Lies, damned lies and SEO statistics – what should SEO be reporting on?
Google Analytics provides you with a wealth of information about the way your site is being used, but as the pandemic has shown us, statistics can be misleading. Data can be sliced and diced in such a number of ways it can become almost meaningless.
What’s more important, is the context and analysis that runs alongside this data. From market trends, to seasonality, to low and unrepresentative figures, the numbers don’t always tell the whole story, but digging a little deeper into the ‘why’ is what makes a more reliable SEO report. And most importantly, a more useful one.
After all, the value in reporting isn’t just to hold you accountable or to provide the stick to beat you with. Nor should it be produced as a box ticking exercise that you complete once per month and move on. As an industry we need to get better at using the findings from our reports to influence future decisions – not just from an SEO perspective but at a business level too.
Whether you’re reporting internally or directing your SEO agency on what to include in their monthly roundup, there’s an art to dissecting the right data and presenting it in a meaningful way.
More than just search positions
I’d like to think the days of SEO agencies simply firing off lists of ranking reports are long behind us, but there is still a tendency to focus too much on search positions. Yes, you want to get to page one, but why? To get more traffic, more click-through, and ultimately more business – so report on this too.
Arguably your most important stat from an SEO perspective, traffic should be at the top of your priorities for reporting. Set your parameters to year-on-year to get a more representative view that takes seasonality into account. Consider what might have been happening this time last year to explain any growth or decline, whether that be a shift in the market, or the fact you launched a new product. Look at overall traffic, look at organic traffic, look at how they compare. Examine traffic for key products and pages to see whether spikes and dips here might be manipulating overall trends. For example, we might see that traffic to the Browser Media site is up 10% and feel satisfied we’ve done a good job, but on closer inspection find out we’ve had a surge of interest in our PPC service, and a slight decline in all other areas; something we’d inevitably want to address in our activities for the next month.
You’re working hard to push pages up the listings, with a view to people clicking through to your website – so see if they are! Making your website more visible if only half the battle, so if your report can identify key pages with low click through rate, this might be an opportunity to examine why, and potentially tweak some page titles and meta descriptions, or experiment with Schema markup.
Session duration and pages per visit
Engagement stats are important because you need to be considering the intent behind the organic traffic you’re driving to the site, which in turn can tell you whether you’re focusing on the right keywords. The problem with some of these stats though is that they are often misinterpreted. There can be a tendency to panic if average session duration dips, but it’s not a simple case of the more time spent on one page the better. Instead, it’s important to take into account questions like how much content is there on the page? What stage of the funnel are they at? Is it a considered purchase? What are you asking them to do? A long dwell time can be great, but it might also be an indication that they aren’t converting, and similarly, a high number of pages per visit might suggest they can’t find what they’re looking for. It’s all about context; a Contact Us page with a short session duration is likely fine – chances are there’s just a few lines of text and some contact details so you wouldn’t expect people to hang around for too long. If in doubt, use segments to view these stats from organic traffic compared to overall traffic to see if they behave differently.
Again, it can be useful to see how your organic traffic performs here compared to overall traffic, whether you’re tracking actual sales, or other forms of conversions, from downloads, to contact forms, to newsletter sign ups.
Bounce rate is another slightly contentious stat that can be misconstrued. To use the Contact Us page as an example again, a high bounce rate isn’t always necessarily a bad thing, but of course a sudden hike in this figure needs exploring and could be a sign of usability issues. It can also be useful to view average bounce rates for your own industry.
SEO is about so much more than just search positions, but of course it all stems from here. The key is to focus on patterns and movements over time, which is why Google Search Console is a great place to turn for your SERPs reporting. Taking aggregated positions over time, it provides a more accurate view than a spot check ranking tool, and helps you draw some useful conclusions, as well as map patterns with activity. A list of every word or phrase you’ve ever appeared for from SEMrush on the other hand, is meaningless. The number of keywords you rank for might be the single most irrelevant SEO stat there is. A large portion of these terms just aren’t relevant, so where you sit for them means nothing. Even if it is a phrase you’re chasing, it also means nothing if you move from page 6 to page 4; it results in little to no real world action. Choose a tool like Visibility and track keywords you’ve chosen based on realistic potential, search volume and relevancy, or, better still, take a look manually in Google Search Console.
The ‘So what’ factor
An SEO report, like any marketing report, should be less about the numbers themselves and more about what they mean. Your commentary is your most valuable asset. Use your understanding of your business, your industry and your marketing insights to pick out anomalies, explain what you’ve learnt and what you plan to do about it. That’s what other departments in your business care about.