Take the step, schedule a demo.
We've spent a decade modeling search engine behavior so you don't have to guess anymore.
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Market Brew’s highly-correlated search engine model reveals how the “online colleges” organic search results work.
When it comes to higher education, it’s no secret that the trend everywhere is toward remote learning and online programs. With the obvious limitations of cost, geography, class size, and available resources, it makes sense that traditional universities the world over would be pushing to expand their online course and degree program offerings. Aside from the obvious convenience it affords the students, this shifting environment also enables all those “for-profit” institutions out there an “easy” way to boost revenues through an every growing base of students interested in seeking their education online.
I say “easy” because as anyone who has ever tried to compete online for website traffic in a highly competitive industry knows… getting to the top of a search engine for a keyword like “online colleges” with 228M competing pages is no easy feat.
Without the help of powerful seo tools this kind of task would be hopeless for the less sophisticated. Yet, for those lucky few who are smart enough to jump on the cutting edge of search analytics technology by utilizing an invaluable scientific marketing platform like the one produced by the team at Market Brew, they would be able to crack the code of the search engines and seriously compete in this highly competitive industry of online education and would stand to gain a lot of money for their efforts! You can see from this chart (below) that even traditional brick and mortar institutions make a sizeable chunk of profit per student.
Of course, this chart is taking into account all the costs and spending of the traditional university model as well, but what if we just look at the purely online universities and their profits/spending…would we see anything different? Well according to a private investigation into the for-profit higher education industry conducted by the U.S. Senate’s Health, Education, Labor and Pensions Committee (HELP) led by Senator Tom Harkin, of the 30 for-profit institutions the committee investigated, the average spending per student was only $2050, and specifically regarding the University of Phoenix, their parent company Apollo Group spent just $892 per student. With these new costs in mind, you can imagine how much more profit these kinds of institutions stand to gain with every new enrollment. Source: American Public Media.
With so much profit at stake, it’s no wonder that companies like the University of Phoenix, Kaplan, and DeVry are pulling out all the stops to make sure they are capturing as much of this online “gold” in the form of website visitors as they can. But in an industry that has no doubt matured a lot over the past few years in their online marketing tactics, is the competition over, or are there still any surprises we can glean from the current environment online for this industry?
Let’s see what we can find out by taking a look at the insights within the Market Brew predictive search platform.
In this use case example, we are going to look at the current results for one of the most competitive keyword terms in this already very competitive industry of online higher education – and that is “online colleges”. Why did I choose that? Well, after loading a few dozen keywords into the Market Brew analytics platform (though I could have used many thousands as I’m not limited at all in how many keywords I can analyze/track) and kicking off a crawl/site analysis for a few of the top higher education institutions focused on remote learning (including the three listed above), I took a look at what combinations of keywords and webpages carried the greatest ROI potential if I were to focus my time and resources on optimizing them. What kind of crazy math and wizardry did I have to perform to come to such a conclusion amidst the myriad of possibilities that I could optimize for? Well, frankly… none.
Yes, you heard me right. None. The Market Brew system was designed to make SEO a no-brainer from an ROI perspective (great for all you CMOs and VPs of marketing out there!). For now, I’ll spare you the details of the complicated math, though for you curious folk you can check out our Automated SEO whitepaper, but suffice it to say that after running millions of simulations of potential changes to the underlying metrics that determine a webpage’s organic search ranking for any of the keywords we have entered into the engine, the system will return a list of those potential optimizations (ordered by highest ROI potential), where they should be done (which webpage and keyword), and how to accomplish them (just click on ‘Switch to Task View’ to see the optimizations ordered by task with explanations and the ability to assign each task to team members).
It just so happens that the most valuable optimization I could focus on if I were working for the University of Phoenix is to improve my homepage’s performance for the keyword “online colleges.”
You might be thinking right now, “Yeah, that’s all well and good, but how does the engine really know which pages/products and keywords are the most valuable to me? ” By default, the engine will calculate optimizations based on search volume. To make the optimizations even more accurate, and more customized to your site’s conversions, you have the option to import your own data into the engine by promoting and demoting URL/Keyword pairs. This part is key to getting the most value out of the platform.
