Stripe is an online financial services company, focused on providing startups and growing companies with an easy-to-use merchant API. I have previously recommended and used Stripe for professional, client, and personal projects of various shapes and sizes. It is not a revolutionary company, there have been other online gateways and similar solutions available for years before Stripe became popular. It is, however, radically easier to use than the alternatives and it is an impressively polished developer tool. Everything from the marketing pages to the dashboard is painstaking considered.
While Stripe’s dashboards, and websites, and Dribbble pages are filled with pixel-perfect screenshots, design can’t stop there. Most companies these days are service companies, which means that you’re not just selling a product — your income is based on continued usage where churn is your enemy. While providing a more steady stream of income, services companies need customer support and service become much more important. With services companies, support is something that needs to be designed, just as much as their marketing site does.
Criticizing Stripe publicly
My usage of Stripe was pretty basic. Using the Stripe integration on Memberful, I set up a subscription payment option on a blog for a client. With $3.99/monthly payments, she was able to start receiving small subscription fees for her posts from her intensely engaged readers. Payments were working, depositing a moderate amount of money each week until I got a text from her a few weeks back. She noted that she hadn’t seen any Stripe deposits in a long time. Months.
When I logged into her Stripe dashboard, I saw that that thousands of dollars had failed to transfer dozens and dozens of times. Stripe would attempt to transfer the funds, it would fail, and while they have email notifications for transfers, they don’t for failed deposits, weird. It failed over and over, trying in vain every two days.
Initially, I assumed it was a bank account error, and I went ahead and reverified and re-entered my client’s bank account and rounding numbers. Stripe requires 7 days to verify new bank account information. Their deposit schedule reflected this and we waited. After 7 days, the scheduled deposit failed, again.
I reached out to their support team, which was only available via email. In the two weeks since the first message was sent, the only responses I received were seemingly non-canned emails saying they were “looking into it”, and Stripe would “follow up as soon as possible…” which never happened. I sent many unanswered, one-way emails before I started to get really frustrated. (What’s truly strange is that follow-up messages went unanswered, but sending two additional distinct emails to Stripe’s support@stripe.com account returned separate agents mentioning there was no record of my previous emails…)
Despite being a financial services company, dealing with real money, Stripe lacked non-email support channels. When Stripe refused to respond to my questions or handle this situation within a generous timeframe, I decided to go public, on Designer News.
I don’t like sounding like a demanding prick. I have worked in the customer service industry (as an IT professional) and respond to support tickets at my current job quite regularly. I know that it can be hard to manage an onslaught of email based support requests in a timely manner. However, for a company valued at over $5 billion, I think having basic customer service phone or chat support isn’t too much to ask.
My post landed on the homepage of Designer News in a matter of hours and quickly rose to the top of the DN homepage. Stripe responded very quickly, both publically and privately once this happened. In fact, they solved the error/bug that was causing the problems and the funds were deposited the next day. Ignoring that Stripe appears to be a premier engineering institution which lacked basic error logging, I’m most frustrated on why it took so long for them to resolve this case, why they ignored me in the first place, and why it took a public confrontation for them to follow up with a resolution. (Which apparently was not difficult… As it only required 2 hours from Stripe’s initial comment on DN until they added that the bug was resolved.)
Stripe is not alone
Going to Twitter’s Search page, one can see all of the shitty experiences people have with United Airlines, Comcast, and Verizon play out online. This is in part due to social media making it easier and easier to voice your frustrations in a wildly public arena. I don’t think people would voice their hatred and frustration online as often and with such vitriol as they do if these companies designed their support channels in a customer-focused manner.
Businesses often treat customer service as a cost, not an investment. It’s one thing to have poor service options for your customers, but it’s another thing entirely to have no customer service, forcing customers to reach out to public. So many companies see customer service as a department to make as efficient and low-cost as possible. Often executives can’t see how it can make them money. Comcast tries to make their service department up-sell customers, which is not only poorly timed and frustrating, but distracts from the entire purpose of restoring a relationship with a company.
Customer service departments provide a way to invest in your business and customers and reap massive rewards in future sales and customer loyalty. It’s quite short-sighted to focus on customer acquisition without adequate ways to support and ensure success in your customer base! This is why I’m often not that frustrated about wait lists for services. (Though that’s for another time…) Cutting support department budgets will lead to increased churn and dissatisfaction with the product, service, or platform.
One comment I see on support pages that frustrates me: “We’re a small team and can’t offer support by phone/chat/anything but 2-week delayed email at this time.”
