“No, Bill. You don’t want wider seats.”

12
151 views

Share on LinkedIn

At a recent dinner party I explained what I do for a living. One attendee responded, “Well, then can you please call Delta, and tell them I want wider seats?”

I responded, “Actually, you don’t.  People say they want wider seats, but their behavior says that they really don’t.”

“Oh, you mean the hypothetical general public doesn’t want wider seats?”

“No, Bill.”  I responded. “I mean that you specifically don’t want wider seats.”



—————-

The problem with many customer experience surveys is that they recommend the equivalent of “make my seats wider.” It’s a common practice to ask customers to rate importance for different factors, then compare that to satisfaction. But it just doesn’t work.  Since you measure each item in isolation, everything is free.  And so there’s nothing to ensure that respondents’ answers match their actual behaviors. Expensive things like wider seats have just as much weight as free peanuts.

To show what I mean, let’s play this out.  I call Delta and somehow find the magical IVR prompts to reach the right person. She hears my plea and responds, “My goodness – you’re right!  We’ve been looking at this wrong! We’ll fix that immediately.”  So they remove one chair from each row to allow for wider seats.  What will happen? Will travelers flock to Delta to take advantage of the space?

Well, probably not. To make up the revenue shortfall, Delta now needs to charge 20% more.  Rather than driving more business, they chase away more customers than they attract.*

Asking customers to rate importance feels intuitively like a good idea. And sometimes it’s your only option.  But there are usually alternatives. By analyzing stated  importance, you don’t get a balanced viewpoint. It’s better to forego asking importance altogether and instead use business metrics to derive importance.  Don’t ask how important things are, bring in sales or usage data and use that to find out what actually drives your customer loyalty.

If you’re Delta, link satisfaction scores with revenue, and use this to derive true importance.  If increases in satisfaction with seat width correlates with increased spending and loyalty, then you start tearing out seats – at least as a test. But, more likely you will find that increases in seat-size-satisfaction have little link to loyalty. When you see that, you know that you need to focus elsewhere. If revenue isn’t available, then at least link it to your high-level relationship scores, such as NPS or satisfaction, then find what factors drive these scores.

This isn’t just a B2C issue. In fact, I find it even easier to do this analysis for B2B customer measurements.

In my review of the 2012 Temkin Award winners I found that most of the winners incorporate business metrics with their customer experience measurements. For example, JetBlue has linked on-time departure to NPS. By understanding this linkage, they were able to better allocate staff at airport. And while Oracle did not give any detail, they apparently link drivers not only to their NPS scores, but also to financial results.

Unfortunately, most of us are using the easy and intuitive approach of assuming that ranked importance matches behavior. And it often isn’t true.



Your turn.  Are you taking the time to determine whether your metrics actually matter? Or are you telling your teams to create wider seats?

—————-

*No, I’m not saying that a company cannot be successful with a premium offering.  Simply that this offer does not match Delta’s value proposition, so will be unsuccessful for them.

12 COMMENTS

  1. Completely agree that stated importance, a questionnaire approach and analytical technique that equates to ratings of expectation, has long since outlived its usefulness. However, derived importance, usually determined through correlation analysis or simple regression against a dependent variable like overall satisfaction or future purchase likelihood, also has challenges when endeavoring to determine what really drives customer loyalty behavior. This is what I’ve sometimes referred to as the “Venus Flytrap of Logical Fallacy”: http://customerthink.com/correlation_is_not_causation_big_data_challenges_and_related_truths_that_will_impact_business_s/

    Would suggest a more actionable approach, such as discriminant function, or swing voter analysis, such as shown in one of my CustomerThink blogs from a few years ago: http://customerthink.com/corporate_reputation_and_advocacy_linkage/

  2. JIm

    ON DATA, CAUSALITY AND INNOVATION

    An interesting post. And an interesting response from Michael.

    How do you use business metrics to rate things that have never been done before, particularly significant innovations like new services or new business models?

    Even if you could break down business metrics to the level of granularity that would allow a statistical association to be established (good luck with that!), and even if you had detailed structural models that show causality (as common as snow balls in a furnace!), neither will help you derive the importance of something that has never been done before and for which, by definition, there can be no useful data.

    Graham Hill
    @grahamhill

  3. Graham,

    Well, I did spend my Friday night shoveling, so maybe that says something?

    The types of research I’m focused on answer different questions than you are posing. Not that yours aren’t important! It’s more about where are the opportunities. The area I’m talking about is the stage before your innovation – finding out where to focus for your customers.

    You’re right that, while this can suggest places to innovate, it doesn’t really tell you whether your new business model will be successful – only whether your existing business model works.

  4. Per Graham, quantitative research for concept/positioning development and refinement has its own special set of analytical challenges, and is not often applied. That said, I’ve used both SEM (Beyond Philosophy has been doing this for some time) and DFA to aid in sculpting product and service concepts, positions, and communications platforms.

