A while back, in one of our blog posts, we wrote words to the effect of "marketing without data is like playing darts blindfolded". Not the most glamorous of similes, admittedly. But it's true - you may know that you're hitting the board, somewhere, from the sound of the darts, but beyond that, you've no idea what you're hitting.
Before you can market effectively, you've got to measure. So. If you're starting from scratch, or re-starting a campaign, or just have this nebulous feeling of "we need to do better", here's a straightforward guide on how to go about things.
1 - get a handle on where you are now
Whether you're marketing or revamping your website, doing so without knowing how you're performing at the moment is, frankly, pointless, as you won't have a clue what impact any changes, or any marketing, that you make will have.
So the starting point is to measure. In fact - for many companies - the starting point is to work out if you're measuring or not. You would be surprised at how many small (and even medium) sized businesses we work with who don't know if they've got Google Analytics installed on their site, and if they have, don't know who controls the account.
We covered this in a recent blog post - how to check if you've got Google Analytics in place, how to get access to it, and so on, and so rather than rehash it, we suggest that you have a quick read through that.
2 - metrics relevant to all sites
Some metrics that you want to measure will be consistent irrespective of what your site's about. Overall visitors, where they're coming from, and whether they disappear off from your site as soon as they arrive (the dreaded "bounce rate"). For those that stay, you probably want to be checking how long they stay on the site, and the average number of pages that the view.
If you target a particular geographic market, then you probably want to pull out the same data, but only for users within that geographic market. Whilst that is possible to do in Google Analytics, as we'll see a bit later, we're not going to be using Analytics for viewing our data, so we can bring all the information we need from diverse sources into one place.
Of those metrics, you want some (visitor numbers, how long they stay, average number of pages) to go up, and some (bounce rate, for example) to go down.
3 - metrics relevant to your site
This will depend on what your site's key focus is. But broadly speaking, if your site's got any purpose at all, that purpose should be converting visitors into...something.
If you're selling online there and then, it should be converting those visitors into customers - getting them to the products, to add the products, to get through the checkout and to buy. That's not to say, though, that e-commerce sites should only focus on immediate sales - it can often be beneficial for them to track other prospect engagement as well.
To do that, and for sites which aren't selling online - so the site is only there to drive prospects to get in touch with you - then the site should be converting those visitors into prospects. That conversion may take a variety of different forms - clicking on the phone number on your site to call your sales team, submitting a contact form, susbcribing to your newsletter, downloading your brochure, or whatever - but all of those forms should be (a) measured, and (b) valued.
The value is the hard bit - where selling online, the value of a customer is clearcut - it's the amount that they spend (mercenary, we know, but...). Where you're not, then the value can really only be ascertained in more general terms. For example (using nice and easy numbers), if every 10 phone calls you get to your order line results in a sale, and the average value of a sale is £50, then each phone call is, in effect, worth £5, and can be valued accordingly.
If your typical newsletter send to 1000 people results in £2000 worth of sales, then a signup to your newsleter is, in effect, worth £2 per newsletter sent out. And so on.
Only you can really decide - whilst the more accurate you can be, the better, to a degree it doesn't matter hugely, as what you'll be looking at is relative improvement, rather than absolute numbers.
4 - changes you'll need to make elsewhere
For a consistent approach, you will want to ensure that any visitors that you're driving to your website from other places are - to the extent that you are able - tracked as well so that you can report on them. Whilst there are many, many, places, the two most obvious are your newsletter, and your emails (your company email footers do have links to your site, no?). If you use a platform such as Mailchimp for your newsletters then ensure that you've linked it with Google Analytics so that all incoming visitors via links in your newsletters are tagged as coming (a) via email, (b) via Mailchimp and (c) from the relevant campaign - that's essential, particularly given that relying on Mailchimp for e-commerce analytics is patchy, at best. Any links in your email footer need to be tagged with UTM tags (which we'll cover in another post).
5 - measuring performance
Once you've worked out what metrics you need to be tracking, the next step is to make sure that you're measuring all of them. As mentioned above, some will be measured by Google Analytics out of the box. And some engagements with your site - clicking on links, for example - will automatically be measured. But you will need to be able to distinguish between them.
The actual "how" will vary from site to site, but broadly speaking, the easiest implementation is to ensure that Tag Manager is installed on the site. That, in turn, loads Google Analytics. But you then also have various other tags created which fire custom events into Google Analytics when certain things happen. Whilst someone starting to fill in any form will fire a standard event in Google Analytics, you may want to fire a custom event when someone starts to fill in your contact form.
Likewise, submitting a form will fire a standard event, but you will want to fire a custom event when someone submits your contact form. With those two bits of information, you can then see how many people start, and finish, filling in your contact form - a metric that you can then easily measure and track improvement on. 100 people starting but only 10 people completing it? Perhaps it's too complicated - simplify it, and see what those percentages turn out to be the following month.
For e-commerce sites you will want to ensure that you are sending e-commerce data through Tag Manager to Google Analytics, and ideally also tracking your sales funnel - from someone coming into the site, viewing a product, adding a product to their basket, starting the checkout, and completing the checkout. With that information, you will be able to see how the various stages of the purchase chain are performing.
6 - other sources of information
You should be set up with Google's Search Console - if you're not, then have a read of our recent introduction to it and make sure that you're set up with it.
Whilst a lot what Search Console provides is focused on making sure there are no issues with how Google sees your site, the key section from a marketing perspective is the performance - how many impressions and clicks your pages receive in a certain time frame (and therefore your click-through-ratio), as well as your average position in the results.
7 - reporting on performance
In our post last week we covered why we're hugely (if belatedly) in love with Google's Looker Studio. In our view, there is no better place for this information to be collated, so have a read of that post to see why we think that.
8 - making changes
Once you've got that far, stop. We would suggest that you have at least one clear calendar month (more for sake of ease than anything else) of all your measurements flowing in before changing anything. That gives you a baseline against which to measure future months.
Then, it's a question of what you want to improve. The possibilities are endless - at least, too wide to cover here. But you need to know what you're trying to achieve. That - in turn - will focus your work onto what needs changing. If your goal is improving the bottom line, and you're an e-commerce store, then you'll want to be looking at (a) getting more people through the door, and (b) converting more of them into visitors. That, in turn, will guide you towards improving your search engine rankings and optimising your website for conversions.
If, however, your goal is to get more subscribers to your newsletter, then that will focus your efforts in a different direction.
The key thing, though, is to know what you want to improve. Work out how to improve it. Make changes, measure the results - which you can now do - and review the changes. Not everything will work. Not everything will work immediately. But when you're set up to measure, at least you'll know, one way or the other.
To see how a structured measurement and content strategy played out over a full year in practice, read our case study Looking Back on a One Year SEO Campaign for Vital Direct.