In part 1, we discussed some of the findings of the recent UK Institute of Customer Service report on Customer Satisfaction Index and highlighted some customers using Splunk and machine data to improve customer experience. There’s a great video from my colleague Rahul that gives you a good summary:
We often find that companies start using their machine data in Splunk to spot issues and get alerted to possible incidents. Very quickly the people who are looking at the data and what they thought was a system of record realise that the same data in Splunk can be used as a system of engagement. Consider the screenshot from a demonstration of Operational Intelligence (click to enlarge):
In this case the machine data collected is used for many purposes (how the architecture is performing, possible security incidents, etc.) but in this fairly simple dashboard you can see:
- How many transactions we have right now
- Average basket revenue
- Top products in the last 24 hours
- Top items sold in the past day
- Top items removed from a shopping basket
- Where sales are happening globally
Consider the customer examples from Domino’s and a leading online retailer described in part 1 and how they use their machine data for real-time promotions and apply it to the dashboard above. If you can see that a lot of people are removing a particular product from their shopping basket in real-time (possibly because a competitor has a promotion on the same item) then Splunk can alert you to the fact. The retailer could then automatically send a voucher to the customer who took the item out of their basket (or notify them in real-time on the site) that there is now 20% off the item they just removed. If you can combine the machine data with social media data about competitor promotions then operational intelligence suddenly becomes a very important way to differentiate with customer experience to positively affect the revenue and customer loyalty.
The key here is that this is unstructured, real-time data that has always been seen as “the data that IT is interested in”. This is often the case but the potential of this same data that is already collected and available for improving customer experience is often untapped. We’re seeing an increasing trend of organisations using Splunk to exploit the potential of this data beyond our core use cases of IT operations and security. If you think about how ITOA is emerging as analytics for IT and how security-driven analytics is changing the SIEM space, next up has to be something in the realm of real-time customer operations analytics.
Consider the following implications for customer experience from machine data and Operational Intelligence:
- Ensuring quality and experience of omni-channel services. Mobile and online are as important, if not more so, than the in-store channel. Making sure these channels are performing and delivering the right experience is crucial for success. Machine data and Operational Intelligence give you a very good chance of getting this right and making sure all your channels work together to deliver a consistent experience.
- Modelling the customer journey.com is a great example of this. Previously they thought they knew all the possible routes a customer could take through their e-commerce site. When they modelled the actual customer journey from their machine data, they found reality was different from their model. This has enabled them to improve the customer experience online.
- Personalisation through a 360-degree view of customer. If you can combine your historic customer data and what is happening real-time across multiple channels, a retailer has an opportunity to personalise offers, advertising and a streamlined experience. Historic data may reside in a data warehouse, Hadoop or a relational database. Real time customer behaviour comes from machine data and OI. Put these together and you’ve got a good view of your customer to make more informed real-time decisions.
- Customer analytics in real-time. You may have heard me mention “business moments”, a term I heard first at a Gartner event. The idea of real-time analytics from “big data” to make a more informed decisions is a great example of this. If you can measure and visualize what is happening with your customers in real-time from a large number of fast moving data streams and start to predict what is going to happen next – you can start to make some very interesting decisions about customer experience. They key here is to democratise the data and allow anyone to “self serve” and create their own analytics (with the right data privacy framework in place).
- A real-time, single view of product. The other side of the 360-degree view of customer coin is the question of “how do you see a real-time view of a product?” For example, we have a leading EMEA retailer measuring who is buying which products. They have effectively created a single view of product from a combination of structured and unstructured data. For example, who buys a particular chocolate spread, when do they buy it, how much in stock etc.
- Measuring release success. With the rise of Agile, DevOps etc., measuring the success of an application, site update or new version of a mobile app is important for customer experience. With much faster release cycles, it is important to make sure that the customer experience is at least the same, if not better with each release. This comes from machine data. This is also true for the introduction of new features. For example, Shazam measure the success of the introduction of new features in their mobile app. With a combination of A/B testing and monitoring new feature adoption in Splunk they can now promote new capabilities and be confident that customers are seeing and using what’s new.
- The role of IoT and sensor data. This is starting to become a hot topic and we’re seeing an increased interest in how to use the data from iBeacons in-store with mobile apps and historic customer data. These are obviously very different data sets that probably haven’t been put together in this combination before. As someone asked me at an event recently, “how to I get as good a view of customer behaviour in-store as I can get online?” There is an important consideration here for retailers – what’s the right balance between using the data that can be captured and not annoying customers with overly intrusive in-store push promotions.
If you’re already using Splunk and want to see how to get started using the product for monitoring customer experience there there’s a five minute video below:
I’ll try and finish where I started – customer experience improves business performance.
Those organisations that get customer service right are growing. Those that don’t are being significantly outperformed by their competition. The complexity of pulling together data from lots of different channels, the combination of cloud & on-premise IT and the need to deliver self-service analytics in real-time isn’t making delivering an excellent customer service any easier. From what we’re seeing at Splunk, machine data and Operational Intelligence is making a difference when it comes to customer experience. If you want to find out more, a good place to start is the Splunk solution guide for retail.
As always thanks for reading.