The Elusive “Effortless Experience”

“The role of customer service is to mitigate disloyalty by reducing customer effort.” – The Effortless Experience

It’s been 3 years since the publishing of “The Effortless Experience,” a book chock-full of research confirming what customers have long found obvious but that too many vendors failed to understand: the path of least resistance is best. Why, then, do so many support organizations continue to make things so difficult on their customer? More importantly, how can your support organization actually deliver a low-effort experience?

Just in the past month, my own personal support experiences have included:

• Google.com providing better search results than my vendor’s own website.
• Navigating through IVR hell, only to repeat the information to a live agent.
• A third-party forum offering a more accurate answer than my vendor’s own knowledgebase.
• An email support agent telling me to call a phone number for “technical support” because she only handles “billing issues,” despite me selecting “technical” as the nature of my problem and choosing email as my preferred channel.
• Being forced to pick up the phone, wait on hold, and spend 15 minutes explaining the issue to a live agent simply so they could access a document that should have been published to self-service.

One recent experience frustrated me beyond belief. Shortly after signing up as a premium member with a social networking service for cyclists, I experienced a bug with one of the very features that required becoming a paying member. Their self-service knowledgebase wasn’t helpful, so I submitted a case online. An automated email explained that it could take up to 7 days before an answer would be provided. Unless I’m a premium member, at which point I should reply to this email with the phrase, “this ticket is resolved,” and then copy/paste my original submission into a duplicate ticket on a separate “premium support” form. I couldn’t believe such a process actually existed instead of automatically recognizing my premium status and proactively routing accordingly. Ironically, the vendor never did solve my problem although I’m sure their “First Call Resolution (FCR)” metrics are quite high due to this process of resolving unresolved cases. Ironically, I quickly found the answer on a third party blog and then canceled my premium service out of principle.

So how can you actually deliver an “Effortless Experience” to your customers? 

“The Effortless Experience” authors identified four pillars of low-effort service. Three of them are predominantly related to assisted service, which I consider to be mitigation efforts from failing to deliver on the most important pillar: channel stickiness. Your customer almost always prefers web self-service just as they’d rather use the ATM instead of the bank teller or the airport self-check-in kiosk over the long lines to speak with an agent.  If they claim to want assisted service first, that’s usually because your self-service site is not useful to them. As such, I advise my clients to focus first on self-service, making that channel as “sticky” as possible.  Below are five steps to do exactly that.

“The Effortless Self-Service Experience.”

Step 1: Know Your Customer

Your customer knows what products they own, their support tier and SLAs, the information they’ve already provided you, and the value of the information you have provided back to them. They probably even recall their last experience with your company, especially if it was a bad one. You need to know all of this and more.

If a “360-degree view of the customer” sounds too daunting, just start with offering information in context of the products your customer owns and the environment they are using it in. Showing a knowledge article about the wrong product or even the right product on the wrong operating systems is one of the most common failures I see in knowledge management systems that deliver answers purely based on the occurrence or proximity of keywords instead of a complete contextual match.

You should also consider your customer’s previous support cases and determine if any of those cases remain open or could be related to the current issue.

Step 2: If you have the answer, make it available to self-service

Countless surveys by the Technology Services Industry Association (TSIA) show that the knowledgebase is not the only source of knowledge but one of many sources. Many times the KB isn’t even the most useful or heavily used resource indicated in these surveys. So why do so many companies limit self-service to only this one repository?

Search engine technology has existed for decades to unify additional repositories like peer-to-peer support communities, training, product documentation, and even known issues and bug databases into a single self-service portal. Unfortunately, some companies are still unwilling to make an investment in such an important technology or are simply afraid or even too prideful to share all of their knowledge.

[tweet_box design=”default” float=”none”]Don’t force your customers to channel shift, because they might just shift to your competitor instead.[/tweet_box]

Step 3: Don’t Be Selfish

I eluded to this in the previous step, but it bears repeating: your knowledgebase doesn’t contain all of the answers. Vendors need to leverage the wisdom of the crowd and use a peer-to-peer community forum. Your support employees are obvious moderators of those communities, but don’t forget to grant those same privileges to your customers or partners who often times know your product better than you do!

Third party sites not owned or even controlled by your company should also be fair game. Almost a decade ago I helped Research in Motion index third-party sites like blackberryforums.com to supplement their own support assets. This was pretty groundbreaking at the time but should be part of any support leader’s strategy these days to ensure that the vendor’s website is the starting point for their customer’s diagnosis process even if the resolution could point to a third-party source.

[tweet_box design=”default” float=”none”]Don’t be selfish. Use #IntelligentSearch to offer more than just a #knowledgebase. #KM[/tweet_box]

Step 4: Get better over time

Until recently, the phone center had one huge advantage over self-service: human agents could learn over time and make up for gaps in the content or even facilitate workarounds to poorly designed support processes. But thanks to recent advances in machine learning technologies and the elastic computing power of the cloud, self-service can be as good or even better at adapting and automatically improving over time with algorithms analyzing so many more data points than one individual call center agent could possibly consider.

My client Logitech told a great example of this at a recent industry event. Their Coveo search analytics were constantly reporting what looked like random alphanumeric phrases being searched on. Nobody appeared to understand what this was. But Coveo Reveal, a new machine learning technology recently deployed by Logitech, automatically noticed that a select group of people eventually modified their search from this seemingly random string to an actual Logitech product ID, at which point they quickly solved their issue. Coveo Reveal started to automatically suggest the product-specific answers when people queried the strange phrase despite nothing in the content suggested even containing that phrase. It turns out users were entering a SKU imprinted by a third-party manufacturer on the Bluetooth dongle instead of the actual Logitech mouse or keyboard product number. Coveo Reveal was smart enough to put two and two together on this and many other similar examples to provide via self-service what no call center could ever achieve.

Step 5: Avoid channel shifting at all costs, but make it easy when absolutely necessary

How do you know if an issue truly required the assistance of a live agent?  Have your agents ask.

Fidelity Investments is a great example, always asking me if I tried to solve my issue online before calling in, and always sending me a link to a document I can view on self-service for future reference. They do this very politely, not trying to imply that they didn’t want me to call in, but making sure I know that I didn’t have to so that I gain confidence in solving this and other issues on my own in the future.

If my transaction can’t be completed online, the Fidelity phone rep will often ask me if I would be comfortable doing so if that feature did exist on self-service. According to a few sources inside the organization – they actually prioritize self-service feature deployments based partly on this survey data. This type of feature backlog management isn’t that different than how Knowledge Centered Support (KCS) organizations prioritize the publishing of content to self-service based on closed case linkages.

Take all the information you know about the customer from “Step 1” and pre-fill their “Create Case” form as best you can. Since people always enter more information in a “form” than they do a search box, leverage that additional context plus what you’ve learned in their self-service session to create a more concise and precise set of search results in your final deflection attempt before creating that case. And if you do create a case, pass all of the context from that self-service session to your agent, including the searches performed, documents viewed, etc. Perhaps you even route the case differently based on the customer’s contract or SLAs. Simply put, make sure your customer is shifting channels, not starting over.

Will Support Ever Be Effortless?

As frustrated as I become with certain support experiences I encounter that are anything but effortless, more and more organizations are learning how to do low effort support correctly. Most that do are experiencing a tremendous return on investment.

I think the next breakthrough on this theme is “Self-Solving via Self-Service” – that is, automatically enacting the resolution instead of simply displaying instructions on how to do so. I’ve worked on some really cool projects along these lines and will cover this topic in an upcoming blog post.