs.

Armed with a list of indexed HVT ranking influencers and their associated Spearman correlational coefficients, it’s time to apply that data to your own site. The first step is to conduct an audit of your targeted pages to identify which HVT factors should be implemented. One way to accomplish this is to breakout the HVTs into categories of concentration, for instance: Business Signals, On-Page Signals, Link Signals, Review Signals, Social Signals Behavioral Signals, etc.

When you audit your target pages, you may use tools to programmatically scrape the code from your target pages, and then identify and isolate HVT gaps. You may use regular expression patterns to flag HVT gaps and alert you. Such coded macros are especially helpful if you have more than just a handful of target pages to evaluate. By the end of this phase, you should have a clear view of which HVTs have been implemented and which ones have not been.

Next, run a time series regression forecast model against a look back window of two to four years of historic traffic and conversion data. The goal is to forecast monthly, one year forward. Ideally you will want to ensure that you’ve already discussed conversion goals with critical stakeholders to ensure your evaluation is highly aligned with your organization’s needs and end-state goals. With seasonality factors identified through an evaluation of slope and intercept, you will have the right data you need to drive actionable HVT insights.

For the list of HVTs that your targeted pages stand in need of, isolate the spearman correlation coefficients (e.g. ranking influence parameter) for those tasks, and blend with your current traffic and conversion data, along with your historic regression data adjusted for seasonality, and identify apportioned traffic and conversions by each needed HVT factor. Identify your non-parametric Spearman Rho Coefficient of Determination (p^2), which is the proportion of shared variance in ranked variables. Forecast this data forward by three-month intervals to establish look-forward benchmarks for outcomes in anticipated apportioned traffic and conversions for HVT tasks that need to be implemented.

Ultimately, HVT analysis will provide you the crucial evidence and newly established milestones you need to convince executive management that it’s worth the resource bandwidth and budget allocation needed to move the needle on optimized ranking and on-site engagement performance with the anticipation of newly acquired traffic and conversions for targeted pages by task. More importantly you will have established a roadmap of prioritized critical tasks to focus on through the upcoming year with the specific intent of boosting ranking and increasing revenue. In a grand finale style of classic Hollywood endings, you will find yourself riding into the sunset after having saved the day by telling a compelling data-driven story, telling that story right, and telling it right now.

Related reading

Business searching data infographic flat design vector illustration
Simple Share Buttons