opinion


The value of AI: are operators missing the big picture?

Machine learning , artificial intelligence , ai , deep learning and future concept. Wireframe Brain connect with circuit electronic graphic , binary code and technology icons background.Blue tone

Telecoms.com periodically invites third parties to share their views on the industry’s most pressing issues. In this piece Albrecht von der Recke, founder and COO of fonYou, takes a look at how the telecoms industry is approaching the vast potential offered by artificial intelligence.

Artificial intelligence (AI) is disrupting telecoms more than any other industry. From automating business processes to optimising performance, AI technologies have the potential to transform networks, and ultimately provide a better experience for subscribers.

The need to embrace AI is not purely for altruistic reasons. Mobile carriers operate in an intensely competitive market, which has seen new OTT players such as WhatsApp and Snapchat competing within carriers’ traditional realm of voice and messaging services. Business-as-usual is simply not an option.

Given the current market conditions, therefore, and with expensive spectrum, network upgrades, and marketing requirements to contend with, there may be a case for operators to harness the power of AI to build their businesses in addition to tweaking and optimising their networks.

Personalised and contextual

Operators are sitting on a goldmine of personal and usage data, and AI is the key to unlocking its full value.

Powered by AI and with access to this wealth of data, operator can employ sophisticated micro analytics to automatically assess the needs of subscribers, analyse their usage and purchase history, and formulate an offer that matches their specific requirements and context.

Consider a prepaid customer that has run out of credit; an event that will trigger a response on the network in which the customer’s usage and payment history is analysed in real time. Having finalised a risk profile, a ‘just-in-time’ offer could be automatically sent out to the customer, such as a financed data package or the opportunity to top up by credit card. This offer will be delivered using the channel to which the customer traditionally best responds, before the payment is processed and the selected service provisioned.

In this example, AI algorithms will have intelligently computed the right sale to push, specifically aligned to the customer’s immediate needs. Not only is it completely personalised but, by using the insight gleaned from analysing the customer’s historic data, it also completely contextual. This may not sound a particularly taxing task in and of itself, but it’s worth considering that tasks such as this will occur millions of a time a month on any given mobile network, all in real time.

It’s clear, then, that combining AI technology with the right transaction execution and the right channel engagement can provide a real boost to operators’ bottom lines – particularly at scale. Indeed, several operators around the world have already seen an increase of several percentage points on their ARPU having implemented such solutions.

At the centre of all customer interactions

Each month, hundreds of millions of events are overlooked by operators which, if properly exploited, could represent an opportunity to generate revenue. Indeed, even by monetising just 10 percent of these events, operators would enjoy significant additional revenues and increased customer satisfaction.

By way of illustration, take the example of a subscriber on the verge of running out of data from a 1GB prepaid bundle, who doesn’t have a credit card and whose bank balance is too low to allow him to purchase an additional bundle. Under the traditional prepaid model, mobile data service would typically be interrupted or downgraded once the bundle’s data allowance has been consumed. Applying advanced AI technology, however, will enable the delivery of a service much better suited to the individual characteristics of a specific customer.

This is particularly attractive in a prepaid-dominated market with low banking penetration. A smarter prepay platform might, for example, leverage historic data which indicates that this particular customer shows unwavering loyalty, and has a perfect credit record. This being the case, the platform’s micro analytics engine will push out an offer to the customer of the same 1GB bundle he normally buys, but this time in advance, and with up to five days in which to perform a top up. The decision is made, and the offer pushed out in real time and in the digital space, removing the need for the customer to visit a physical top-up point, and avoiding interruption of his mobile data service. As a result, the customer will be satisfied with the experience, and the operator will enjoy increased revenues and lower costs.

Understanding usage and context

Very often, simply recommending the right offer at the right time will result in a customer accepting a deal which they consider to be fair and that meets their immediate needs. By employing artificial intelligence to make these personalised, contextual offers, based on continuous analysis of the huge volume and variety of user data, will not only improve the customer experience, but will bring consistent new revenue to the operator over time.

Intelligent repackaging of an operator’s existing portfolio could create additional new revenue and ARPU increases, too. Bespoke offerings could be created around application-based packages including Netflix or Spotify, for example, social media bundles, or new handset deals.

Fundamentally, all of this comes down to understanding usage and context to ensure that each offer meets the specific requirements of any given user at any given time. And with AI in place and cloud tools, it’s possible to do this without requiring humans to do the heavy lifting.

AI is already transforming many key areas in telecoms, improving network efficiencies and customer service delivery and, with digital transformation projects well underway, more change is yet to come. Looking beyond this, however, and utilising AI-led offers powered by micro analytics, operators have the opportunity to direct impact sales and generate revenue, and more confidently face the challenges of today’s increasingly competitive marketplace.

 

Albrecht von der Recke_fonYouAlbrecht is one of the founders of fonYou and has over 18 years of hands-on experience with mobile carriers. He has led successful service launches with mobile network operators – One (Austria) and Orange (Switzerland) and, as an entrepreneur, in several launches of MVNOs (mobile virtual network operator). Prior to fonYou, he cofounded NextGen, a consultancy specialized in telecoms. Albrecht holds a degree in international business management from the University of Vienna.

  • Telco Automation Everywhere

  • 2020 Vision Executive Summit

  • BIG 5G Event


One comment

  1. Bernard Szederkenyi 16/02/2019 @ 4:57 pm

    Contrary to Gartner’s Hype Cycle, AI is likely at early stages of the “trough of disillusionment”. Inflated expectations re the ease of implementation, the millions of use cases are certainly creating some disappointment.

    It doesn’t have to be that way. We are a Canadian mobile carrier with a few million subscribers in our base. Predominantly postpaid customers. Like all companies with that many users (banks, cablecos, telcos, insurance) one of the larges operating costs is customer service. Notwithstanding all the digital tools and increasing adoption of them, companies still spend tens or even hundreds of millions a year to respond to customer inquiries. For us 30-40% of these inquiries were low value, low complexity contacts.

    In order to break hardwired customer behaviours that companies have built over the years (contact us if you have any questions) we took, what some call a drastic decision. Before any customer is able to reach a rep they are put through a triage system. A path that leverages data (for personalization), assisted AI to communicate and Genesys’ scheduled callback feature if they still need the help of a rep. No live hand off, no more inbound contacts.

    This e2e approach freed up millions of OPEX $ that we are now reinvesting in higher value activities.

    So, it can be done. But it’s not as simple as visiting your local Best Buy, grabbing a $50 box labeled AI for All, slapping it on your website, contact centers, social platforms and expect to all go well. :-)

Leave a Reply to Bernard Szederkenyi Cancel reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Polls

Should privacy be treated as a right to protect stringently, or a commodity for users to trade for benefits?

Loading ... Loading ...