Why Your Help Island Knowledge base Software Isn't Helpful for Your Customers

If you’re an online business, you may have a knowledge base attached to your website. Popular examples are Zendesk’s Help Center or desk.com’s Support Center. You hope that when your customers have a question, they will go to your knowledge base, search, and find an answer. You hope that by using these self-service technologies, customers will answer their own questions and spare your company the endless deluge of email that comes in as support tickets.

But you may have noticed… this doesn’t work in practice.

Why not?

Here’s a hint: You do not have a content problem. You have an experience problem.


I was at the headquarters of a popular online business recently, and they showed me the “help link” embedded within their SaaS application, a complex product that requires significant learning on the part of their customers. Their help link was, as many help links are, at the bottom of the page. (But at least it was visible without any scrolling.) When clicked, this link led to the top generic entry-point of their article-filled knowledge base, presenting their customers with a big white text box in which to search.

I went out on a limb: “You know, less than 1% of your users will bother to ever click that link.”

They thought I was being unfair, so - to my amusement - they accessed Google Analytics right there on the spot to look up the actual usage of that link.

I had been wrong. It wasn’t 1%. It was 0.1%.

That’s one out of every 1000 users. And although we didn’t look it up, I’d wager the bounce rate from the entry-point to the knowledge base was over 90%. If so, that would mean 1 in 10,000 customers are actually using the help link and bothering to search. Who knows how many people actually find the answer they need. Even fewer.

But we all agreed that far more than 0.1% or even 1% of their customers have questions when using their app. That’s a lot of unserved customers. And a lot of customer support tickets that never should have been created in the first place.


At AnswerDash, we call customer support knowledge bases “help islands.” Help islands are stand-alone products bolted onto sites that are not integrated into the user experience. And while they can be useful for housing articles that customer support agents can link people to, as a self-service technology designed to deflect support tickets, they fail. And writing more articles won’t help.

Four reasons why “help islands” aren’t helpful

Here are four key reasons why your help island knowledge base isn’t providing effective self-service. And they all center on the poor customer experience that they create.

  1. They require customers to leave their point-of-action.

    We must remember that customers did not come to your site or app to use your help system. They came to get something done. Having to go off and use a knowledgebase takes them away from their point-of-action, which is where they had their question in the first place. The further customers get from their point-of-action, the less likely they are to come back.

  2. If users are going elsewhere to search, they are going to Google, not your knowledge base.

    We have all had the unpleasant experience of searching for something on a website and having our search results come up empty. That almost never happens at Google. Customers willing to “go off and search” for an answer are many times more likely to do so at Google than on your help island. And in the process, they are going to find your competitor.

  3. Your knowledge base puts the work burden on your customers.

    Ideally, your site or app is so well designed that questions don’t arise in the first place. But no such site or app exists in the universe. Your customers are going to have questions. And when they do, you need to make it as easy as possible for them to get answers without having to do too much work. A help island places a lot of work on customers. Not only do they have to travel to it, but after arriving, they have to formulate search queries, dig through search results, and scour articles for an answer. It is no wonder so few customers bother doing this.

  4. Not every question requires a full article. In fact, most don’t.

    Reading online is arduous and most people do not like to do it beyond a certain point. The online world is not a long article-sized world, like knowledge bases presume, but a Tweet-sized world. Quick, efficient answers that speak exactly to the customer’s concern are more valuable than long articles, even if the answer is buried somewhere within it.

AnswerDash solves these problems

AnswerDash was built to provide great user experiences and remedy the problems with help islands listed above. AnswerDash offers point-and-click contextual self-service for websites and web applications. It brings answers quickly to the customer’s point-of-action with just a click or tap. In recognition of most people’s desire to solve their own problems rather than contacting customer support, AnswerDash makes getting self-service answers easy. In fact, most customers never type a single word when getting answers with AnswerDash. Our data shows that 5-15% of customers on a website or web app use AnswerDash to get answers to their questions. That’s 50 to 150 times as much usage as the business I mentioned at the start of this post, and most online businesses that rely on help islands to deflect support tickets.

Get your free AnswerDash demo to learn how you can start giving your customers the answers they need, right where they need them.