The idea is compelling: customers handle simple requests themselves, around the clock, and the service team gains time for the difficult cases. That is why almost every company now has a help center, an FAQ page or a customer portal. The results, however, are sobering. According to a Gartner survey of 5,728 customers, only 14 percent of service issues are fully resolved in self-service. Even for issues customers described as “very simple”, the figure was just 36 percent. At the same time, 73 percent of customers use self-service at some point in their journey.
So the willingness is there – it is the execution that fails. For small and medium-sized businesses this is good news: most of the causes can be fixed with manageable effort if self-service is designed around the customer’s request rather than around the website.
Why self-service so often ends on the phone
Gartner names the most common reason for failure as customers not finding relevant content – 43 percent said so. Behind this is usually a structural problem: help pages are built from the company’s point of view. They explain products, plans and procedures, but do not answer the question the customer actually arrives with – such as “Where is my delivery?” or “Why is my invoice higher than last month?”.
There is a second effect: many requests cannot be resolved with information alone. Someone who wants to change their delivery address does not need an article about how to change the address, but a function that lets them do it. If the help center ends with “Please contact our customer service”, it has not prevented the contact, only delayed it.
Start with contact reasons, not with the FAQ
The most sensible starting point is a sober look at your own data: why do customers get in touch at all? In most service teams, five to ten request types account for the bulk of the volume. Analyzing contacts by reason over a few weeks quickly shows which of them are suitable for self-service. Good candidates share three characteristics:
- They recur frequently and always follow the same pattern.
- The answer lives in a system – shipment status, contract term, ticket status, invoice copy.
- They require no judgment call, meaning no goodwill gesture and no case-by-case weighing.
With a contact center that records calls, chats and emails consistently with contact reasons, this analysis is not a special project. In myContactCenter, requests can be categorized and evaluated across channels – the basis for deciding where self-service pays off and where it does not.
Functions instead of text: what a customer portal should deliver
An effective self-service offering does not just answer questions, it completes tasks. For SMBs these are typically a few but central functions: checking the status of a request, booking or rescheduling an appointment, updating master data, registering a return, downloading documents.
The status check is particularly underestimated. A considerable share of calls to service teams are follow-ups on cases already in progress: Did my email arrive? When will someone get back to me? If customers can see the status of their ticket themselves at any time, a contact disappears that benefits neither the customer nor the team. It is important that the displayed status matches what the team sees internally – a portal that shows “in progress” while the case was closed long ago creates more calls than it prevents.
The way to a human must remain visible
A common mistake is to build self-service as a barrier: phone number and contact form are hidden so that customers “search for themselves first”. That may reduce contact volume in the short term, but it costs trust. Gartner found that 45 percent of customers who started in self-service felt the company did not understand their needs.
A better approach is an open exit at every point – one that carries the previous journey with it. When a customer starts a chat or submits a form after reading two help articles, the agent should see which pages they read and what details they have already provided. That way the customer does not have to start over. In an omnichannel solution like myContactCenter, such a contact arrives as a ticket with its history in the same system as calls, emails and WhatsApp messages.
Measuring correctly: resolved is not the same as did not call
Many companies evaluate self-service by page views or by a so-called deflection rate: how many visitors used the help center and did not get in touch afterwards? That sounds plausible, but it only measures that someone did not contact you – not that their problem was solved. Some customers simply give up.
Three questions are more meaningful: Does the customer get in touch about the same topic within a few days after all? How often is a task started in the portal but not completed? And how does the volume change for precisely those contact reasons for which a self-service option was created? Reviewing these values regularly reveals which content is missing and which functions do not work.
Conclusion: start small, measure against real requests
Self-service only relieves a service team if it actually resolves requests. The path there does not lead through the most comprehensive help center possible, but through the three or four most frequent contact reasons for which there is a real function instead of a text – with a path to a human that is visible at all times, and measurement that counts resolutions rather than page views.
Would you like to know which of your contact reasons are suitable for self-service? Book a no-obligation consultation via the online calendar at ilogixx.de – we will show you how self-service offerings, ticketing and telephony work together with myContactCenter.