When I was at PayTango — an invoicing and automated accounts receivable company — the founder and I had in depth conversations about how important it was for us to have “above-and-beyond” service. When were just an invoicing platform, it was easy for us to convince ourselves that we weren’t mission critical. Meaning, if we went down for a few hours or half a day here and there, it wouldn’t bring any business to its knees. This isn’t a great mindset to being with. It’s a worse mindset, if your product is in fact, critical to a business’s operations.
While building out the product, we launched a new level of service that had our customers to hand-off their accounts receivables tasks to our team. We’d hunt down their unpaid invoices and handling the billing relationship with our customers’ customers. This decision had the side effect of us needing to implement lighting fast support, including a 24/7 toll-free phone number. Our company consisted of only two full-time employees and a few contracts. It was, however, blatantly clear that because of the industry we had entered (financial services) we had to be nearly always reachable. We were handling people’s money. We were depositing checks, handling wire transfers, credit card, and ACH payments as well. Not to mention following up with customers, via email and phone calls on behalf of our clients.
Financial services are incredibly mission critical. You can’t fuck it up.
Our two person company gave all of our customers the option to call us 24/7, which simultaneously rung all of us, and if it fails sends texts of the voicemails transcripts to our entire team. While it isn’t always an option to provide 24/7 service, email only support wasn’t on the table. Luckily for companies today, Twilio and Grasshopper and Phone.com provide wonderful, easy-to-implement phone and texting support channels. For us, it wasn’t acceptable to have limited support options. While I am aware this is one anecdotal, personal choice, I truly believe that no matter what scale you are at, you have the choice to invest in customer service.
Customer Service vs. Damage Control
There is a huge difference between Customer Service and Damage Control. What happened on my Designer News story was clearly Damage Control. Designer News is a somewhat influential news site and community of designers who can impact a business’s decision on which merchant service to choose.
This is just one of a few similar comments. Stripe most definitely lost customers who felt that this example of poor support meant Stripe would provide a similar level of support to them. It would have cost significantly less money to have just handled or escalated this issue.
Customer Service is often most successful when it’s not done in public. (Notice how most companies push for users tweeting angrily at them to follow / DM the customer service account. It’s not just for account security, it’s often to help let customers be heard and vent in a private space.) I’ve had so many bad experiences with customer service representatives. Even at companies which are known for their incredibly effective support systems, like Amazon and Apple, there are always bad apples.
I find the best setup for customer service success is when an agent can take the time to listen empathetically, and are personally empowered to solve problems without heavy management oversight or escalate without repercussions.
Be Human
Customer Service isn’t an exact science. There isn’t one motto that works for every company and every situation. A lot of what needs to go into customer service is a combination of long-term thinking and empathy.
Your employees and business must have empathy for your customers. Your customers aren’t calling to just chat or listen to your hold music (except if it’s this song). They are frustrated and probably quite irate. Even the best frontline agents’ can’t solve the clusterfuck when systemic problems, a nerve-wracking situation, some external/personal drama, difficult to reach representatives, long hold times, lack of updated documentation, and annoying ever-present marketing stack on top of one another.
If you’re a support, success, or operations team member, there’s a lot you can take from bad Customer Service “horror” stories. The solutions are not simple but require nuanced and a good deal of deep thinking and validation.
Customer Service Beats Damage Control Every Time
Can you give your agents the autonomy to make judgment calls on small stakes problems without management repercussions? Do your frontline people have the ability to escalate situations in a framework. A framework that is clearly explained to them? Can you make it easier for customers to not have to deal with complex IVR (interactive voice response — a.k.a. phone tree) systems? Can you back the thinking all the way up to make policies that can prevent the need for customers to call in the first place? Spending a significant effort on testing your current system, new support options and looping in feedback is a great way to improve the status quo.
In Stripe’s case, the fault didn’t lie entirely with the engineering teams or QA people. I bet it had a lot to do with policy. When I reached out the first time, a clear outline of what Stripe was going to do should have been laid out (not just “I’ll look into it”). Sure, they responded initially in 24-hours, but the response was just as meaningless as their canned “support case created” email. Stripe could have easily provided users who email in with an escalation path in their initial automated response. For instance, they could have included a phone number or non-technical chat option that is staffed during business hours for these type of situations. I didn’t need technical support– I needed someone who had empathy and power to not just file a bug report, but immediately find a way to rectify the situation. Could the agent have asked for Stripe to wire the pending funds immediately? Could one of the agents
Design isn’t just pixels. It’s mostly empathy.
Remembering that no matter how well a product is designed, it matters more how it works that how it looks. How it works isn’t skin deep. It goes all the way to business policies and company culture. Design isn’t just interfaces, animations, and pixels. It’s also policies, support interactions, and how services are designed for how they work when everything is perfect and how they work when the service falls short.