11 February 2011

1 + 1 = 2? Nokia and Microsoft Better Hope Not

There's a lot of topical interest in the decision by Nokia and Microsoft to form a partnership around Windows Phone 7.

From a UK perspective, quite a lot of attention has been given to the market shares, as reported by comScore (original Data Gem here). I've reproduced their chart below (my estimates of their numbers, reading quite carefully from the graph).


Obviously, it is anything but automatic that when two companies combine the result will resemble a simple sum of their pre-combined market shares. But health warnings apart, it's interesting to see what happens if you do just add them.


Unsurprisingly, it doesn't look any better.

Is the Nokia-Microsoft partnership a good move? Who knows? From my perspective, it looks unambiguously positive for Microsoft, but far less clear for Nokia. I think Stephen Elop's memo suggests he has diagnosed the problems Nokia faces rather accurately; whether embracing Microsoft is a way out is much more open question. It will be very sad, both for Finland and for Europe, if Nokia dies. As a long-time Nokia fan, it took the iPhone to persuade me to buy a non-Nokia phone, and I'll personally be very sad if the company does fail. But I have to say, embracing Microsoft is not a way to win me back.

08 February 2011

Demand Generation: the Salad Shooter (Bloom County)

Back in the day (a mere 22 years ago, it seems) The Guardian ran the "holy trinity" of cartoon strips daily—"If...", by Steve Bell, Doonesbury, by G. B. Trudeau, and Bloom County, by Berkely Breathed. (Remarkably, two of the three featured penguins among their lead characters.)

Sadly, the Guardian stopped running Bloom County many years ago (long before the paper's evil deputy editor, Ian Katz, even tried to remove Doonesbury), and I assumed that it had ceased to exist. But there was a pair of strips from Bloom County that I particularly cherished, and have frequently described in the two decades since they ran. This morning, through the power of the internet, I found them. Enjoy (in order), The Salad Shooter Sagas:

Salad Shooter 1

Salad Shooter 2

Even more wonderfully, it transpires that Bloom County still exists, and can be mailed to your inbox daily.


01 February 2011

How to Lose a Customer (Part 1)

Regular readers of this blog will know that the primary focus of The Scientific Marketer is customer analysis as it pertains to marketing. But one of the major themes from that analysis is that marketing surprisingly frequently causes negative effects—sometimes, so negative that it results in the loss of a customer. This reminds us that, valuable though the role of analysis can be, it is ultimately actions that count.

Customer service is hard to get right, but the sheer hamfistedness of much that passes for customer service is breathtaking; this story simply recounts my experience with one company that called me this morning, and an attempt to analyse it from the perspective of customer experience.

The Call

This morning, my work was interrupted by a call from a company I placed an order with on 25th January last year (2010). I’ll call the company NorthSouth. A pleasant woman said she was calling because my company, Stochastic Solutions, has an account with NorthSouth and they noticed that I hadn’t used it recently; she wanted to know if there was a reason.

I recognized the company name and knew I had placed exactly one order with NorthSouth. I wasn’t sure what the order had been for, but did remember that I had paid a reasonable amount for next-day delivery, because (funnily enough) I needed it the next day. I also remembered that they had failed to deliver on time. After the call, I looked it up, and the item was a Mac/PC-compatible 2GB 256-bit AES encrypted flash drive made by Kingston (the Data Traveler Vault Privacy Edition). Looking at my diary, it becomes clear that the reason I needed it was that on 27th January I went to a client site, and the security policy I was working under required any drive I used to be encrypted. I now recall that the lack of a suitable encrypted drive was, as expected, a major problem, and caused a lot of time to be lost. In the event, I called up to cancel the order, since the immediate need had passed, but NorthSouth refused and merely agreed to refund the next-day delivery charge. In a further amusing twist, I happened to be preparing paperwork for my accountant the other day and noticed that NorthSouth had actually taken over six months to refund the next-day shipping charge.

