Author: Filip Matous
Does this sound familiar? When your team reviews web analytics in your monthly marketing meeting, the person in charge of analytics reports says, “We got 123,456 visitors to our website this month.” Perhaps someone makes an effort to tie analytic metrics to business targets, but in no real depth, and the next item on the meeting agenda swiftly takes over without any action items being set as a result of the review.
The problem with this all-too-common approach is there is so much more to your website data than counting visitors — and there is more to your web analytics than reporting a number in a vacuum and moving on. By only giving one metric a cursory look, you are obscuring signals that point to both risks and opportunities in your business.
Instead you need to connect to your data at a deeper level by avoiding mental shortcuts, focusing on the metrics that really matter to your business goals, and communicating those metrics in useful ways. By doing so you can make better decisions about how to leverage your limited marketing resources.
Avoid Mental Shortcuts
When performing web data analysis, you are often looking to understand what is behind a particular trend — why is your site suddenly experiencing higher traffic levels? And why has your sales team just seen an influx of new leads — are those things related? In searching for an answer it’s all too easy to become guilty of what psychologist Daniel Kahneman calls System One thinking: using fast, instinctive, and emotional shortcuts to jump to conclusions.
With analytics, there are two culprits to avoid in particular: using aggregate data, and mistaking correlation with causation. When web data is clumped together — as when you’re looking at all traffic, rather than segmenting by source, referral, or user demographic, for example — there are often too many variables at work to understand what is really causing the pattern you are seeing. For analytics insights to produce actionable solutions, you need to favor System Two thinking: slow, logical examination of separate facts to find out what’s behind them.
Let’s look at an example. Say that you and your team have observed a boost in traffic to your luxury real estate website that correlates with a boost in leads.
Instead of just celebrating, or making an investment across the board to bring in even more traffic, think about how you can dis-aggregate your data so you can better understand cause and effect in this situation. To efficiently increase what’s really important — the leads — you need to know: which traffic sources are yielding most of the leads?
You have the team look at the numbers more closely, separating out the traffic data by each source in your marketing mix: email marketing, Facebook, public relations, AdWords, and SEO. By matching each source to a lead, and tracing it all the way to purchase, the team is able to identify the top performers: AdWords and Facebook.
As a result of this System Two thinking, your team moves more resources over to Facebook advertising, which multiplies leads because of the improved allocation of your marketing investment. Looking at your numbers carefully and digging into disaggregated data allows you to take much more educated and focused next steps.
But to do this kind of deliberate analysis successfully, you need to be tracking the right metrics and communicating them clearly to begin with.
Match Your Business Goals to Metrics
Making tough decisions is easier and more effective if you have a good grasp of the core metrics that matter most to your business. There shouldn’t be more than a handful of these if you want them to be potent — but you need more than one — pageviews isn’t enough. These metrics should be measurable month over month, year over year.
You can break these up into macro and micro metrics. Think of macro as the main objective and micro as an assist. Especially if you are selling products or services that cost six or more figures, plenty of micro actions will often occur on the way to a macro result.
Both macro and micro metrics can target different needs, common leads, sales, and brand building. Here is a list of sample metrics at both levels that you might explore:
- Your contact page is used to send an email
- Your listed phone number is called
- A specific company you’ve targeted visits your site
- Average customer acquisition cost (there might be tiers)
- Average order value
- Amount of new customers
- Branded site visits
- Newsletter subscriptions
- Whitepapers / Lead magnet / Case study downloads
- Amount of company videos watched
- Tripwire / loss leader offer purchased
- Webinar sign up
- > 3 pages visited + > 3 minute sessions
- Referral traffic from PR or content marketing
By regularly reviewing more than simply pageviews, you can find causation and uncover the marketing channels that consistently deliver.
Communicate Metrics Effectively
Having the right metrics is a good first step; then you need to communicate them clearly. Good metrics are usually most effective when represented as ratios because they allow you to compare data — you’re not just looking at isolated numbers.
For example: this year one of our clients wanted to increase their online sales numbers. We needed to decide whether to spend the marketing budget on increasing the amount of traffic coming to the site or on improving the site itself.
We first looked at sales figures and traffic month over month, but this proved confusing. Different months had different levels of sales and of traffic, so it was hard to spot any patterns. But when we combined the data as a ratio we found that that ratio was consistent month over month: for every 100 human website visitors, the client acquired one customer (worth a few thousand dollars).
We felt that this ratio wasn’t good enough for that market. It had room for improvement so the whole budget was allocated to a new website redesign, which, upon completion improved the ratio to one in 60. Traffic levels stayed the same but sales increased.
Seeing how the effort on redesign impacted the sales conversion rate motivated both the customer service team and business owner to work on other areas that improve this key metric.
Once you know which metrics you’ll be tracking and how they will be represented, schedule consistent reporting. What I’ve found works well is creating a two-page report that takes 15 minutes to review at a meeting and that can be used month after month. Keeping the report format consistent fights the natural urge to explore different analytic findings each time. Of course if you are in campaign mode, there will be special campaign reviews, but your ongoing brand marketing needs stability and understandable metrics to review each month.
As you start digging into analytics at this deeper level, egos may take a hit, especially if team members have made assumptions that differ from the outcome of your analysis. If, for example, it turns out that the traffic that helps you reach most of your goals is actually coming from one particular channel, the individuals in charge of the others might not like the news. This is where your emotional intelligence as a leader comes into play; employees need to feel safe to learn from and improve the metrics rather than feeling threatened that everything didn’t go well.
Recall the last time you discussed your website analytics. Did it result in clear actions? Remind your team at your next marketing meeting: the whole point of reviewing data is to act on the real information you have — and once in a while, to celebrate success.