Trustpilot goes all in on self-service and gets results
Published June 23, 2020
Last modified July 13, 2020
Most companies realize that providing robust self-service options to their customers no longer falls into the “nice to have” category—yet creating and maintaining self-service that keeps up with ever-changing customer preferences and business needs can seem daunting. But as Trustpilot has discovered, making content easily accessible, improving agent efficiency, and harnessing actionable analytics can lead to the kind of self-service that drives scalability.
Trustpilot, an open online review platform where 90 million people (and counting) can share feedback about their buying experiences with other people, helps 400,000 companies on the service improve the value they deliver to their customers. And from the get-go, the company has been investing in knowledge-centered support.
“Eight years ago we started with Zendesk, and now we handle more than 150,000 tickets per year and more than 2 million self-service requests,” said Tonni Buur, vice president of global support for Trustpilot. “This year our self-service grew more than our business, enabling us to scale even further.”
By leveraging Zendesk Guide and its close integration to the Zendesk Support Suite of products, Trustpilot has been able to create modern experiences that empower its team to take an agile approach to knowledge and activate it in impactful ways. Let’s take a closer look at how Trustpilot accomplished this and how making knowledge work for your company can yield similar results.
Self-service best practices—the key to success
When Trustpilot decided to go all-in on self-service, it didn’t base the decision on a hunch—after all, data shows that 63% always (or almost always) start with a search of a company’s online resources when they have an issue, according to Zendesk’s 2020 Customer Experience Trends Report.
For Trustpilot, the key to meeting these customer expectations meant creating a framework that enables the company to build and scale its architecture that keeps pace with growth, which has been 50% year-over-year over the past decade. And that framework rests on a trio of best practices:
- Customer self-service: make knowledge available and easy to find for end users
- Agent efficiency: give support employees the tools to work smarter and more effectively
- Extend and measure: shape your knowledge strategy with the right insights
By implementing Guide with a clear self-service vision, Trustpilot has been able to grow, build, manage, and use knowledge to deflect tickets and improve agent capacity. That’s led to a 62% YoY growth in self-service, a 98% self-service success rate, and a 1,272% annual ROI on the platform.
Deflection and efficiency
So how did Trustpilot get here? It helps to start by breaking down the different use cases of self-service within your support strategy and create an underlying knowledge framework that can support this strategy.
- Deflection: this is likely the self-service use case that comes to mind first. Deflection covers any use of self-service, prior to or in the place of engaging with a support agent. Deflection can include instances where customers navigate to help organically, as well as those where content is served up to them proactively in context.
- Agent productivity: this is the use of knowledge and automation after a ticket has been submitted. Investing in self-service can play a pivotal role in arming agents with the right knowledge, decreasing agent ramp up times and increasing customer satisfaction on agent-assisted interactions.
By taking both customer self-service and agent efficiency into account while developing a knowledge strategy—while also being armed with measurable analytics—companies like Trustpilot create virtuous cycles in which self-service improves agent efficiency, and agent efficiency in turn feeds self-service. This tracks with what customers have been saying: 66% of them report that valuing their time is the top sign of great customer service, and 43% of customers point to automated systems that make it hard to reach an agent as top indicators of bad customer experiences.
Trustpilot’s advice to other companies
As Buur sees it, a framework should illuminate what needs to be built, which types of articles will be required, and which channels they must be optimized for. Beyond that, Buur argues, companies would be wise to shape content based on insights from data and analytics, and that simple, clear language works best.
At Trustpilot, Buur has broken down the company’s knowledge needs based on three factors:
- Who the end user is. In Trustpilot’s case, this is based on the stage of the customer journey, such as whether they are a prospect, new user, or experienced customer.
- The knowledge the customer needs. Are they most likely to be looking for high-level product capabilities or detailed troubleshooting tutorials?
- The channels they use to find help. While new customers are often navigating to the help center through third-party search, existing customers are typically leveraging in-product links or knowledge shared by agents.
That approach has helped Trustpilot boost website traffic by 46% and SEO rankings by 39%, with more than a third of traffic coming from organic search. In instances where Buur has enough information about what types of knowledge a customer may need, he leverages automation to reduce friction. “We don’t want customers to search for content,” Buur said. “We send them directly to the relevant knowledge based on what they need.”
For example, Trustpilot has built in product self-service to address the top needs its customers have at various points in the product interface. The company leverages the web widget on product pages to show specific articles that answer common issues from within that product area, optimizing keywords and headlines to work in that context. Similar exercises are conducted to optimize the customer experience with SEO, in-product referrals, and in-ticket knowledge sharing, providing Trustpilot several opportunities to optimize customers’ experience with self-service content.
On asynchronous channels such as email, Trustpilot has found that Answer Bot works well to provide immediate answers to customers. That further drives traffic to self-service content, freeing up agents to work on more complex support requests.
“Answer Bot is part of our first line of defense for deflection,” Buur said. “We use Answer Bot to help us solve tickets coming through webforms and email. We’ve learned that Trustpilot users prefer Answer Bot in the webforms—they seem to want to get it done in one interaction.”
Doing far more—with less
As Buur notes, one of Trustpilot’s strengths has been its focus on continuous improvements, with an eye on the long game. That far-reaching view has enabled the company to go from 60,000 self-service sessions to 2 million with just 22 agents—a fraction of the 125 employees Buur would have needed without the framework. That fine-tuned efficiency means that Buur’s organization is far from being seen as a cost center. Looking to the future, Buur hopes to turn support into a source of revenue, a testament to Trustpilot’s investment in scaling with self-service.
And those agents, using the Knowledge Capture app, continue to improve content by flagging and commenting on articles; those efforts are augmented by powerful data and analytics provided by Content Cues, which breaks down how content performs and helps content managers align articles to customer needs.
“You’re not done when you hit publish,” said Buur, who notes that content writers review all content once a quarter. “Relevant and updated content is everything.”
Take a closer look at how Trustpilot creates modern customer experiences with self-service—watch the webinar now.