What is your idea of good web analytics? What should it be like? Where should beginners start? Please give some tips.

A good web analyst should first and foremost ensure competent data collection.

Guaranteed data delivery can be one of his KPIs. And this should include setting up goals, conversions for the site, not just standard interactions like page views.

__________

If we’re talking about Google Analytics — it’s a minimum of setting up filters to exclude unnecessary data, linking to advertising systems, checking the quality of the goals set up.

The second is understanding the KPIs of key performance indicators for the main types of business. At first glance it may seem that web analytics needs to track only 3 – 4 key universal metrics such as CR (conversion rate), CTR, CPL (cost per lead). This is partly true, but if we start digging deeper, then, for example, behavioural metrics are very important for a blog, and for e-commerce sphere — metrics on sales of specific products, their brands, etc.

And the third is the ability and willingness to extract useful information from the data. An experienced analyst likes to dig into the numbers, dig out information, and look for patterns. It’s like solving a problem without clearly defined conditions and a vague answer that may not exist. But if you find a pattern, you get a reward in the form of the effectiveness of this or that audience, a competently set up sales funnel, etc.

You also need to be able to react quickly, to create so-called ad-hoc reports — reports on demand. In my lectures I talk about 2 types of metrics: Exploratory metrics and reporting metrics.

Reporting metrics we track on a regular basis — these are monthly or daily standard reports.

Exploratory metrics are about looking for patterns in our volume of data. We look for these metrics to better understand the audience in order to further adjust the settings of targeted advertising. Or to conduct a/b tests, experiments to further improve key business metrics.

Often standard reports in the same Google Analytics are not enough, and you have to upload data in tabular form and generate reports that answer a specific question. Therefore, knowledge of Excel or Google Spreadsheets is very useful.

For beginners, I would advise you to create your own website on the same free WordPress, fill it with whatever content you want, put in a Google Analytics counter and start collecting data. When you work your way from scratch, setting up goals, filters and bundles – it will be very valuable.

There are also great official courses from Google at Google Ads Academy, and they are completely free. I’ve taken them myself in the beginning stages. Moreover, you can get a certificate if you successfully complete the course and pass the tests, which will greatly increase your value in the labour market. The only downside is that the videos are in English, but with Russian clear subtitles. Tests can also be taken in Russian.

If you work in the e-commerce sphere (online sales of goods ) you will be helped by a free demo account of Google Analytics — Google merchandise store. This is an open analytics of google shops for selling merchandise with Google logo. As you realise the analytics is set up perfectly there, you can navigate as an example.

__________

Exploratory metrics is about looking for patterns in our data volume. We look for these metrics to better understand the audience to further adjust the settings of targeted advertising. Or to conduct a/b tests, experiments to further improve key business metrics.

What is the best system to choose for analytics and how to do it right?

Must have for a beginner analyst is Google Analytics. This system is free and one of the best for web analytics.

Yandex Metrica is interesting with behavioural visual reports such as click, link and scrolling map. And it is also very convenient to track the effectiveness of forms on the site by default, without additional settings.

It is quite normal to have several analytics systems installed on a site, so you can safely put both systems.

In this article we are talking about web analytics only, so these systems will be quite enough to properly track all metrics.

Do all projects need web analytics? Is it possible to do without it?

Absolutely. We live in an era of high competition, so it is hard to build an effective business without a data-driven approach.

Data-driven is an approach based on data. I.e. for a company to be data-driven means to make decisions based on built analytics, not on emotions and guesses, even if they are suggested by the management.

Please share some of your successful case studies.

I can give you an example from my experience at Citrus. One of my first tasks was to set up compound goals – user movement along the purchase funnel, which included the following steps — adding a product to the basket, entering shipping and payment data, and completing the payment.

Having set up the goal, literally in 2-3 days I noticed that at one of the steps I was losing quite a large number of potential buyers – the data entry stage. In fact, there was a purely technical problem, and it was not obvious enough in normal testing, but on a large volume of data surfaced.

Having solved this problem, we managed to improve the conversion rate of this stage of the funnel by 35% and, consequently, the number of sales.

I.e. an analyst should be an indicator of any changes, jumps, drops in sales or traffic, and the essence can be of different nature, the main thing is to get to it.

How do you improve your knowledge?

Earlier it was email subscription to several useful resources, but now it is exclusively Telegram channels. Almost all the top resources have started their own Telegram channels.

https://t.me/webmagic — useful channel on contextual advertising and web analytic.

https://t.me/webanalyst — the channel is more focused on Google products – GA, Google Data Studio, Google Optimise.

https://t.me/neokotler — channel that aggregates the most interesting selections from all popular marketing channels.

What trends could you highlight now in this sphere?

I would highlight the emergence of the position of “Marketing Analyst”. What is the difference between a web analyst and a marketing analyst?

A web analyst focuses mostly on website performance metrics such as conversions, leads, etc.

__________

A marketing analyst digs much deeper. His tasks include building end-to-end analytics. Cross-cutting analytics is not only tracking actions on the site, but also linking these interactions with CRM system, business efficiency.

The simplest example is tracking the effectiveness of marketing channels of an online shop. With standard tracking we will see the number of requests, but not the number of actual sales. To understand the facts of sales we need to import this data from the CRM system to Google Analytics. So we can calculate CAC (cost of customer acquisition, not requests) and therefore the return on investment in a particular marketing channel.

It so happens that often the questions of web analytics are closed by the marketer in the company. With small budgets for advertising, the marketer can really close the main questions on analytics.

But more and more medium and large companies allocate a separate position for analytics, because a good marketer simply does not have time to simultaneously set up competent data collection, dig into the figures and at the same time generate ideas for advertising activities.