Monitor and manage: how website analytics can help increase conversions

‘In 1993, when I came back to Coca-Cola and developed the first round of television commercials, the first thing I did was show the footage to Roberto Goizueta (Coca-Cola's chairman and president) in his office. - ‘I don't like these commercials,’ he said. - ‘Listen, Roberto,’ I replied, ‘if you're going to sell all your products all over the world, then I'll make the adverts you like. Otherwise, I intend to do what consumers like. When you realise that marketing is about selling products, then the money you spend on it becomes an investment, not an expense.’

This is an excerpt from a book by Sergio Zayman, who for many years was the chief marketing officer of such companies as Coca-Cola, PepsiCo, Procter and Gamble and Microsoft.

Even back in the 1990s, the success of such companies was largely determined by the attitude of employees who understood that analysing consumer interests and behaviour was not only the basis for an advertising campaign, but also an opportunity to scale their business. The communications market has changed a lot since then. And if earlier it was enough to create a bright advertising message ‘for everyone’ and send it to the most popular TV programmes, and then analyse the changes in sales and awareness, now everything is a little different.

Digital has changed the way we think about analysing advertising campaigns. Just 10 years ago, it was enough to put a Hotlog counter or a popular Ukrainian startup, Bigmir, on a website that could hardly be called e-commerce, but which already sold, for example, enamel pots with a non-stick coating, and then visit it every day, enjoying the increasing number of hits and hosts.

So now, the most popular, but by no means the only, Google Analytics tool in its basic configuration is an empty space!

At the same time, I still very often come across projects whose task is not just to beautifully print their website address on a business card made of designer cardboard, but to sell it on the Internet, using Google Analytics in a ‘black box’ mode.

What is it? The black box mode does an excellent job of analysing data:

How many users came to the site and how many left;

How many from mobile devices and how many from desktop computers;

From which channels, with what bounce rate, and how many new ones;

And a few more data of this level. Great, you might say, a huge amount of information, and you'd have time to analyse it.

I disagree and will try to explain why. First, the growth in the number of Internet users is slowing down. And while it has grown by 10% globally over the past year, 4% in the US, and 2% in Ukraine, according to research by the Ukrainian Internet Association, the number of Internet users in Ukraine has not been growing since May 2017 (currently about 23 million users). At the same time, the number of e-commerce projects and corporate businesses shifting their focus to finding customers on the Internet is growing. The average growth is about 20% per year. This means that competition is growing.

Secondly, the dynamics of the modern world demonstrate another important contradiction for digital business. The average daily time spent on the Internet is growing - over the past 5 years, it has doubled, and now the average Ukrainian user spends at least 2 hours on the Internet. At the same time, the time spent by a user on a particular website is decreasing.

Over the past year, the average time spent on the site has decreased by 31% and is now 2:09, compared to 3:03 a year ago.

Wow, that's a lot of statistics! And what's next! And then there's the analysis!

1. Decompose and turn your site inside out, learn everything about your visitors, understand the relationship between the desktop version and the mobile version, determine the most popular time of activity of your audience and use it in advertising campaigns, analyse in detail the weaknesses of your site - where customers fall off, which links/buttons are not clicked at all. We had a project in which a huge search box occupied the central part of the site and, according to the manager, was supposed to facilitate specific sales. After installing in-depth analytics, it turned out that this ‘gorgeous’ window accounted for only 0.006% of all clicks on the main page in a month. As a result, the window was removed, freeing up space and raising the promotional offers higher, which increased their conversion rate by 18%.

2. Calculate ROI - it is possible and necessary for digital projects. And to make it obvious, set up not just goals in your analytics, but calculate the value of each of them. You may not know it, but Google Analytics allows you to take into account even discounts and sales of your product in the analysis. Keep in mind that nowadays it is not enough to just know the purchase channel. Build omnichannel sequences and identify associated conversions. Nowadays, everything is not so black and white, and to make a purchase, a user will find you in many channels. And your task is to identify them all, understanding the contribution of each.

3. Keep in mind that quantity is not everything. We took on a project that showed excellent growth in traffic, and organic traffic, and a decrease in conversions. In the first general analytics shown to the director, everything looked very nice (as a manager myself, I like it when the graph is constantly growing). When the promotion strategy was changed six months later, traffic slowed down, but conversions increased by 43%. And it was all about matching the interests of users with the entry points to the site.

4. Experiment. After analysing the current state of affairs, drawing the right conclusions, and implementing changes (and you will definitely implement them - the data will be so obvious), proceed to tests. Namely, A/B testing. Here are two cases to inspire you. Consulting services. Changing the name on the button after 2 months of A/B testing showed an 11% increase in conversions from the button with the new name.

E-commerce. Changing the location of the ‘checkout’ button in the shopping cart based on testing increased the number of checkouts by 8.7%. And you must admit that with unchanged investments in advertising, this is an increase in profits out of nowhere.

5. Think about the future. And it is in personalisation, automation of work with huge amounts of data and samples, and, oddly enough, in visualisation of what is happening. No matter how deeply we dive, our task is to show our clients, business leaders, who should not be as versed in these rather specific systems, the main data in a clear and compact way. Therefore, develop reporting systems with indicators that are ultimately important to your client, not your PPC specialist.

Beautiful creative campaigns and online marketing as a set of digital tools are no longer in existence. Marketers who effectively use individual tools or channels without building an analytics system architecture and analysing the entire project comprehensively lose in the competitive battle for the consumer to logic, consistency, and a focus not just on analysis, but on predicting the future of consumer behaviour.

Irina Mushtina

CEO of CF.Digital

The material was published on Delo.ua