Product innovations can blow up. They can blow up big, or they can blow up bad. Sometimes it is hard to tell which is which in the end – sort of like every time Kanye West tweets something.
Unlike Kanye’s publicist, who just has to hold their breath with every keystroke, Product Managers can predict and prevent the bad kind of blow up by listening to the “Voice of the Customer” (or “VoC” for short.)
In my last post on “How to Communicate Product Changes to Your Users,” I called out a couple of big cases of user backlash, including Spotify and Couchsurfing. I also called out the would-be backlashes that were turned into big wins by Airbnb and Tinder. The latter two listened to and monitored their community of users closely, really understanding what their users thought and felt about the product’s evolution. The former either underestimated their users or undervalued them. This time, I will run down what a Product Manager could learn from three very big blow-ups that could have been prevented by leveraging a VoC program that prioritizes listening.
What is Voice of Customer?
Traditionally, VoC is customer feedback usually gathered through a Voice of the Customer Survey, where customers provided feedback about their experiences with and expectations for a product or service.
Modern VoC is much more than that. Yes, surveys have moved from traditional focus groups, paper and pen to integrated widgets right in your app with forum features that I hear are just fantastic, but your user “voice” comes from many outposts: Engaging in forums (in-person and online), listening to feedback during beta tests, observing actual user behavior, running social media tracking and sentiment analysis, staying close to customer support & success to track and collect insights, running user tests… *takes a breath*… Now, that is a lot, but a holistic VoC program will not only prevent terrible product blow-ups that could trigger your company’s downfall, it can turn those launches into big wins.
Despite the push toward VoC, companies keep making the same mistake: they don’t listen enough to their users, or they listen to the wrong things. In light of the recent user backlash over big changes to popular multiplayer game Clash of Clans, I thought I would do a post-mortem on some of the biggest.
For each, we will look at what happened, the product manager’s perspective, and how a VoC Program could have helped.
Product Blow-Ups That a VoC Program Could Have Prevented
Uber’s Surge Pricing – New Year’s Eve & Hurricane Sandy
Uber – oddly also a bit like Kanye – seems to not only cultivate negative publicity, but often seems to come out stronger after each backlash. One of the first of these came during New Year’s Eve 2012, quickly followed by Hurricane Sandy.
In 2010/11, Uber introduced a feature that would increase the price of a trip during times when demand for trips would outstrip supply. The company called this “Surge Pricing” and labeled it as a means to encourage more drivers into surge areas to meet demand. In 2011, before a price surge, Uber would send a mass email to users, make a blog post, then send a notification through the app if surge pricing was in effect.
On NYE 2012, however, the demand was so heavy in several areas that some complaining users stated they paid over $100 for a 1.5 mile ride. Even if they had seen the notifications, they still used the app. Others did not see it or notice it for technical or other reasons, and most underestimated how much it would cost. Uber claimed only 97 complaints were logged on the night, but it caused a social media and PR backlash, with plenty of word of mouth about Uber, for better or worse.
Sample tweets from that night:
Wow… $135 to go 12 blocks on @uber last night. 3.75x surge pricing alert needs to have flashing red light on it! Really guys? @uber_sf
— Dan Whaley (@dwhly) January 1, 2012
While I’m glad I’m home safely, the $107 charge for my @Uber to drive 1.5 miles last night seems insanely excessive. 🙁
— Aubrey Sabala (@Aubs) January 1, 2012
Later that year, during Hurricane Sandy, this happened again, but Uber actually switched off Surge Pricing on the user side during the backlash and absorbed the extra pay to the drivers, costing an estimated $100,000 on the day, as well as even more negative publicity.
Some users reportedly do not necessarily trust Uber’s motives in how it uses surge pricing now, even if they understand the need it is supposed to be filling.
Let’s take on the hypothetical perspective of a Product Manager at Uber tasked with considering Surge Pricing back before this all started. It probably made a lot of sense. Dynamic pricing in travel has been around for a long time to balance supply and demand. With all of the benefits Uber provides, the main pain point that Uber has successfully addressed is the need for an accessible ride within minutes, even during peak traffic – a need which was not, in the user’s view, being reliably filled by cab companies. Surge Pricing helps satisfy that customer need.
We know that our users are willing to pay the surge amount on a busy night like New Year’s Eve to get home safe. We know that 3.7x Surge Pricing will attract more drivers onto the road to meet that demand. But if we were talking to a handful of users or testing behavior and response in a handful of small markets, our otherwise all-star PM work might still miss the risk for such a large backlash, even one that looks obvious with 20/20 hindsight.
How a VoC Program Could Have Helped
The best evidence of the benefits of a VoC program is a blog post by one of Uber’s devoted users who was shocked by that NYE Surge Pricing. Their recommended redesign mockup of how Uber’s Surge Pricing notification UX should work seems eerily close to what Uber actually changed to mitigate their issues.
The user suggested:
- The Surge multiplier become much more prominent than before with less distraction.
- The explanations are more to the point and less cluttered.
- The user takes an action that clearly confirms that they understand the pricing.
- (Not on the mockup) The blog also mentioned the user should see the final pricing, not just the multiplier – which you can see in the new implemented design on a per-mile basis along with a minimum fare.
These are all minor nuances that a closer touch with the users could have uncovered, avoiding one of many very large and costly backlashes.
User’s Suggested Redesign:
Uber’s Eventual Redesign:
If Uber had engaged with users beforehand using a more thorough, user-focused VoC program, they would have identified that even users who understood the basic reasoning behind Surge Pricing would experience emotional triggers, surprise, and potentially even feelings of mistrust at unexpectedly high fares in their moment of need.
