Sorry SEO Specialists, You’re Going to Have to Brush Up on Your Coding and Stats Skills

For the past few years, organic search marketing has been fuelled by keyword-targeted content. But as iPullRank points out in a MOZ blog post (which I highly recommend), there’s a technical revolution going on in SEO. And it looks like one of the more predictable and long-lasting changes in the field since its inception.

Back in the mid-1990s SEO was just beginning. It was a mostly technical skillset that required creating the best site possible on an architectural and performance level, to better the chances of getting it included on web directories. Creating websites with clear link structures and fast-as-possible-performance was critical back in the dial-up days, when users would spend longer sessions on sites and rack up more page views.

Then the search engines came, and with them came black hat SEOs. Content creators at the time realized they could increase organic traffic with formulaic keyword-based writing and meta data stuffing. Users started having bad experiences with misleading, poor-quality websites ranking well in search.

Google and the other search engines have been fighting this battle ever since. Their latest weapon is advanced natural language processing: and it’s working.

There’s a Shift in Ranking Data

A SearchMetrics study actually shows a positive correlation between relevant phrases, proof words, and organic traffic; and a negative correlation with specific keyword targeting. And when you look at data Neil Patel collected, longer blog posts are consistently outranking shorter word count posts.

And although a well-written meta description can improve your organic CTRs, there’s been another change on SERPs that’s even deprecated that: Even when you code in a well-crafted meta description, Google will often override it and display text they scraped from your page. Because specific keyword targeting doesn’t work anymore, Google frequently finds better, more relevant phrases from your webpage to match whatever the specific query was. And they even put relevant and proof keywords in bold to boost visibility.

Google’s message here is clear: They are enforcing a premium on in-depth, naturally well-written content over keyword targeting.

The Tools: Data and Coffee

There are a number of SEO tools out there that give you individual pages’ SERP ranks, how your pages are ranking for specific keywords, overall domain authority, and several other metrics to help determine your website’s health in search. But all of those stats are completely worthless when you think about it. That data is collected as if searching as a brand new user with no history – how often is that going to happen in real life?

To get a true idea of how your SEO efforts are working, analytics tools are your best bet. Drilling down into the fluctuations in organic traffic for page groups and individual pages is the best way to identify problems and opportunities. You should also familiarize yourself with the more technical issues to look for, like cross-device performance and page load speed issues.

This is where coffee comes in. And by coffee I really mean JavaScript, but if I put JavaScript in the title I knew about 20% of you would have immediately stopped reading. But if you’ve looked even a little into Google Analytic’s new page speed tool, you’ve probably seen issues with the advice to “Minify CSS/Render-Blocking JavaScript”.

Google is making two things clear here: Page load speeds really matter now, and you need to correct these issues. And in a couple of years, if at this point you have to go find someone else to do this task, then you can’t fully perform your function as an SEO specialist.

Render-blocking JavaScript actually means something pretty simple. As Google is rendering the webpage in your browser, it has to stop and wait for JavaScript code to load before finishing the rest of the page. For the past few years Google has been working on ways to better-read content rendered via JavaScript, so loading scripts and indexing the content they produce is the best means of analyzing true relevance. But it’s not ideal for speed.

There are newer variations on the JavaScript language that can mitigate this issue. Angular JS is a server side scripting language, which means it renders JavaScript before pushing to the user’s browser. This helps with the time it takes for a user’s webpage to load.

Minifying CSS means to literally compress the code and remove all of the extra spaces and unnecessary code. Again, a simple process in theory, but without knowledge of how webpages are created and the tools necessary to work with them, you can end up doing real damage to a site.

And yes, this does mean that familiarity with HTML, CSS, and JavaScript is going to become pretty much mandatory. But it’s really not as impenetrable as it sounds. With free tools like Code Academy you can develop a strong foundation in these languages and more with just a few hours a week practice.

Using Developer Tools to Gather More Data

Once you have experience with JavaScript, you can leverage a powerful data investigation tool: Browser-based developer tools. Chrome DevTools is one of the more powerful examples. What this tool does is open a window on the side of your browser that shows all of the code and certain stats and elements of the webpage. One of the tool’s options is called Console, and this is where JavaScript comes in. Using JavaScript, you can interact with the website in interesting ways. For example, in the phenomenal blog post by iPullRank featured in MOZ (yes, I’m plugging this post again – it’s that bloody useful), they talk about how they were able to set a trigger to start recording whenever a user’s meta data included ‘Googlebot’. Then they could actually playback the recording of the Google bot navigating around their site.