Speaking to the person who called this morning, however, I didn’t remember all that. My only recollections were (1) I ordered something I needed with next-day delivery (2) they had failed to deliver on time (3) this had caused me a problem. So my response to “is there a reason?” was along the lines of “Yes. I ordered something from you with next day shipping and you failed to deliver it on time.”

The woman I was talking to sounded surprised, and slightly off balance, but came back with “Won’t you give us a chance to prove we can do better than that?” I declined and she said she’d make some notes on the account so that I don’t keep getting calls.


The point of this story isn’t to excoriate NorthSouth for their appalling customer service; I have only one data point, and I certainly don’t want to read too much into a single incident. But I think it’s still possible to learn a number of things from this one incident.

My candidate list would be these:

  1. Restoration. Mostly, when people (especially businesses) pay for next-day delivery (in this case, increasing the effective price of the item by a third) they don’t do so for the hell of it: they do it either because they need it or because they are impatient. Simply refunding the next-day charge, while clearly the minimum that the supplier can get away with when it fails to supply on time, is unlikely to leave the customer satisfied.
  2. The Primacy and the Recency Effect. It is a well known principle of presentations that the most important two parts are the start of the presentation, when you either succeed or fail in hooking your audience (the primacy effect), and the end of the presentation, when you have the opportunity to offer a take-away as their last impression (the recency effect). I tend to argue that customers have long memories, particularly for unexpectedly poor service, but also for unexpectedly good service. But I think the primacy and recency effect are relevant in customer experience too. In this case, my first impression was caused by the failure to deliver on time. My most recent impression (before the call) was that they took a remarkably long time even to provide the inadequate remedy that they offered. In this case, there was probably no reasonable way of winning me back as a customer after screwing up the first order and then handling the aftermath poorly.
  3. If you’re going to operate a win-back/reactivation programme, at least look up the customer first. To its credit, NorthSouth clearly has a customer win-back programme, perhaps triggered a year after an order. The woman who called clearly knew at least a tiny-bit about me in the sense that she knew I’d ordered on the internet. This suggests that either she looked up my customer record or (perhaps more likely) was given a list that included at least some information about the order history. But what is equally clear is that she didn’t have all the information that the company has (or should have). Almost certainly, the company has some kind of record of the fact that it screwed up, even if it’s only the credit to the account that was made when they (eventually) refunded the next-day shipping charge. The woman who called could have been much better forearmed or could have saved herself the effort of making the call.

What Could Have Happened?

As I say, it’s not clear that, in this case, there’s anything economically worthwhile that the company could have done to win back my business; I think you have to accept that once you’ve screwed up the first order, you’ve probably lost the customer unless you are in some kind of monopoly situation or have such compelling differentiation that the customer might plausibly come back anyway. (The bucket airlines tend to benefit from at least one, and sometimes both of these effects.)

But I will offer a couple of possibilities:

  1. Overcompensate. There are lots of reasons for (from the vendors perspective) over-compensating customers when you screw up. I would include in these

    • it emphasizes to the customer that you take the problem seriously;
    • it shows that your claim to ‘care’ extends to being willing to make a financial sacrifice;
    • perhaps most importantly, it creates a clear economic incentive within the company not to screw up;
    • you potentially exceed ths customer’s expectation, thus replacing a negative perception with a more recent positive one. In fact, there’s quite a lot of anecdotal evidence that an exceptionally good response to a mistake can lead to a better customer perception than getting it right in the first place.
  2. Whole Customer View. It’s not exactly a radical new insight, but the more connected is your view of the customer, the easier it is to interact appropriately. In this case, if the woman who called had known that NorthSouth failed to deliver on time, she would certainly have been in a better position to discuss reactivation.

  3. First purchase. As noted above, the first interaction is particularly important, and if you screw it up, I guess you’ll rarely be given a second chance. If you want to try to undo the damage of a broken first interaction, you’re going to have to work particularly hard and give the customer some plausible reasons to believe that the situation was atypical.

Labels: , , ,