2. JC Penney’s No-More-Sales Snafu
Back in 2011, JC Penney famously hired the creator of the famous Apple Store experience, Ron Johnson, as their CEO. Mr. Johnson set out to end the “fake prices” set out by constant markdowns, essentially ending the term “Sale” at the store. It sounded revolutionary, but it ended up being a large case of company-customer disconnect.
JC Penney spoke boldly about the plan, had a sputtering start in advertising the new idea in-store (mainly just not having “On Sale” signs), and eventually lost over half a billion dollars in the 4th quarter of 2012. Ron Johnson and the JC Penny President were both ousted. Mr. Johnson only lasted 17 months as CEO.
A Product Manager’s Perspective
If a Product Manager can learn just one thing from Ron Johnson’s stumble at JC Penney, it is this:
What worked for your last product will NOT necessarily work for your next one.
There is no silver bullet other than understanding your users.
Mr. Johnson was brought in from Apple with a mission to deliver what he had at Apple in the JC Penney stores. As a PM, you know how much internal selling you have to do to get things done, with both compromise and improvement as a result. Mr. Johnson and his team did a lot to sell a change in brand to the press and to the internal corporation itself, but it forgot to really check in with its customers.
How a VoC Program Could Have Helped
JC Penney was underperforming its category leading up to Mr. Johnson’s appointment, and in desperation, didn’t take the time to truly engage with customers and identify nuances that could make its product reboot a failure. (This is also a good lesson on why doing a “product pre-mortem” is key. We’ll be writing more on that in the coming weeks.) The outcome was a big risk and a big tank.
If the company had invested even a little more in understanding the consumer, they would have listened to customers by testing questions about their consumers’ perception of sales and “luxury brands” using traditional VoC methods such as consumer surveys, customer interviews and focus groups, and even collecting consumer feedback at the point of sale verbally or with devices.
What they ended up doing by jumping right in is testing a different, riskier question: “Is JC Penney an aspirational brand *right now?*” They got their answer.
Understanding your users may not always give you detailed solutions, but they will certainly help you ask the right questions.
3. Netflix’s Price-Hike
2011 & 2012 were rough years for product backlash, it seems. Industry-darling Netflix could not miss out.
In 2011, Netflix CEO Reed Hastings planned to split the $10 per month subscription, which included DVD rentals and unlimited on-demand video streaming, into a $7.99 on-demand streaming subscription and a $7.99 DVD subscription. The company predicted some backlash at first, but assumed it would cool.
The company eventually lost 800,000 subscribers and 77% of its stock valuation in 4 months, as well as distribution rights to Disney and Sony Pictures films. You read correctly – nearly a million subscribers and three quarters of its market cap gone.
Several internal sources at Netflix allegedly mentioned that communication was low at every level. They suggested that there was not a lot of discussion around the decision to split and spin off the DVD business that had been Netflix’s cash cow for a decade.
Even after the initial announcement created backlash from subscribers, the DVD spin-off business was launched and taken down after only 3 weeks.
Netflix has obviously recovered and maintained Reed’s goal of having the most streaming content and subscribers, and you can still rent DVDs today (Baby Boomers and Gen Xers would not just make a jump like DVD to streaming overnight), but this could have all gone down far more smoothly.
A Product Manager’s Perspective
Price adjustments, focusing on future growth, and sending old products to pasture are all part of a balancing act. In this case, Netflix’s decision to unbundle and increase prices to focus on the growing streaming market with an improved product was the correct strategy – only with poor execution. There is an ability to underestimate the time it will take for historical user behaviors to fall away. These often come with emotional ties as well. Netflix has since been able to keep the old while growing the new. If they had listened to users, perhaps they would have not ripped the band-aid off so harshly.
How a Voice of Customer Program Could Have Helped
In this case, the subscriber’s voice was heard everywhere in the company except in the C-suite. Reed Hastings had been praised by the press and shareholders and that sort of trust can create a bubble. This can create a stable vision, but can also lead to missteps like this one. Even the most steadfast product or company executive must have his or her pulse on the user, especially when affecting something so ingrained in their habits. A healthy VoC program would either put C-suite execs directly in front of customers or in front of a Product Manager who is a trusted voice of the customer in the organization. That trusted voice would be someone who shares VoC data from surveys, focus groups, etc., with the C-suite regularly before changes are made from on high.
Thankfully, Netflix seems to have learned its lesson: It’s 2014 price raise actually gained a positive response from customers and an up-tick (rather than crash) in its stock price.
We have learned that:
- You cannot assume that your users will understand why a feature is needed.
- Unengaged users may not always trust your motives even if you explain your changes.
- Things that worked on your previous products will not necessarily translate to new products or users.
- You should not get stuck in a bubble separating you from your users, just because of recent success or praise.
In the end, this all comes down to one thing: Listen to your users at all levels. More than just one-on-one. More than just from support complaints. Actively engage them. Ask for feedback on future products.
Of course, users can’t always predict their own habits. Occasionally, there will be ideas that users claim to loathe in theory that they may end up loving in practice. (For example, Facebook’s News Feed is now a core need-to-have feature among users, but created one of the major user backlashes in recent memory, with 100,000 of their then <10 million users joining a group condemning the feature when it was first launched.) In the end, though, a healthy VoC program can help you listen well and read signals before even seemingly positive changes blow up in your face.