I have noticed a pattern in my conversations with retail marketing and loyalty teams looking to make the most of their customer data: a shortage of ideas is rarely the problem. Instead, the conversation usually turns to the limitations of the tools they are using. They know what they want to do. But the systems they use to access customer data are too slow, too rigid, or controlled by other teams with different agendas. It makes sense that many retailers are currently re-examining the tools they use to analyze customer data and use it for personalization and monetization.
The dominant model in the past decades has been the service model, where retailers outsource the storage, analysis, and access to their customer data to a third-party provider. This model has several obvious advantages. It allows the retailer to leverage expertise the service provider has gained from working with a range of clients around the world and does not require a heavy time investment from internal resources who are busy with other projects. In many ways it functions as a ready-made, turnkey solution.
But the approach is not without its downsides. Because it is service-intensive, it comes with a high, ongoing price tag. Many retailers also find that the algorithms and data science they use to drive key business decisions function as a “black box”. Service providers may share the output of a business question but not reveal the logic that got them there.
Of course, outsourcing to a service provider is not the only option. Some retailers decide instead to build solutions in-house. Again, there are obvious advantages here. A solution built by internal teams can be highly customized to your business. There is no costly service contract after the project is completed. And of course everything is completely transparent – you never have to wonder how a particular calculation was made.
But building the solution yourself is not straightforward. Projects are usually being balanced with multiple internal priorities and can easily run over time or over budget, often resulting in reduced functionality. Customer data science may not yet be a core competency, so the work often does not reflect leading-edge thinking and practice. And these solutions are rarely flexible and scalable enough to adapt as the business evolves.
Obviously for many retailers, one or the other of these approaches can be the right solution. It is easy to list winning retailers on both sides of the fence who have mature and effective customer data practices. But there is a third option emerging that combines some of the advantages of each approach.
It is now possible to purchase a customer data solution that is sold as a product, not a service. Think of it as the difference between hiring someone to do your accounting for you, and purchasing a software package such as Quickbooks that will allow you to do it yourself. These software-as-a-service solutions can offer many of the benefits of the outsourced service model, but without the high price tag. Retailers using one of these solutions will have direct access to their customer data and a set of tools designed to make it quick and easy to analyze and visualize the data. The approach will not be right for everybody. But for those who have tried one of the prevailing models and are looking for a change, it may just be the best of both worlds. Let’s examine four key advantages of this model.
1. Avoid Outsourcing a Core Function
If you are opening a restaurant, you may consider outsourcing the cleaning, bookkeeping, or interior design. But you probably wouldn’t outsource the cooking. Every business has elements that are core drivers, and elements that are support functions. Outsourcing support functions can make sense. But can you imagine a retailer outsourcing category management or store operations?
Understanding your customers through data analysis and building a relationship with them through personalization are not peripheral to retail. They are at the very heart of the business. Trusting a third party to take care of is a risk that many retailers are no longer willing to take.
2. Fast Time to Value
A custom solution can be powerful. But it can also take time to build. Waiting months or even years for an internal project to be completed may be an option for some. But for others, every week counts. If time is critical, a product that is ready to go “out of the box” can be ideal. Even for large-scale, complex retailers, implementation can be done in weeks.
3. Stay Current
Internal IT teams who are laser-focused on getting a project done can deliver outstanding results. But what happens after the project is completed and resources move on to other projects? Features that were not included in the original scope (or that were pushed to the dreaded “Phase II”) may never happen. Likewise, changes in technology or advances in current thinking may be tough to integrate into a legacy system. This is especially true if there is no burning platform and what you have today is “good enough.”
One big advantage of buying a dedicated software product for your customer data needs is that you will benefit from a steady stream of upgrades and enhancements.
4. Keep a Bigger Slice of the Pie
A key feature of many outsourced customer data models is the requirement for a revenue sharing agreement in which service providers keep a portion of funds from monetizing data to suppliers. There is no doubt that some of this cost is justified by administrative support, sales support, and customer service support from the service partner. But at the end of the day, most retailers do not like sharing their revenue with someone else. Purchasing a platform that allows you to monetize data with suppliers without being forced to share the revenue means more money in your pocket.
Different retailers are at different stages and have different capabilities. Ultimately there is no one solution that will be right for everybody. But if the high cost of outsourcing and the hassle of building it yourself both seem to come up short for you, consider the middle ground. It just might end up being the sweet spot.