  5. Hi Jim

    Thanks for your thoughtful reply. Unfortunately, it has not assuaged by concerns. Let me explain.

    If you look back at Ted Levitt’s prescient HBR article on ‘Marketing Myopia’ you will see that he proposes an outside-in, three-step process for going to market. First, find out what customers want. Second, work out how to give it to them profitably. And third, tell them all about it. It is quite simple and it works very well.

    As companies (and their growing number of competitors) built catalogues of products they became more reluctant to use Levitt’s process due to the time required and its cost. Instead, they preferred their own inside-out, three-step-process. First, Adjust the products they have to take account of competitors activities. Second, find customers who you could potentially market them to. And third, tell them all about it. This inside-out approach is taken by the vast majority of large corporations today.

    The approach typically only comes up with incremental, me-too innovations. The incrementalism is exacerbated by the devilish triumvirate of short-termism, P&L focus and siloed product management that have established themselves in large corporations. As a consequence, real innovation, the sort that creates radically new products has largely been abandoned to smaller competitors and particularly, to brand new competitors.

    Back to Delta and their seat width problem. I think it is safe to say that Delta hasn’t produced any noteworthy innovations at all in the past 30 years. American created the frequent flyer programmes that are now more valuable than the airlines themselves. Southwest created the low-cost carrier model that has democratised air travel. And Virgin Atlantic has pioneered the servitisation of travelling in style for upper-class passengers.

    By only looking inward at itself and its rather lifeless US FSC peers, Delta can only use its business data to tell itself about the most incremental of changes. It couldn’t use it to, e.g. innovate new things like FFPs, the LCC model, or servitisation. All these new things required out of the box thinking, radical innovation and completely different business models.

    Legendary entrepreneur Steve Blank talks about ‘there being no facts in the building’. By that he means you have to get out and talk to real, live customers to see what they really want, before you can create the new products they all want but that nobody has thought up before. Gazing longingly at mountains of inside-out data is not going to give you the next generation of industry disrupting products.

    So next time someone badgers you about seat width at a dinner party, listen to what they really want and think about how you could give it to them profitably before turning to your (t)rusty spreadsheets. Delta could certainly do with your help.

    Graham Hill
    @grahamhill

  6. Graham,

    I think we’re answering different questions. My post was not on how to disrupt the market, whether innovation is good, or how to conduct innovation research.

    Rather, it’s that using rational methods (Simply ask people what they want, then do that) is fraught with failure. While this certainly applies to disruptive innovation research, that wasn’t my point.

    Rather, I’m looking at that more incremental research, where most of the activity comes.

    While I’m not familiar with Levitt’s article, from your description of it we’re quite aligned. It’s in this first section “Ask customers what they want,” that I’m speaking. Rather than simply asking what they want, bring in behavioral data to verify that what they’re asking for is truly impactful.

    I hope nothing in my post says you should only look at spreadsheets, and never talk to customers! Simply that you have to be more methodical. And most surveys I see still use a rational approach of just asking customers what they want.
    Jim

  7. Agree with Jim. Getting customer feedback is a good thing, but just blindly acting on every request is a recipe for disaster.

    For years Southwest passengers have been asking for assigned seats. But Southwest doesn’t provide because it would increase plane turnaround time, which ultimately would increase costs and fares.

    So in this case giving customers what they want would be one step toward encouraging them to stop being a customer! Because the core value that Southwest customers want is low cost.

    Consider Walmart’s “$1.85 billion mistake,” what some experts say the retailing giant lost in an ill-conceived “Project Impact” to reduce inventory. Customers said “reduce the clutter in the store.” After spending $millions to do so, Walmart actually decreased same-store sales. Clutter was an annoyance, for sure, but reducing inventory undermined Walmart’s brand promise of a broad selection.

    Also agree with Graham that it’s tough to research truly new/innovative ideas and know in advance what will work. But I don’t think Jim’s example qualifies as real innovation: increasing seats –> fewer seats –> higher cost per passenger –> higher fares. There’s already a solution for that — it’s called Business or First Class!

  8. Hi Jim, Bob

    UNDERSTANDING CUSTOMER JOBS DRIVES PROFITABLE INNOVATION

    I think you can have your cake and eat it; both the small incremental slices and the whole disruptive cake if you want.

    Nowhere in my comments did I say that you should slavishly pander to customers expressed wishes. As you both point out, that is a recipe for high-costs, but not necessarily for higher profits. On the other hand, simply ignoring customers will not get you very far either. Henry Ford infamously said that customers could have his cars in any colour they wanted, providing it was black. Alfred Sloan at GM new better and offered his customers not only a larger range of colours but also a larger range of cars. The rest is history. Ford Motor Co never recovered from Henry Ford’s manufacturing myopia.