This data is infinitely useful when combined with analytics data and overarching digital marketing goals. But it’s only one small example of what’s possible when SEOs and digital marketers take it upon themselves to become specialized in how websites are built, maintained, and evolve over time.

But it’s Not Just More Hard Work. It’s Also More Payout.

Yeah, learning three coding languages on top of the myriad other skills and tools that digital marketers need to be experts in is a real pain. But keep in mind the reason it sounds like such a pain: programming is a hard, technical skill that only select, trained people can do. That’s why programmers make sky-high salaries. And that’s why adding it to your repertoire will increase your worth exponentially over your career.

That’s just the more obvious, concrete reason to professionally adapt to this shift in digital marketing. Another reason is that knowing basic coding will make the rest of your work more precise, easier, and if you’re passionate about your work, more fulfilling. Not only is SEO not dead or dying, it’s about to become an even more specialized and in-demand skill than ever before.

Forget Meta Data. Let’s Take a Look at Micro Data.

Google is veritably a massive database. The entire web is scraped, poked and prodded to generate as much information on content and relevancy as possible. With the advent of semantic mark-up, you not only have the ability to give Google more information about what your content is, but you can expand your pages real estate on search engine result pages.

Marking-up content, or making “rich snippets,” is what enables search engines to show reviews, ratings, authors, additional links, and more within an organic search result. Like the more traditional meta data tags (title, description, etc.), rich snippet mark-up is added right into the <head> of your page’s HTML. Builtvisible has a great, in-depth blog post on the different ways to code and utilize rich snippet mark-up languages.

Although this is yet another area where Google will not equivocally state whether or not it contributes to search page rank, doing your own independent research will give you a pretty good idea. Go ahead and Google just about any query, and at the very least the top several organic results will display additional links underneath the meta descriptions. For larger-market, more competitive search queries, you’ll see blog author names, star ratings with numbers of reviews posted, and (mainly for branded search queries) additional site links with accompanying description text snippets.

You will often see these additional site links for your website even if you haven’t coded any mark-up snippets. That’s because Google is pushing for SERPs made of more content from fewer, higher-relevance results. Manually adding mark-up snippets is also a great way to control what Google displays, if you’re seeing additional data or site links that you don’t want shown. You can also deprecate certain links in Google Search Console.

For a little bit of effort and exploration, you can turn your site’s organic search results presentation into billboards with more page real estate and information than even paid advertising.

Research – and the Lack of it – Around Digital Marketing Psychology

As psychology became an accepted member of the scientific community, it was only natural that its applications were applied to a broad range of fields – including advertising and marketing. These professions are, after all, exercises in understanding and influencing human behavior.

Enter digital marketing, a specialized modern offshoot of a relatively modern profession. Digital marketers are collecting massive amounts of big data every hour of every day – and it’s all insight into human behavior. In the early days of consumer psychology, data was collected through surveys, focus groups, and media monitoring. Not exactly unbiased, uninfluenced contexts to try and make determinations about natural human behavior. But now that we have all of this pure, digital marketing data, we’ll have tons of research being published yearly about consumer psychology in the current digital environment.

…Right?

A published paper by Andrew T. Stephen, who’s a L’Oreal Professor of Marketing at Oxford, identifies the five main fields of research into digital marketing psychology, and critiques the most notable contributions in each. These areas are:

·      Consumer digital culture

·      Responses to digital advertising

·      Effects of digital environments on consumer behavior

·      Mobile environments

·      Online word of mouth

It’s a fascinating introduction for someone who wants to become acquainted with this branch of research; it’s also disheartening for anyone who is looking for more advanced research to delve into. Mobile and digital advertising are important aspects of digital marketing, but beyond these two examples there is no substantive research on specific channels and strategies.

In his conclusion, Stephen points out how the research in this field is failing to keep up with current trends. For example, he could find no research at all on consumer behavior around privacy implications. This would be an area that’s easily tested, and could have immense implications for webmasters and digital marketers.

In one of my blog posts, I discuss the ways in which digital marketers can tap into other fields of study to add new data to their toolbox. But this was anecdotal personal observation, not rigorous scientific evaluation. In a growing industry with more money for R&D than most academics, and more data available than just about any other professional field, it’s sad that research has fallen so far behind.

