How PayPal uses predictive analytics to proactively plan

Ahead of his session in CX Network Live: Customer Insights and Data Analytics, Adeel Ahmad shares his insights on how PayPal is staying on top of the latest customer experience technologies, how to evidence ROI for new systems, and what new companies can do to stay relevant in the ever-changing customer experience field. 

CX Network: What is the most important CX initiative that results in frictionless, convenient customer experiences?

Adeel: At PayPal, we try to make our services effortless and frictionless and we’re creating experiences to align with that customer championing method.

We are embedding analytics into our processes to understand how we can serve our customers better so they don’t have to reach out to us.

The most prominent point of friction is when customers have a problem and they have to reach out to us. We want to ensure this process is as easy as possible, or instead, have self-service capability at the point of service.

We have been prioritising services for our most engaged customers and introduced new channels of interactions, like live-chat and other social channels, to give our customers easier ways to contact us. These strategies have provided very significant financial savings and efficiencies. 

We have also seen significant improvements in our contact metrics like contact rates, wait times over this period.

Read: How TGI Fridays uses VOC data to build personalised VIP experiences

CX Network: You mentioned that you use analytics data to map out where you can be proactive in terms of CX. Do you have any advice for Customer Experience professionals looking to overcome data silos? 

Adeel: To derive more value from data and deliver deeper insights, you must pursue a delicate balance between providing enterprise view insights for customers, versus high-value specifics to lines of business.

Silos are not always problematic. The real problem is when your only method of gathering data is through silos. The optimal organisational model is more of a Centre of Excellence model, where there is governance, programme management and best practice being provided from a central group. However, they’re not the ones doing all the analytics.

The analytics still need to be balanced between the central group and the departmental silos. But there has to be a very important balance of power between the two of them.

Read: The value of time in the experience economy

Ultimately, silos decrease the value of the analytics as they might not follow a customer throughout the entire journey, but it’s still important to strike that balance and keep your silos organised.

CX Network: Do you have any tips for brands at the start of their journey with artificial intelligence (AI) and machine learning?

Adeel: You’ve got to lay the foundations in terms of the platforms you’re using to make sure data governance is good. AI and machine learning have become buzzwords in which we sometimes perceive as a magic bullet to solve all of our problems. This isn’t the case. I see that enterprises are fumbling the opportunity to capture the data science and the whole opportunity – and are not being strategic enough with their use of AI and machine learning.  

My tips are to:

• Target business outcomes and do not focus on research studies at the start. While research studies have their place, target a business outcome. Once you have proven your point, do your research study.
  
  
• Build hybrid teams where there are interactions between the data science team, the data engineering team and the IT team. People miss out the IT teams often and that leads to failure. Plan a conjoined ecosystem where the three of them can operate.
  
  
• Build powerful platforms that augment collaboration, reduce the rework, and speed up adoption of the data products that you are developing. This is essential, because communication is key. It doesn’t matter how great your analysis is if you’re not able to engage users or engage people to enjoy it.

CX Network: How do we evidence the ROI from a new system or platform?

Adeel: Our customer experience investments are tactile and strategic, so it is hard to measure the ROI.

This is how we approach it. We try to select a quantifiable success metric. Preferably, a monetary one, but it doesn’t always have to be. It should be and there should be a success metric.

The most compelling story that you can tell your stakeholders is monetary related, so prioritise bottom-line metrics and measure hard currency wherever possible. When it’s not possible, don’t worry about it. As long as it’s quantifiable, it can be translated into dollars at some point.

We and many other companies have adopted customer-obsessed metrics, such as customer lifetime value and customer satisfaction, to gauge success. They are important metrics, but always ask yourself the question;Is this relevant for me?  Do not use the same metrics because other organisations are using them.

Read: How to be a CX leader: Lessons from FedEx

The metrics must be influenced by the project and be enforceable. You should be able to measure the impact of the work that you are doing and as a result be able to justify the ends to the means.

Also, when embedding analytics at the point of interaction, embed the success metrics here to make data gathering much easier. As you are trying to build out that customer journey and your various interaction points or friction points with the customer, associate metrics on what you will measure as you try to improve the process.

For example, if you are focusing on customer retention, measure your retention rate and renewal rate. If you are trying to measure your cross-selling and upsell potential, measure how much you are cross-selling or upselling.

Individual interactions and individual touch points need to have their own metrics. With CX, since it’s again tactile, the whole is not the sum of the parts. Your metrics should always be benchmarked against the industry. Otherwise, you won’t know whether you are improving compared to the rest of the market.

CX Network: Where do you think the future of customer experience is heading?

Adeel: For this question, I want to focus on analytics. From what I have seen, predictive analytics has entered the mainstream. Most of today’s customer experience analysts are not only focusing on customer response behaviour; they’re expecting behaviour as well. Predictive analytics has arrived big time for the customer experience space. 

I’ve also seen the rise of unstructured and untraditional data sources on the rise. Traditional data sources are still there; their growth has been relatively static. Traditional elements of CX such as customer emails and phone calls can now be analysed through current technology.

Read: Why the experience economy is key to CX success [Oracle Interview]

Last, streaming analytics, AI and deep learning, I think there are multiple opportunities out there for these technologies. They have become mainstream in a lot of senses, but I think the industry still needs to mature.

CX Network: What would be the CX challenge that you think businesses need to overcome to make progress and stay relevant in 2020?

Adeel: By 2020, analytical technologies, data science, AI, deep learning will enter the mainstream. More companies will utilise deep learning and AI will have a big part to play in how we drive customer experience.

Augmented experiences, augmented realities, and virtual experiences are no longer experimental. If the experiences will be driven by these new interactions, the new interfaces that we are going to have, analytics needs to play a key role.

If we are not analysing how these new channels are behaving, it will be hard for any company to understand how they are performing from an experience perspective. These new channels and interactions will be so wide that it will be necessary to get ahead of the curve and understand that it will be a reality.

CX Network: For those companies that do not adapt, it will become even harder to keep up and the gap from laggard to leader will only increase

Adeel: Indeed. From an experience perspective, we have always seen that customer experience delivers its value by filling that gap between the customer expectations and the profitability expectations of a company.

As companies utilise new channels of interactions with customers, and as machine learning and AI are starting to augment these new channels, there will be a greater need for filling in that gap between expectations and actual profitability expectations for the company.

Read: How to use data to create the ideal customer experience

If companies don’t do that, if companies focus only on profitability and don’t focus on customer expectations, I think they are setting themselves up for failure.

At PayPal, we always want to be ahead of the curve, so we are making investments and we are locking in these new channels and interactions.