5 Critical Digital Marketing Metrics You’re Probably Still Missing
24th of Nov 2018
Connecting what consumers do online to the black box of what they do in the offline world is a mystery digital marketers have been dying to solve for millennia. Or it would have been if digital marketing had been around that long!
Joking aside, Forrester research revealed that understanding customer interactions across all touchpoints is the #1 challenge for marketers. No wonder, it’s a data jungle out there.
And often the biggest disconnect is the chasm between the online and offline world. While humans may be just “one self” in the real world, in the digital world each human is represented by both software and hardware: a browser that is connected to them (which they may or may not be logged into or allow to record their browsing history), and a devices that are connected to them.
But here’s the problem – what happens when other people use your device?
Our team has often laughed about the “Peppa Pig Effect”. This is what we call the effect of trying to target women in a certain age bracket (say between 25 and 40) on any subject. If you’re not careful with targeting exclusions, you’ll come back to a campaign a day later to find you’re appearing on a variety of kids YouTube channels from Peppa Pig to The Wiggles. I’ll give you a hint, it’s not the 35 year old woman who is watching the video…
Or what about when you’re shopping for Christmas presents on your work laptop connected to the office wifi, then head into Myer later in the day with your mobile phone to make your purchase. Without constantly being in a logged in environment, mapping the virtual you to the real you is harder than you’d think. To put that in numbers, a recent study from Conversant, Forrester IPSOS and Comscore found 67 percent of shoppers started on one device before moving to another to make a purchase.
Yes, yes, I know “microchips are the answer” is what you’re thinking. And that future is certainly closer than you might think. But we’re not there yet. Thankfully though, there are some simple ways to approach marketing analytics that make marketing insights easier, more accurate and more effective.
Decide what data you ACTUALLY need
I liken data to looking at a menu when you’re hungry. You haven’t committed to what you’re going to choose yet but that’s ok because the entire menu looks good and you’re pretty sure you could eat it all.
But just because it CAN be tracked, doesn’t mean it SHOULD be tracked
To start, these questions can be a really helpful framework to curb the hunger for unnecessary data:
- What decisions am I trying to make?
- What hypotheses am I looking to prove or disprove?
- How do I need to validate the information or conclusion I come to?
- When I make decisions based on data, how will they be actioned?
When you go through the following list, think about your own website. What are you tracking at the moment, do you know? How far back or forward does that tracking follow your customer on their journey? Usually 5 or 6 of the metrics outlined below are not being tracked. Not all metrics are relevant to all organisational types but this should give you good visibility on some potential gaps that may exist for you.
- Track visitors, and their movements
Most of this gets tracked “out of the box” so to speak. When you place the tracking code on your website, it will automatically track stuff like visitors to your website and where they came from. The originating channel is still pretty broad though and it’s all set around defaults. So for instance clicks on your GMB (Google My Business profile) are automatically lumped in with Organic Search traffic. But if you’re a brand with hundreds of local stores, you’re going to want to break that out.
- Track specific goals completed
These are actions that need to be specifically defined by you based on how important you think they are in telling a story about the customer’s behaviour. It could be anything from how far they scrolled down the page to read your article, or signing up to a newsletter or making a purchase. Advanced ecommerce tracking is another optional extra form of tracking that ecommerce sites should all have in place if using Google Analytics and Google Tag Manager.
- Track movement from a website lead form submission to a sale in your CRM
Doing this in Google Analytics requires the use of Google’s Measurement Protocol. At a simplistic level what happens here is when a site visitor submits a form on your website, a hidden field with a unique identifier for that visitor is sent with it and stored in your CRM along with the opportunity or lead record. When a sales person then marks that opportunity as closed in your CRM, that sends that unique identifier, along with any other data that you wish (not including anything that could identify WHO the customer is), back into Google Analytics.
This means that in GA or from any dashboard connected to GA data, you could for instance, see exactly which channels had the highest sales conversion rate, regardless of the initial visitor to lead conversion rate.
Now not all forms on websites automatically connect to a CRM and that’s where the complexity sets in. This must be the case in order for you to connect the dots. So if your website form/s output only send an email and don’t automatically connect and send the data to your CRM, that’s an operational shift as well as a technical integration that would need to occur before you can achieve this tracking.
- Track movement from a Live Chat instance to a sale in your CRM
Zendesk, Intercom, Drift. These and many other platforms are now prolific on most corporate websites. And many of them also have sales lead tagging capabilities – in Drift’s case it offers further sales and marketing automation capability. If you’re lucky enough, someone might have even connected it back into your core CRM. Nevertheless, this is likely a separate system again where the Measurement Protocol would need to be sent.
- Track movement from a phone call via your website through to a purchase
Tracking a phone call requires some extra setup again! Yes, Adwords gives you call forwarding numbers and you can now import into GA what Google refers to as “conversion” data from phone calls with Google Forwarding – but in the case of a non ecommerce site, that conversion is only to a lead, it never tells you if an actual sale was made.
