Braintrust Query: Why Does So Much Customer Segmentation Analysis Lead to So Few Results?
Through a special arrangement, presented here for discussion is a summary of a current article from Cultivating Your Customers, the M Squared Group blog.
As I recently thought about customer segmentation analyses I have worked on over the past 10 years, I realized that the actual segmentation initiative has succeeded in fewer than 50 percent of those companies.
While many analyses were deemed successful by identifying clear and distinct segments, truly successful initiatives are characterized by specific actions that will add value — clear, identifiable customer value.
There is a wide range of reasons why organizations are resistant to implementing customer segmentation initiatives. These three reasons represent some of the greatest barriers to driving customer and company value out of customer segmentation:
- Mismatch Between Metrics and Desired Behaviors: If company metrics are quarterly-based or, even worse, monthly-based, employees will be unlikely to focus on initiatives that can yield higher returns in the long term. Since segmentation increases marketing costs (creating different offers, creative, etc. for each segment), and the returns may not come for several months down the road, you see the problem.
- Conflicting Priorities: A segmentation initiative may be the top priority at the outset, but over time other projects appear more important. Enough short-term projects falling one right after another may cause segmentation to fall off the list entirely. Since segmentation at the beginning is an unknown, and workloads are already high, the project is often delayed in order to "make the numbers, now."
- No Repeatable Process: If no process exists for evaluating marketing programs on a regular basis and the segmented effort is also not measured, then the effort as a whole is doomed to fail. If you do not determine that segmentation is successful as a strategy, why should you repeat it?
In addition, segmented marketing programs are more work the first time around. The only way they get easier is if you plan for multiple versions and then put the work into its time slot. When a marketing department is either overwhelmed or just basically reactionary, any approach but the simplest goes by the wayside in the name of expediency.
And there goes segmentation…down the drain.
So if you are starting a customer segmentation initiative, or find yourself in the midst of one right now, you should take a few minutes and examine these three factors to see if they apply in your organization. More often than not, they will.
Discussion Questions
What are the reasons organizations are resistant to implementing customer segmentation initiatives? What’s the best way to overcome such hurdles?
Newton is dead. The universe is not organized around ‘segmentation’, it’s organized around alignment. It’s all ONE thing.
Maybe this resistance or “hurdles” are the answer, not the problem.
Stand back and enjoy the butterfly and stop dissecting the cocoon. Breathe people, breathe.
Organizations are inherently resistant to change—they exist to provide order to chaos. Most segmentations I have seen argue for change, usually big change, so resistance is natural. Add to that the apparent inability of American corporations to think long-term and the effort is blunted from the beginning.
We haven’t been very successful at overcoming such hurdles yet—don’t see that we are going to change American corporate culture in the near future either.
I think the simple fact is that many fail simply because they lack sustained management support. Put another way, segmentations are seen as useful when building PowerPoint decks, but not so useful when a company is scrambling for results.
If management teams truly believed in the power of segmentation to deliver improved operating results, they would invest in them, make them the highest priority, and institutionalize the process.
In the end, marketing tools rarely make the final prioritization cut—especially when results are shaky.
In addition to Mark’s three barriers, you can add: employee turnover, internal inertia and the simple fact that consumer-focused marketing is much harder than product-focused marketing. Oh, and one more—many CPG marketers are so in love with their products and so tied up in internal justification processes that they have no interest in segmentation marketing. Let’s see how many other barriers people can come up with….
Segmentation is an interesting problem for data analysts to address. However, if it is a data analysis job, it is not likely to succeed. Segmentation needs to be part of a multifunctional, integrated approach to marketing.
First, knowing and understanding the consumers must include more activities than just analyzing a discrete data set. Second, if and when clear differences are identified there needs to be testing to determine whether these groups really respond differently to messages. Third, the marketing program needs to design messages appeals so that each segment receives a different message. Fourth, metrics and measurements need to be used. Assuming segmentation is just a data analysis project will not lead to success.
Another key barrier to success with segmentation is that these studies often don’t give actionable insights that are easily translated to retail partner discussions or activations. Plus, retailers have their own segmentations, and the crossover of data is rarely easy.
A million dollar insight is always trumped by an actionable insight gathered through keen observation and understanding of the purchase journey, the retail situation and the emotional drivers behind the brand selection.
Many small business folks out there can not even grasp this idea in the first place, as it is overwhelming to digest. I am a fairly simple man, who stay’s on top of the business trends, and have not used this initiative before. Larger corporations who can use this initiative may be able to utilize this to their advantage, but is a daunting task for the rest of us.
There are a ton of ways to reach out to customers, but once they stop in, the critical part is keeping them coming back, and that is where the Human Experience kicks in. If you can build trust one-on-one with someone, than the rest is very easy, as the relationship is secured for many years to come.
Too often, segmentation becomes the flavor of the month within marketing initiatives. Without commitment of resources—effective amount of time, and sufficient (and efficient) amount of funds—this initiative can quickly fall by the wayside. In addition, companies too often don’t prudently narrow the number of segments they are pursuing. No marketer has been successful in attempting to be “all things to all people.”
