Living up to Expectation: AI,Scalability and Automation
CIOREVIEW >> Canada >>

Living up to Expectation: AI,Scalability and Automation

David Gonzalez, Head of Big Data and Advanced Analytics, Vodafone Business
David Gonzalez, Head of Big Data and Advanced Analytics, Vodafone Business

David Gonzalez, Head of Big Data and Advanced Analytics, Vodafone Business

Big data and artificial intelligence (AI) are topics that have been on the agenda of every business event in recent years. There continue to be several impressive use cases across different verticals, but too often these are bespoke proof of concepts with little return on investment.

Despite the hype, the branch of science has yet to really hit the mainstream and fulfil its potential. We are yet to see an embedded process that drives long term results across an organisation.

To overcome this and make big data and AI a real success which drives business impact and adoption, the most important factors to consider at each stage are scalability and automation.

Starting the journey

To properly demonstrate a significant business outcome, it is first important to differentiate between bespoke and scalable projects. For an early adopter, any use case is great progress, but it is important to understand that this is just the beginning of the journey and will not drive long-term value alone.

The retail industry is a great example of this, with AI and predictive analytics being used to deliver results already. Many retailers are now using chatbots to answer customer queries quickly and efficiently.

 With the modern customer more demanding than ever, big data and AI can help retailers deliver a personalised experience​ 

While successful in the task and a nice-to-have, they are unlikely to really drive profits and impact a retailer’s bottom line. With a more holistic and scalable perspective, there are many other touchpoints on the retail journey which could yield additional impact.

Scaling up use-cases

With the modern customer more demanding than ever, big data and AI can help retailers deliver a personalised experience. For example, AI can be used to show live customer recommendations or tailored content marketing at each stage of the customer journey. This is likely to drive engagement and ultimately increase customer spending.

From a broader network perspective, predictive analytics can be used to identify the need for enhanced resources on a given day. This can deliver tremendous value for retailers who offer regular seasonal sales which can put a strain on infrastructure. Predictive analytics means that retailers can predict and overcome this additional strain.

On top of that, retailers can tap into the vast potential of anonymised data for further customer’s insights. This can range from credit card usage to predict GDP and inflation to the economic potential of different locations. This makes it quite possible to go from delivering tens of use cases a year to multiple hundreds or thousands.

Rather than designing a dozen or so predictive models, organisations should be looking to implement an end-to-end analytics factory capable of building hundreds of models in weeks.

Maximising data’s potential

Despite this, there has been widespread conventional understanding of AI for several years. This is especially true for use cases with a clear objective, such as reducing customer churn in long-tail organisations like Netflix, Amazon or traditional telcos, which target a large number of niche markets in a highly competitive sector.

As digital adoption grows, so too does the data with it, and this provides an opportunity for organisations to maximise their customer interactions and monetise this momentum.

There are five things that organisations need to achieve this:

1. A clear business opportunity - There must be an actual opportunity to improve the business, as well as an understanding of its potential value

2. Good data knowledge – This means an organisation has in place data governance, metadata, dictionaries, and processes to ensure good quality data is ready to be used

3. A cloud platform ready to build and process large amounts of data at scale – AI requires the processing of large amounts of data, which an on-prem solution probably won’t be able to handle

4. The skills required to apply AI effectively – It is very difficult to find experts in this field, so in-sourcing this talent is a priority

5. An agreement that AI adoption is for everyone – The outcomes generated must be integrated into all digital channels to maximise results in a seamless way

Driving real success Realising the true value of AI isn’t easy and requires many teams collaborating and aligning on objectives. This means that internal transparency across all departments is essential.

To get past just delivering demonstrations and case studies, businesses must approach AI with scalability and automation on top of mind right from the start. This will ensure any investment drives real change, and we will see big changes in the effectiveness and value realisation of AI and data analytics.

Read Also

Cloud At The Edge

Duncan Clubb, Head of Digital Infrastructure Advisory, CBRE

Edge Computing - Where Does It Fit Today And Tomorrow!

Adel Bekhiet, Senior Director of Infrastructure & Cloud Services, Northwestern Mutual

The Evolution of Digital Banking Landscape in Indonesia

Altona Widjaja, Head of New Digital Venture, Bank OCBC NISP

Banking Preference Shifted: Moving Away from Traditional Banks

Supaneewan Chutrakul, First Senior Vice President, Kasikornbank

How Opendoor Platformized Inspection Tooling for Self-Guided Assessments

Salman Jamali, Head of Engineering, Strategic Initiatives, Opendoor