by Tomasz Lechowicz, business advisor, Gemius

It has long been known that observing the behaviour of clients who come to a shop or client service desk helps you understand their needs better, and through this, boost sales. E-commerce works by similar rules. However, the analysis involves digits rather than the flesh and bone customer. How to use data on client website behaviour to foster a mobile operator's e-business? Just take a closer look at so-called small conversions – hints Tomasz Lechowicz, Senior Business Consultant at Gemius, who advises telecom companies on how to use Big Data to the benefit of their business.

Telecoms' offer is for years

E-sales conversion in case of telecoms is by far more complicated than for any other online store, such as shoe shop. This is due to the specific characteristics of the services (e.g. a contract under which the client accepts a two-year subscription). Any e-sales figures analysis should then start with defining what conversion actually is.

Abandoned cart does not matter anymore

In the context of telecoms, conversion is a different matter because of characteristics that are typical for a mobile operator's client. First, by taking a decision to purchase, the client undertakes to pay specific amounts for a long period of time – usually over a year. Second, the purchase decision is connected with expiry of the previous contract with the operator. Consequently, the behaviour they display on the website is extremely difficult to measure. For instance, a client goes through the purchase steps in the e-shop over and over again, just to compare the offers. What is more, they will search for other operators' offer. This means that abandonment of shopping cart is by no means representative of anything, as clients will terminate the buying process on many occasions before they find and pick the right one. So, before you start talking about measuring conversion, you must correctly define the groups of clients.

Client behaviour on a website matters most

Let's imagine a John, who wants to buy a mobile phone with a subscription plan. Assume, he wants to buy an additional offer or is not a client of the operator in question. First, then, there are the potential clients - they show interest. They compare and look around to narrow down the field of search. If they decide on the price range, or a particular model of phone, they will start inspecting and selecting the offer, thus becoming a searching client. Once the offer is chosen and the purchase pursued, then we have a decided client.

What I mean when I speak of ‘conversion’ is two types of situations. First, let's call it a 'big conversion', is a positive finalisation of a purchase, of course. Yet in researching client behaviour on operators' websites, there is much more to take into account than just sales figures. Other types of behaviour must be considered, too. Something I call the 'small conversion' is of help as well – i.e. the fulfilment of some pre-determined conditions by the person visiting a researched e-shop.

I may buy a phone with subscription

Let's come back to John. He runs through the offers. He is not planning to change his mobile or the current operator, but - who knows – he may some day buy some additional subscription?

One cannot conceal the fact that potential clients are the hardest group to define. They are the individuals who entered the operator's page for the first time. Apart from that, the group could also include those who concentrate on one offer only. They simply compare several similar offers of different operators. Another premise indicating that an operator's website is visited by a potential client is, naturally, their interest in the offer tailored to new subscribers. In a nutshell: if data shows that a user is exposed to an offer for the first time, it is then that they should be treated as a potential client. In this user group, what we have called small conversion can be defined as performing an action on the operator's site. For example, moving a slider adjusting the offer range, scrolling down the page so that the hidden content can be seen, or other similar actions may be considered a potential user conversion. Also meaningful is the time spent on learning about the offer. The potential clients only compare the main price component, without delving into rules or details. This group will then be defined with relatively short visit time.

I seek and I compare

John knows he wants a phone for up to 300 PLN, it must be a smartphone and the amount of free minutes to all operators cannot be less than 250. He will then be defined as a searching user. He is going to compare, select offers and telephones matching his requirements. He is also bound to come back to the page frequently.

On top of the visit time and recurrent views, the information on where the user came from is also important. Entries originating from search engines after typing in concrete key words or after clicking on concrete advertisements give the operator the information on the user's preferences as well. If a user spent more time on the website that the average or used some of the internal tools to compare phone models or offer – this too warrants the ‘searching user’ label. For, in case of such user, the so-called small conversion may be defined as following the operator’s hint. As a result, the operator should actively support the user and precisely measure what steps they take within the website.

I have decided I want to buy that

If John wants to buy a Samsung Galaxy S4 with unlimited call subscription, this defines his as a decided user.

In this situation, the operator's actions are to be similar to the ones taken in the searching users case. Here, the telecom will have to employ its tools the right way, so as to facilitate the client's precise selection of offer and - finally – the purchase. The only thing such client needs from the operator is some limited assistance in personalizing the offer. A decided user can be identified by previous visits on the telecom website and narrowed scope of search. Such client also concentrates on those parts of the operator's website that help effect the purchase: shortlisting and option checking. A decided user is also characterised by a low come-back rate. In our case, a small conversion will take place if a client uses tips adjusting the offer or selects proposed additional services (not necessarily paid) on the way to finalize the shopping cart.

Shopping is a string of events

The entire shopping process is a string of events, particularly evident in case of telecom devices and services. Apart from the tools for traffic and website behaviour tools, it is important that the targets to be attained at each sale stage are defined. For example: John was looking for a phone for 300 PLN and subscription with monthly fee up to 100 PLN. As a result of the operator’s activities, a proposal of phone for 320 PLN and subscription for 130 PLN a month, John changed his scope of interest and went for a more expensive model. What he did not take into account is the more expensive subscription fee. The main aim of the operator was, however, to sell a more expensive subscription model – the goal was then not attained and the so-called small conversion did not take place.

Not before a full analysis of all components of a user contact with a telecom website can the range of conversion – both the ‘main’ or selling one, and the individual stages or the ‘small conversions’ as we called them – be determined. Thanks to the tools that can be used by clients themselves to define criteria for observing user actions and that can measure many different factors, the analysis can be complete and may include information on small conversions. To learn them is key in case of telecoms.