    Henry Ford was disdainful of listening to customers at all. He also said that if he had asked customers what they had wanted they would have asked for a faster horse. Ford was not only myopic but also ignorant. He didn’t recognise the distinction between asking customers what they want (Step 1 in Levitt’s process) and deciding how to give it to them profitably (Step 2). All innovators struggle with this innovation dilemma, sadly most of them get it wrong; evidence suggests that 80% of new to market innovations fail on entry and 60% fail on re-entry.

    Independently of each other, Prof Clay Christensen at Harvard and Tony Ulwick at Strategyn both identified that asking customer about the ‘jobs’ they were trying to do and the ‘outcomes’ they wanted from doing them, could be used to resolve the innovation dilemma. Using a mixture of qualitative ethnographic research (commonly used by the design profession) and quantitative validatory analysis, innovators can create a catalogue of jobs customers are trying to do, segment customers by job and use the jobs as the creative catalyst to innovate new products, services, experiences, even capabilities and business models. It is the company’s job to innovate new solutions to help customers do their jobs better, not the customers’.

    This jobs-based approach has been around for almost 10 years now and studies by Strategyn have shown that it has an average 80% SUCCESS rate. That is a lot better than the 80% FAILURE rate from traditional innovation. And there is even more potential for improvement. A paper by Goldenburg et al on ‘The Idea Itself and the Circumstances of its Emergence for New Product Success’ shows that up to a 95% success rate in innovation can be achieved by responding really fast to customer needs. The sky really is the limit for jobs-based innovation.

    Back to Delta. Perhaps if Delta spent a little more time understanding the functional, emotional and social jobs its customers were trying to do, it might do a better job of creating incremental innovations that meet their immediate needs and still grow Delta’s profitability. It doesn’t have to be a zero sum game played out as the ‘passenger’s dilemma’.

    You can have your innovation cake and eat it if you start with understanding customers and their jobs, before you reach for your spreadsheet.

    Graham Hill
    @grahamhill

  9. Graham’s point re. “understanding customers and their jobs”, as a platform for innovative products, services, communication, and/or positioning is more important today than ever. We interpret ‘jobs’, as the customer journey, the moment-by-moment travel through the entire customer experience. In the days of TQM, this was Quality Function Deployment, where employees and customers participated in innovation, both separately and combined. Now, we see this through devices such as market research online communities and ‘moment mapping’ ( http://www.beyondphilosophy.com/services/moment-journey-mapping/). These would be effective approaches for companies like Delta to follow if they are to realize success from their seating space modification initiatives.

  10. I’m not sure I agree that “jobs” are the customers journey through the customer experience. Jobs are the “what” and to me, the experience has solution “who and how” built into it. As Graham points out, 80% fail to succeed with new innovations and quite often it’s because the focal point are things like “ideas” or “experience” when they should be the job a customer is trying to get done (functional, emotional & social). We can design the experience later once we have our facts.

    First we have to understand the job, and *all* of the needs used to measure its success. That’s quite different than asking a customer what they need; they don’t really know. This is not voice of the customer or NPS. There is a scientific method here that’s worth exploring and Graham is far more eloquent in the way he describes and defends it, so I’ll leave that to him.

    BTW…

    I’m 6′-4″ tall. Just being that tall means I’m probably wider than someone shorter than me. I can tell you I DO WANT WIDER SEATS AND MORE LEG ROOM! And more head room in the lavatory!!!!

    LOL

    Graham is dead on and just gave laid a trail of bread crumbs to the truth.

  11. I see two competing sides this interesting issue. Maybe more, if I thought about it long enough.

    On the one hand, Henry Ford is maligned for his ‘tin ear’ for customer wants, needs, and preferences, saying that if he had asked customers what they had wanted they would have asked for a faster horse. (Thanks, Graham). OK, that comes across as insensitive and callous.

    On the other hand, as Graham writes, “It is the company’s job to innovate new solutions to help customers do their jobs better, not the customers’.”

    These ideas are not diametrically opposed, but they aren’t in alignment either. The question that hasn’t been answered in this interesting discussion is, can the two be reconciled in the path toward useful innovation?

    By the way, as one who is under six feet tall, I can truly say that the only time I actually gloat about that physical characteristic is when I fly coach.

  12. The answer – to Andrew’s question – can be found in collaboration, or partnering, between customer and supplier (http://customerthink.com/customer_partnering_proactive_winning_cem/)

    In the TQ days, this was achieved through Quality Function Deployment. Today, it is most often accomplished through innovation (in product or service development, marketing and communication concepts, etc.) with the tool of dual customer journey mapping, or as we define it, moment mapping.

LEAVE A REPLY

Please enter your comment!
Please enter your name here