Using the Overrides feature, we are able to tell the engine which keywords and/or URLs are converting the best or driving the most revenue into the business. Input your revenue per keyword and/or webpage to customize the model even further.
Adding this key piece of quantitative information gives the system the proper context it needs to then go and do what computers do best…crunch massive amounts of data! Once completed, the system arrives at a statistical approximation that accurately simulates the reality of your organic search market and gives you the kind of insight that can ONLY BE POSSIBLE by having access to your own search engine – a search engine combined with an analytics platform that can make these calculations and decisions in the proper context of numerous changing variables (solving for the butterfly effect).
So let’s get back to the example at hand… optimizing for the keyword “online colleges”. Starting from square 1, let’s just see what our old friend Google has to say about the competition…
When doing a typical Google search and looking at the results, it may not be clear how close or far the race to the top truly is in this case. We can see that of the top 10 organic results, there are only 4 actual schools making it onto this coveted page of traffic gold. Looking at the top 3 competitors for this keyword marketplace, we can see Kaplan is pulling up the top ranking (currently 4th at the time of this writing) with the University of Phoenix and DeVry pulling up a “close” 6th and 7th finish in the rankings. But how close are they really? And how reliable are these rankings? Might these results jump around a bit, even exchanging places in the Google results from one hour to the next or one location search to the next? How is an SEO professional supposed to honestly, objectively judge this and find any meaningful signal with which to begin the long arduous process of optimizing??
Well, if that SEO professional had access to Market Brew’s search engine model he would be able to look underneath the noisy signal that modern search engines like Google broadcast, and expose the real story going on behind the curtain. Let’s take a look at what we see inside the engine:
By eliminating all the other noise and just focusing on the top competitors that matter to us in this competition to become the king of “online colleges”, we can see that while it may normally be presumable that any of these top three could potentially win out based on the Google results alone, once we look at the numbers that underlie the scoring layer of any major search engine (we allow you to model a number of different search engine environments), we begin to see that Kaplan is far and away the leader (at least as far as this keyword race is concerned). The real story here is the race for second place amongst these top institutions. Depending on when you make the search, Phoenix or DeVry may pull up ahead of the other which shows you that Google already thinks these guys are fairly close in overall relevance to the search term. But how close are they and what are their differences or relative strengths and weaknesses?
A quick hover over Market Brew’s search engine query score for each result begins to paint a clearer picture…
Now these numbers at first glance might not make much sense, but as soon as we begin to compare with the raw numbers for DeVry we begin to see where the opportunities are…
So by comparing the breakdowns between these two (you can also view these detailed metrics side by side and do further analysis by exporting an excel report), we immediately understand how the disparities in Net Total Link Flow and the Semantic makeup of each of these two pages affect the query scores and thus the distances between them making the race interesting. We already knew by looking at the other metrics underneath each result (without going into the query score breakdown) that the University of Phoenix has more Net Total Link Flow than DeVry which we can then conclude that DeVry must be doing a better job on the semantic side of things (page content and html markup) in order to be even in the running. But now that we see these hard figures we begin to understand what that gap really looks like and can hedge against it (which could mean the difference of hundreds of thousands of dollars if not millions if I were working from the perspective of the University of Phoenix and wanted to ensure that DeVry can’t pass me up anytime soon).
Keeping with the perspective of working for the University of Phoenix (someone at that administration should really buy me a beer for what I’m about to share for free 😉 ), the first thing I might want to do is get a better understanding of why we are underperforming from a semantic perspective for one of our top keywords. I will investigate how I might be able to change that scenario using the least amount of time/resources possible. In this instance I have already checked the Top ROI Optimization suggestions that the engine has identified to make sure I’m using my time wisely. I want to be working on an optimization that has the potential to drive serious value into the organization by way of pushing the needle on important areas with revenue impact potential.