Himanshu Sharma does a great job of explaining the basic process you need to go through to get proper commercial call tracking set up and then connect that to your CRM, and then of course using Measurement Protocol to connect that call sale data back into GA.
- Track movement from an email subscription to a sale in your CRM
Email is generally the generator of the most sales. The simplistic thought process that follows is that we invest more in email. But email is a channel made up of people who are at the very bottom of the sales funnel so of course they’ll convert at a higher rate. The real question you want to know is where did they originally flow from? What was the original traffic source that fed your email list that ultimately got you the sale?
Very often email marketing systems are not on the same platform as your CRM. Sometimes they are, like Salesforce and Salesforce Marketing Cloud. But if you’re using Mailchimp or Campaign Monitor, these may or may not have out of the box integrations with your chosen CRM.
Email or marketing automation systems are often used as middle integration points. That is, rather than sending information directly to the CRM, it’s sent to the marketing automation platform first and then to the CRM. The CRM then sends data back to the marketing automation platform which in turn can send it to GA. This type of set up is worth considering also and can sometimes be cheaper, especially if there is an out-of-the-box integration between your website CMS and your marketing automation platform, but not your CMS and your CRM.
- Track movement from website to in store foot traffic
This is probably the toughest one to track.
Companies like Blix and Kepler Analytics provide retail foot traffic insights, but they don’t connect back to web traffic.
Google does have a solution called Store Visit Conversions, but a) its only tracking through from paid search activity and b) they’re also pretty exclusive with it and companies need multiple store locations, thousands of ad clicks, viewable impressions and enough store visit data to attribute back. Let me burst your bubble now before you get too excited. Most companies don’t meet their criteria. Bleh.
- Know at what point they hit different parts of the sales funnel in your CRM, or how much they purchased when the sale closes
This is all about what anonymous data you choose to send back to GA from your CRM. You don’t want to get too fancy with it too soon but things like lead status and the sale value are both very useful data points to be able to tie back to site traffic data.
Once this tracking is set up, you then have a pretty good view of which digital channels contribute the most revenue, exactly which search or paid ads were responsible for a sale, and if you have set up campaign tagging for email or other social and campaigns, you’ll also be able to see exactly which one was the catalyst for sales.
It’s important to note that none of the above refers to analytics systems like Mixpanel which is designed for applications or websites where a user has already identified themselves by logging in. GA tracking NEVER identifies the individual user that was sold to.
Decide what data you ACTUALLY need
When you’re determining your metrics, think long and hard about the levers that you’ll actually be able to use to take action ahead of time, rather than when it’s too late. The best frame for thinking about this is lead and lag indicators. A lag indicator is data that tells you about the past. It’s often the easiest to measure, but it’s already happened so there’s not much you can do it about it! In the context of marketing analytics as it relates to a website as an example, organic traffic is a lag indicator. Once organic traffic goes down, it’s kind of too late, you need to be able to see that it is likely to go down months in advance so you can take action to rectify it.
“But how would you do that?”, I hear you ask. Why with a lead indicator of course! A lead indicator is data that is easier to influence, but can often be harder to measure.
If we stick to our lag indicator example of organic search, the obvious question is, what lead indicator could help me figure out if organic search is going to go up or down? There are a quite a few, but our favourite at Foresight Digital is Domain Authority. Domain Authority or DA, is a score, usually out of 100, that third party SEO tools determine using their deep knowledge of Google’s algorithm. It represents how “authoritative” or “credible” your website is in the eyes of Almighty Google.
These tools sometimes call it different things. For instance AHREFs and MOZ call it Domain Authority, while Majestic calls it Trust Flow and Citation Flow. But they all help you to measure the likelihood of your website ranking for relevant keywords. If your domain authority is dropping this could mean you’re losing either the number or the quality of links back to your website, and as that drops, your rankings drop, and as your rankings drop, you’re seen less often when people search. And as you’re seen less often, less people click. And as less people click, your site gets less traffic. And voila, that’s why DA is such a great lead metric!
Decide on how you need the data to LOOK
Insights Experience or IX is something I’d love to see more attention. Given this is data you could be making critical business decisions with, it’s pretty important that you can easily read and understand it. We’re big advocates for applying the same approach to a marketing dashboard design as you might take doing UX and design for a sales landing page.
*An illustrated example of one of our dashboards
Why is it necessary? It’s a sad fact that many marketing dashboards are a combination of the faux pax’s below. The rule is pretty simple: if the graph or table or data doesn’t give you information you are going to act on, don’t put it on your dashboard.
So often we say “data is a priority” and we talk about wanting to connect the dots in terms of sales and be able to track it back to online activity. While tracking is never 100% accurate, it’s a whole lot better than nothing at all. So make these more advanced, but highly valuable data points a top priority.
In Part 2 of this mini series next month, I’ll cover off things like getting your hands on the data, getting your head around how web behaviour is actually tracked, deciding on an analytics tech stack and more.
Check out more articles about: