How long should a good onboarding experience be?

Question

I'm currently investigating and preparing to add, quite soon, a tool to welcome and guide a first-time users to the service we are offering, show them around, and explain where is everything located.

For this purpose I am considering Intro.js (link), a light script that isolates elements of the markup and adds a tooltip to further expand on its use.

While trying to think of the steps and tips, I came to this question I now ask you: how long should a good onboarding experience be?

Eventually, it's a guided tour that you can stop at any time, but our clients are from a specific niche we assume would want to take full advantage of the service and therefore will go through all of it.

We have 4-5 key features we want to cover, and we will add a couple more over the next few months.

So, what do you think is the best experience for the first-time user (though any user will be able to go through it again by clicking the "Help" button):

A. One "long" tour, covering all the features but not going through every option.
or
B. Short separate guidance for each feature.

I'm leaning towards option A, to avoid interrupting the user workflow multiple times, but I would love to hear your opinions, or alternative solutions if you have some.

EDIT: it's worth mentioning that this project is more of a business-to-business service, where we don't have so much input that we receive from the user but rather tables with statistics, graphs, and so on, should this affect your reply.

Answer 1

How long ? As long as the user needs to learn about the features available in the site or the web application.

Now coming to the options you have

1. Doing a complete introduction of all the features : This is also called joyriding where you give the person a quick walkthrough of all the features and give him a quick summary. However while this helps in establishing the focus of the web app there are a couple of concerns

   • Your user might lose interest or get disconnected in case of which he should again go through the onboarding process the next time
   • The user doesnt really do anything, he gets a tour of the features buts thats about it.

To quote this article

The “Joyriding” Approach

The “joyriding” approach walks the user through the features of an app or highlights the key features. It’s great because it clears up a lot of confusion right from the get-go. I think of this as the go-to approach since it’s what first comes to mind when you really think “onboarding.” While joyriding tends to be the most common process, it can be executed in many beautiful ways.

The disadvantage of this to quote the site again :

Cons: The user learns how it works but may not get far in actually doing it. They aren’t being begged to take action and start using the product.

2. Doing an individual introduction of features : While this gives the user flexiblity in not requiring him to go through the whole process, you still are just giving him an introduction and not getting him actively engaged in the process which basically doesn't get him into the system.

---

My suggestion

Look at an option of continuous onboarding where you get the user started on small steps and give him the option of onboarding himself completely at a later stage. Tumblr and Linkedin are good examples of this.

To quote this article about how Tumblr does it

This new approach is more direct than before since the user is faced with all suggestions to help him get started through a simple list on the dashboard. Each list item on hover highlights the particular UI element so that the user is directed to take action.

Coming to linkedin and other examples

What we’re calling “continued onboarding” are tactics a site uses to keep the user moving in the process of using the web application. LinkedIn, for example, has clear call outs at the top of your profile, most often asking you a question to add more information to your profile and encouraging you to “endorse” your connections. The incentive is getting a 100% complete profile, a continued setup approach, but getting that completion perhaps isn’t even possible!

Empty States are another approach to continued onboarding, which subtly and sometimes delightfully let users know they need to do something. They’re a great opportunity to use that empty real estate that exists when users haven’t taken any action to your advantage. Don’t stop at “You don’t have any friends yet.” Tack on “So get out there and make some!” (for example).

Answer 2

The key questions to ask yourself are how many use cases do you have, and how involved are they?

If you have one involved use case, such as "register for our site", then several less-involved use cases, such as "show report on industry trends", "show report on shortage prevention strategies", etc., then a joyride might help them get through the complex portion, and introduce them to one or two of the reports that give them benefits.

If you have several involved use cases, such as "register for our site", "create a strong connection to a partner", "how to search for partners that exactly meet your needs", then instead of a joyride, consider a "wizard" approach. A wizard is just a tool that walks you through a particular task, step by step.

The nice thing about wizards is that they're always in the toolbox, so you can pick one up any time that you need one. A joyride is really only valuable to an absolute beginner, but really drags on for an experienced user.