For many brands, regionalization of marketing programs represents a key pillar to their overarching strategy. Particularly for those companies operating nationally, it’s often difficult and costly to target consumers on a very granular level.
Regionalization allows marketers to look at trends across a slightly larger area — typically several adjacent states that share some cultural elements — and create campaigns that are more relevant for consumers in that area as compared to national initiatives. However, this strategy can overlook some unique qualities of individual states and cities.
To help address this issue for marketers as it relates to the online space, Chitika Insights released an interactive map in early January examining state-level usage rates of desktop operating systems and Google usage.
As part of this article, we’ll be focusing on the OS-level statistics. For each of these data sets, Chitika Insights examined tens of millions of US-based online ad impressions from November 20 to 26, 2013. We then used each impression’s associated IP-address to glean the user’s location within the country.
To begin, let’s review the data and some associated trends.
Regarding the OS data as it’s currently presented, Washington State exhibits the highest rate of Windows usage — not too surprising, given that Microsoft’s base of operations is located there. However, Windows usage is fairly pervasive across the US, with users of Microsoft’s flagship OS generating more than 70% of all desktop traffic in every state.
Meanwhile, users from Vermont and Hawaii browse using Mac desktops/laptops at the highest and second-highest rates in the nation, respectively. Vermont is unique in that Mac usage represents more than 1/4 of all web browsing from the state.
Finally, Linux usage rates consistently hover between 0.5% and 2% nationwide, but there are a few notable outliers. Wyoming users exhibit the highest Linux usage rates, nearly doubling the rate of second-ranked Virginia, likely in part due to Cheyenne, Wyoming and surrounding communities being home to a number of data centers and associated personnel.
CTR & Browsing Patterns
Now that we understand some of the notable takeaways, to attain some actionable insights, we’ll need to view these data in the context of some other recent studies on CTR and browsing patterns.
Generally, earlier studies have found a correlation between the likelihood of a user to engage with a campaign and the operating system version being used.
In an earlier study, users of older operating system releases (e.g., Windows XP as compared to Windows 7) had a higher click-through rate than users of more up-to-date versions. This trend was more pronounced in Mac users than in Windows, with users on Mac OS X 10.4, the oldest Mac version studied, having the highest of all studied CTRs.
So, how does this relate to the state-level data when it comes to improving campaign strategy? If you know that the state or states in which you’re targeting have a high rate of Mac desktop usage for example, you can look to create campaigns that take better advantage of the strengths of that OS (e.g., an App Store download tie-in) along with a better estimate of the potential for campaign engagement based on the most current OS version distribution.
Additional insight into where and when campaigns see the most success can be found by understanding the daily browsing habits of users. In an earlier study, Chitika analyzed hourly web browser activity broken down by platform.
The difference in activity showed very marked fluctuation, notably around standard business hours. Desktop usage in particular showed the most significant fluctuation, with activity dropping to under 30% of its daily high during the early morning hours, then by and large sustaining its maximum level from 10 a.m. to 12 p.m. PT/1 p.m. to 3 p.m. ET.
This kind of knowledge in conjunction with the state-level data can help marketers strategically plan the timing of campaigns to best target consumers in a given state(s). Creative marketers have gone even further by leveraging data collected through internal website analytics tools to better monetize their different audiences.
Travel site Orbitz famously decided to rank its offerings differently based on a user’s OS, location and other factors in order to increase spend on the site. This kind of hybrid model — using a combination of internal and wider trend data, like the state-level figures outlined here — represents the next phase of “regionalization.”
Marketers understand where they want to target and the commonalities within a given region as they relate to online usage, and then best interact with those visitors based on internal KPIs and engagement patterns.
Opinions expressed in the article are those of the guest author and not necessarily Marketing Land.