By providing marketing associates with useful and appropriate analytical tools to complement databases, organizations position those associates to effectively and efficiently data mine to pursue the most valuable segments.
By providing marketing departments with a Consumer Equity Index of their shoppers, CMOs position their associates to both scale the market opportunity within segments, and launch them in the direction of uncovering the brightest prospects for retention and growth. When the right tools are given to associates, segmentation works because resources are placed on the best strategies in the quadrant matrix of Grow & Protect, Invest, Harvest, and Abandon.
Very interesting topic and one that needs far more discussion than it currently gets. In addition to the three reasons that Mr. Price lists as barriers to segmentation success, in my experience both as a practitioner and a consultant, most retailers that begin segmentation fail to understand how to connect the homogenous behavior of the segment with relevant content, thus rendering the segmentation schema somewhat ineffective.
But the largest obstacle is Mr. Price’s #2…conflicting priorities…meaning that retailers still first and foremost manage their business at the aggregate, district, and/or store level. Going more granular than that is sometimes difficult as progress in shopper segments needs to show connectivity back to the overall business, which does not happen fast enough for many to value.
Good topic!
Mark, these are all great points. One observation we have made is that organizations incorrectly apply a “one size fits all” segmentation scheme across their entire suite of products. For example, in the financial services space, two households may have a similar profile as relates to deposits, but be totally different as relates to home equity, credit card, etc. Organizations should replace such schemes with a product-level segmentation framework to obtain a much more accurate prediction of product-level needs, attitudes, and buying behaviors. The returns on investment will be dramatically higher. This sounds challenging, but the lift in results is well worth it.
It’s only natural for an organization to be innately resistant of any process that multiplies complexity and effort while delivering unclear returns. It would be unfair to blame this on a bad attitude.
Segmentation may indeed deliver superior performance under ideal circumstances, but sophisticated analytics are only the beginning.
Each segment demands a relevant strategy and associated tactics and plans. Each planned action must be designed to be executable and the fact of its implementation tracked and evaluated.
Impact on performance must be measured, and the very painful question must be asked on a continual basis: “Did this targeted activity deliver a marginal benefit that justified the added investment?”
For retailers with multiple categories, this makes life a lot more complicated. For brands selling through multiple retailers, this makes life a lot more complicated. Since that’s just about everybody, we can conclude that segmentation is hard to put into practice.
The analysts and planners may scratch their heads over this, but they may not appreciate that the data analyses behind targeting are really the easy part. Implementation planning is the crucial missing element. Without that, performance management is well nigh impossible.
I have found that building recognition for the need for segmentation and procuring the needed resources is the first hurdle to overcome. Once a project is secured, the points in the article do come up.
As marketers become increasingly aware of the value of data, the issue of conflicting priorities should reduce as data analytics moves higher up the chain.
To ensure that segmentation exercises meet their intended result and become a renewable element of the marketing toolkit, a solid measurement plan needs to be established with clear metrics of success in place.
Does this mean that 50% succeeded?
I am very skeptical of segmentation analysis, although I use it myself. I think a 50% successful usage rate would be outstanding! The reason is that, at root, segmentation is about lumping shoppers into groups so they can be more intelligently targeted. The segmentation is going to be based on an array of behavioral data.
How do you know that the behavioral data you are using is actually relevant to what is needed? A lot of segmentation is simply using nifty math to organize the ignorance of the marketers using it. GIGO.
And even if the segmentation is the very best, as with any business initiative, activation/implementation is the killer.
Then there is the little complication that customers are chameleon-like, with distressing tendencies to ignore your segmentation boundaries, and move from segment to segment with reckless abandon.
Segmentation can be a powerful tool, but its potential requires genius in managing it. That wouldn’t be me. 😉
Could the problem be the commitment made is the C-level nod of approval, but not the commitment to put their name behind it to see it through? So the program is operating, then what? It goes to Tony’s comment about doing what is needed to get them to buy something and, more importantly, come back again and again. Unless that happens, nothing has happened.
Similar issues exist in both retail and CPG. Both have initiatives that are well-intentioned, however, these obstacles and more tend to make positive results difficult to achieve.
I think too much of marketing processes are still “gut-feel” driven, even in the largest organizations. This is one area that technology can lead to the answer. I don’t always defer to technology solutions, even though I’m an IBMer, however, in this case, I know there are great, simple-to-leverage tools available today that capture POS and other data, cleans it, analyze it and create actionable insights from it. Consumer segmentation may be only one potential output.
Thanks to everyone for their valuable contributions on my post. I think that executive support is one of the critical components that is often lacking in customer segmentation initiatives.
The challenge with implementing customer segmentation is mostly about change, and the solution must be consistent with best practices in change management, including executive sponsors, clear, identifiable wins, metrics that are tracked on a frequent and visible basis and repeated, consistent communication among others.
Segmentation is only the start of the process. Once segments are established what is the practical action to be taken? Localized assortments, specialized offers, personalized communications? If the segmentation information isn’t turned into a marketing, or merchandising activity then it’s just nice to know.
A cross functional commitment and an action plan have to be part of the solution. Sometimes that’s harder to do then just parsing the data and coming up with the segments.