Turning VOC into MVP
(Customer) Data is the new Currency!
"Cash is King" they often say, but in this modern world, the new truth is that "data is king" and there's no better and more valuable data in my mind that customer's inputs about your product or service. They can easily provide feedback on every aspect of your experiences -- from products, services, and in general the overall customer experience -- all with just a few clicks.
As a digital strategist, customer service leader, or commerce executive, it's essential to listen to customer feedback and turn it into actionable steps to improve your digital strategy. I've often found that companies are great at gathering the VOC, and don't do a great job incorporating it into digital product management or commerce leadership agendas. In the below paragraphs are a few ways I've found helpful to move insights into actions on your digital assets.
Three ways to move from VOC-to-MVP
1. Gather feedback through various channels and analyze
The first step in turning customer feedback into actionable steps is to gather feedback through various channels. These channels can include surveys, online reviews, social media, customer service interactions, and website analytics. It's crucial to collect feedback from a wide range of sources to get a comprehensive view of your customers' experiences. Don't just go off of random quotes on Sprinklr or the broadest generalized view. Try and combine and tie-together all the data you're seeing.
After you categorize and prioritize the feedback the more important part is to then understand what is causation vs. correlation. Just because there's a lot of feedback happening doesn't mean that what the customer is saying to change is what should be done. The cause of the friction for example can be system architecture and poor tech design, but it's correlated to repeated issues about "staying logged into the Mobile App".. and those login issues may be part of the reason sales dropped, but not the reason.
Another critical miss is that I often see the CX leaders espouse numerous datapoints and share feedback from the likes of Sprinklr, InMoment, Medallia, but there's not a mention of frequency and duration of the issues. The login issues above, while important, may just be happening to 7 out of every 10,000 customers, and only happen one day of the week.... is that worth derailing a product manager's agenda to fix? It may or may not be, but in addition to 'impact' the frequency is also critical to understand.
Recommended by LinkedIn
2. Develop process for review, incorporate into OKR's and agree on plans
Sounds simple, right? Just have the meeting every two weeks or so, and hope that the product manager takes action on it. My experience has been it takes much more intentionality.
Something I did in a prior product role was allocate 15% of the total sprint capacity to "known-unknowns", the things you know were happening or likely to, but were not sure. So in a sprint of 80 pts, dedicate 11 pts. That could be a substantial and impactful "11 pointer" or 2-3 key bugs that have been causing friction. Over time, fixing those 2-3 every two weeks will add up to a better overall experience.
One thing as well is every 3-4 sprints, dedicate it almost entirely to bug/experience issues. From a relationship standpoint, the product and CX organizations can know that there's a dedicated time and space on a recurring cycle to address the major issues. This helps in prioritization and the 'negotiation' that happens on what to fix, and when to fix it. Consider using quarterly increments to tackle the larger topics, which I've found to be more related to architecture, antiquated systems and at an organizational level when there's multiple brands or geographies each with competing interest.
3. Implement changes with a customer-centric approach
The final part is, ensure you have a baseline of the issues and frequency before hand, and then measure the impact afterwards. Objectives + Key-Results will help measure this in a framework, and those "KR's" about specific things will ultimately add up to a better overall "O" (Objective).
So for example, say your objective for Q2 is to increase post-purchase NPS by 3 Pts... you can "measure by" reducing the number of errors of add-to-carts from 15% --> 5%. The glue that pieces together that NPS objective and the key-result of errors is effectively what the VOC is telling you.
A final step could be to use tools like A/B testing or customer satisfaction surveys to track the results and make adjustments as needed. There are options like Adobe Target, and I personally like Optimizely's solution as well.
Wrapping it up!
In conclusion, turning customer feedback into actionable steps requires a systematic approach that includes gathering feedback from various channels, analyzing the feedback to identify actionable steps, and implementing changes with a customer-centric approach. The process of moving from Voice of the Customer (VOC) to Minimum Viable Product (MVP) requires a well-defined process that includes change management, testing, and measurement. By following these steps, businesses can improve their digital strategy and deliver a better customer experience.