“Which CX improvement project should we do first?”
Forrester’s customer experience (CX) research team gets that question a lot. The good news is that picking the right projects and doing them in the right order can be relatively easy — provided that the customer experience team sets up its measurement program to populate a simple but effective prioritization model.
The Two Rules Of Prioritization
Here’s what we tell our clients — the CX projects you should do first are the ones that will:
- Do the greatest good for the greatest number of your customers.
- Do the greatest good for your business — specifically, for its revenue and profitability.
If you know what projects will accomplish both Nos. 1 and 2, then all that’s left is to factor in feasibility and risk, which is something any good project manager should know how to do.
So how can you determine the projects that will do the greatest good for both customers and the business?
CX Measurement Done Right Feeds Prioritization
I recently wrote about two ways to determine what problems are affecting customers and how many customers are being affected by any given issue. CX professionals can do this through surveys and text analytics or by asking their contact center to tell them what people are calling about. When your measurement program asks customers about their intended loyalty as well as their satisfaction, you can also model how big the impact of making an improvement will be.
The details of how to do what I just described are in those two previous posts. So if you haven’t read them, you can either go do that or else take it on faith for now and fill in the blanks later.
A 2×2 CX Prioritization Matrix
Advanced prioritization models can get complicated. But a basic — and still very useful — model consists of nothing more than a two-by-two chart and a scoring rubric. To populate the chart, just estimate:
- Impact to customers. Award one point to a project that will improve the experience for just a few of your customers and two points to a project that will affect many or all of your customers. You get to decide the definitions of “few” and “many” based on your business dynamics. If this becomes a sticking point, then just stack-rank the number of customers for each project and draw a line at the midpoint of the stack.
- Impact to the business. Award one point to a project that will generate revenue or cut costs for your business and two points for a project that will generate significant revenue or cut significant costs. As above, you get to define cutoffs for significance (a stack ranking works here, too). The more important need is to determine the threshold below which a project is insignificant. If you can’t come up with any quantifiable business benefit for a project, don’t do it — this is business, not the Summer of Love.
Once you’ve scored your proposed projects, you can drop them on your 2×2 chart. As always with a 2×2, you’ll have the 2, 2 quadrant in the upper-right corner. Those projects, which will improve the experience for many customers and produce significant economic value for your business, are the ones that you want to do first. The projects in the lower-left quadrant won’t affect many customers and won’t produce much business value, so you should do them last.
What about the two remaining quadrants? One will have projects with the biggest positive impact for the business, and those are the ones I’d do second, unless they are so resource-intensive that you can’t afford them right now.
Speaking of which, before making your final prioritized list, you’d be wise to run the numbers through a basic ROI model that will let you make a one-sentence business case. That little extra bit of effort will help you make choices that will be superior to the ones being made by over 90% of the CX teams out there — including the ones at your competitors.
What I’ve described in this post is all you need to get started. It will put your CX improvement program on very sound footing. And as your CX program matures over time, there is more you can do to make even better prioritization decisions. If you’re a Forrester client, when you’re ready to advance, you can download a more detailed model with more granular criteria, plus the ability to apply weightings and factor in risk.