Maybe larger companies that invest in research are keeping it sealed and outside of the public academic arena. That would be proprietary and understandable. But in an age where there are so many thought leaders and specialized authorities, as a community a larger emphasis needs to be placed on producing scientifically-sound data on this already-ubiquitous field.

What Digital Marketers can Learn from Behavioral Economic Research in the Developing World

Digital marketing analytics and data collection present a great opportunity for modern marketers. We can answer just about every question about human behavior in our online environment – where you acquired them from, how they use your site, when they’ve visited, etcetera. But even with all of this data, determining why someone did something is still an educated assumption at best.

Luckily there’s a wealth of research and data at marketers’ disposal that they can employ to complement their digital data, and create a more complete customer persona. Behavioral economics, in particular, offers a unique resource for marketers who want to take their targeting strategies to the next level .

Behavioral Economics

Behavioral economics is a fascinating field because it focuses on two aspects: What the sensible, prudent choice would be in a financial situation; and what people actually do instead. A nonprofit organization pioneering research in this field is Innovations for PovertyAction (IPA). Hearing that example, you’re probably thinking “What on earth does research on financial behavior in the developing world have to do with digital marketing in a first world country?”

The answer is that people are equally as bad with money and irrational about their spending habits in the first world as the developing world. And on the other side of the coin, sometimes the best choice for an individual is not always what a company might imagine.

Translating Third-World Crop Insurance into Conversions

Take this study for example, which offered farmers in Kenya three options for buying fertilizer: A standard offer; a discounted offer where the price was reduced 30%; and a “harvest deduction offer,” where farmers would pay full-price, but a significant portion of was paid after harvest. 72% of respondents chose the third, Harvest Deduction offer.

Now here’s the dreaded question: why? The researchers posited two ideas.

1.     The farmers have a present bias, meaning they give more weight to costs closer to the present time, and don’t feel the loss as much when considering a further future payment;

2.     Farmers did not have enough lump sum capital to make the full purchase at the outset without selling crops, even with a 30% discount.

So how does selling agricultural insurance in Kenya relate to selling products online? Take, for example, a segment of customers who have been buying from your site for a long period of time, but their purchases are less frequent compared to other customers, and they also spend less money per order. This segment also tends to make wish lists if your site has that feature, or they put items in their cart and leave them there for long periods of time without ever even beginning the checkout funnel. Just like the farmers, they know they want to buy from you, but they also know when it is prudent for them to actually finalize the purchase.

This is valuable information. Now you can send them discounts via email to speed up and encourage the purchase process; you can launch a CPI remarketing ad campaign targeting this segment to keep ad costs low and conversions high; and if you notice a segment like this comprising a large swatch of your customer base, you may want to consider introducing payment plan options like the researchers did in Kenya. This would make the customer buy earlier, possibly buy more frequently, and greatly increase their lifetime value.

Lessons in Direct Marketing from Ghana

Another IPA study in Ghana tested whether text message reminders straight to peoples’ cell phones increased their rate of completion for malaria treatment. The results are significant, but not staggering. There was a 5% increase in completion in the treatment group. Two different messages were used – a short one, and longer one with an additional warning – and there was no statistically significant difference between the two.

However, the short text message reminder more than doubled the completion rates amongst women, yet had no significant effect on completion rates amongst men. It’s also worth noting that participants who attended private clinics had a 14% increase in completion compared to participants who attended public hospitals and the control group.

Again, how does this relate to digital marketing? Simply put, it reinforces that profiling is not only ok in the marketing world, but it has more value than marketers may even give it credit for. Granted there is more research needed and every industry is different, but taking medication for a sickness you want to get rid of is a pretty convincing control.

According to this research, a direct marketing message reminding someone of something they want to do, that also provides a direct reminder of next steps to take, can be very effective with women, but not so much with men. Also, the private clinic participants could be considered your segment of reliable customers – they’re looking for you to reach out to them, and they’re more ready than their counterparts to take action.

In short, to get ahead and stand out in the marketing world, you can’t limit yourself to marketing data. Yes, digital data is nearly infallible when collected correctly. And no, spending your free time reading up on behavioral economic theory may not be your ideal Friday night. But if you want to get closer to answering “why?” you’re going to have to look further than Google Analytics.