The past 12-18 months have seen a great deal of attention around the concept of Big Data in the marketing world. The numerous ways in which Big Data promises to increase insight into online behavior, customer profiling and more has been written about extensively. At Conductor, we agree that Big Data will lead (or has led) to new levels of insight for those marketers who choose to leverage its capabilities.
But, any marketer digesting the ongoing Big Data commentary over the last 12 months or so could be forgiven for drawing the conclusion that when it comes to Big Data, the only requirement for a marketer’s extricating valuable insight from the mountain of data is… data, and lots of it. That is… all that is required to produce actionable insight that will positively impact a site’s traffic and revenue is gobs and gobs of data… or so the thinking goes.
However, this kind of thinking — large quantities of data without context and container — results in a marketer who is data rich but insight poor. We’re going to use this article to demonstrate one way in which a marketer — without changing data quantity or sources — can make small adjustments to how her data fits together, and vastly increase her insight into her online traffic.
For illustrative purposes, we’ll use the example of ‘Ann,’ the SEO Manager for Acme Co., a medium-sized clothing retailer with both a brick and mortar and online presence.
The Typical Non-Integrated View Of SEO Metrics
As the SEO Manager for her company, Ann is responsible for both the SEO practices that will guide their tactical approach, and measurement of the metrics that will identify the impact the strategy is having on natural search visibility and, ultimately, traffic and online sales.
She tracks search rankings weekly for her most important keywords, and monitors traffic and conversions in Web analytics software. She manages a small team consisting of an analyst and two SEO tacticians.
When it comes to measuring the impact her SEO activities are having, while she is able to mostly track natural search visibility changes as a result of activities, and, separately, is able to measure changes in traffic and conversions, she has difficulty tying the two together.
She has a sneaking suspicion that in visualizing the two (rank data and traffic/conversions) independently, rather than in one cohesive screen, there might be insight she is missing that could tell a different story about the true impact of her SEO activities.
Put another way, she has all the data at her disposal, but realizes she may not be bringing it all together to gain the insight she needs.
Building An Integrated SEO Metrics View
To test the hypothesis, that there is insight she is missing by not having her KPI’s together in one place, she exported traffic, conversions and revenue data for two months and charted rank-data on top of it.
Next, she annotated major activities that she anticipated would have had an impact on natural search visibility and a corresponding impact of varying degrees on traffic, conversions and revenue.
When she finished, she sat back and saw that the exercise had clearly revealed what had previously not been obvious. She learned that while ‘SEO actions A’ had resulted in a strong rankings improvement, it had led to a smaller than expected corresponding growth in traffic, conversions and revenue.
Conversely, activities in ‘SEO actions B’ had resulted in smaller rank improvement than the actions following ‘A,’ but had resulted in considerably more growth in both traffic and conversions/revenue.
When Ann dug into it further, she saw that in ‘SEO actions A,’ her team had focused intently on on-page optimization and link building activities. These activities propelled the keywords up the search rankings in only a few weeks’ time.
However, they did not result in the traffic and conversion/revenue growth she would have expected based on the keywords’ search volumes, rank and corresponding click-through rates based on their rise in ranking position.
Upon closer examination of ‘SEO Actions B,’ Ann saw that her team had stumbled across the discovery that the snippets on the search pages for many of the keywords were poor, resulting in a much lower click-through rate, thus accounting for lower traffic than expected.
Further analysis showed that the team had also discovered that the landing pages for the keywords were poorly optimized, geared to generic brand searches rather than to conversion for the specific purchase queries.
After optimizing the landing pages to line up better with the specific optimized keywords, conversions and revenue improved, hence the post ‘SEO actions B’ increase in traffic, conversions and revenue she was seeing on her chart.
Integration Of Data Sources Reveals New Insight
After completing the analysis, as Ann sat back to think about what she learned from this exercise, she recognized that, in some ways, she was lucky her team had stumbled across the snippets and conversion issues that had curtailed her keyword’s traffic and revenue.
She realized that, going forward, if she could combine the data together into one view, she could be better assured of these kinds of issues not slipping through the cracks undiagnosed and impacting her company’s bottom line.
She also resolved to explore other ways in which she might bring diverse data sources together for increased insight and began researching technology solutions that might automate the process and also introduce new ways to leverage her data for actionable insight.
Use The Data You Already Have To Gain Insight
The length and breadth of data search professionals collect and analyze can vary greatly based on their maturity level. No matter your maturity level, however, there are likely to be instances whereby making a small change in how you view your data substantially increases your insight into what is and is not working — in manners both similar and dissimilar to the example scenario described above.
Ultimately, you may find that this newfound insight can have a real impact on your bottom line.
Opinions expressed in the article are those of the guest author and not necessarily Search Engine Land.