I already know from the query score breakdown (QSB) that the basket of keyword terms that the engine has indicated represent the content footprint, or Market Focus, of the page in question (in this case it’s the homepage) is scoring around 25% of what DeVry’s page is scoring, so this is a major area of concern. Because the search engine is integrated into the analytics platform, I can quickly drill down directly from the search results themselves into the proper section of Market Focus data that I want to analyze. In this case, all I have to do is click on the term “university phoenix school” in the search results to be taken to more information regarding this metric.
However, search engines are a bit more sophisticated than merely doing a shingle analysis of the on-page content itself… they also consider the surrounding link graph and any clues that those inbound links may give away as to what a page could be about. But not all links are created equal (as we well know), so when we click on the Anchor Text tab above, we can see how much relevance (by way of link flow) is flowing through each linked word of the anchor text on inbound links to paint a more detailed semantic picture.
Perusing this list clearly shows the overwhelming focus on branded terms for those inbound links which sadly don’t drive nearly as much traffic as more industry focused terms. Luckily though, there is already a good amount of link flow coming in for the term “online”, we are just missing much needed anchor text link flow for the term “colleges” in order to boost the weighting of the full term “online colleges” within the basket itself:
Combining the unique phrase occurrences with the incoming anchor text link flow, we see a whole basket of phrases with the associated weightings that the search engines can then use to determine our page’s relevancy to any search query. Of course, while branded terms are quite natural and do drive traffic depending on the strength of your brand, having a Market Focus Basket entirely dominated by branded terms is not good. Serious steps need to be taken in order to shift the semantic picture of this homepage toward terms that will drive more traffic and revenue overall. Naturally, the quickest and easiest changes to be made are increasing the number of occurrences within the content for these high traffic terms and also changing the anchor text (or ALT text in the case of images) of any internal links that may already be pointing to the homepage. Of course, if possible, attempting to change the specific wording of links on other domains that may be sending a decent amount of link flow to your domain can help tremendously if you are successful in making such a change. Usually, however, a good PR piece with intelligently placed, keyword optimized links will do the trick!
Now what about the penalty situation? Is the content on this page possibly negatively impacting the pages performance in any other ways?
Clicking back to the search results again and this time clicking on the Market Brew Score takes us into a deeper look at how any potential penalties may be limiting the performance of the page.
Not surprisingly, due to the lack of any single phrase occurring significantly more than any others (which we just saw in the MF basket section), there is no keyword stuffing penalty being assessed however that doesn’t mean there aren’t a number of other potential issues going on. One of those issues which is actually reducing the ranking power of the page by almost 12% is due to content duplication within the internal pages of the site itself.
After clicking on the penalty location button for the Duplicate Content penalty, we are taken to this screen which gives us a breakdown of all the top offending pages and the average percentage of content that is the same between the two pages (each link is in respect to the homepage because that is the penalty scorecard we are looking at in this example, but you can view this for any page on your site or your competitors sites too!). With the way modern websites are designed in a template based fashion with common elements like header and footer navigation being common throughout the whole site, it is nearly impossible to totally eliminate any duplication, but as a rule of thumb we suggest keeping the level of duplication below 50% to stay in the safe zone and not risk being penalized by the search engines.
In this case, a duplication percentage of 83% is significantly impacting the performance of the University of Phoenix homepage, and though it may not be something most SEOs would think is a top concern, with the power of this platform we can see statistically significant metrics like this that quite affirmatively do impact the overall performance of key landing pages and could very easily be the difference in thousands of visitors if DeVry were to optimize any areas where they are currently underperforming and pass the University of Phoenix right up!
So basically, in just a few minutes of exploring the Market Brew analytics platform, I could have prevented the University of Phoenix from going ‘up in flames’ (forgive the pun) and saved them many thousands of dollars. But when it comes down to it, optimizing a website for any number of keyword search terms can often be a multi-pronged effort with changes positively impacting some areas while negatively impacting others. This can be frustrating and confusing for brands operating on merely a bit of non-contextualized intel from other tools and whatever historical best practices they have acquired. In this modern world of Big Data and predictive analytics, the high stakes game of for-profit higher education demands more.