Stop Asking Me to “Sign Up”

SaaS Startups Using Sign Up Buttons

The fate of many startups depends almost entirely on one conversion point: When a visitor becomes a user.

All too often, this pivotal role falls on the shoulders of a pitifully generic “sign up” button that’s lucky to get even a minute of consideration during development.

If you take a moment to consider the wording of your signup button, you can drastically increase how many of your visitors turn into users.

(What happens after the conversion is important, too, but things get significantly easier once you have someone’s email and opt-in.)

Why “Sign Up” buttons don’t work.

  • They’re ignored. When visitors see common elements repeated on many sites, they begin unconsciously ignoring those elements (aka “habituation”). Doesn’t matter if they’re blue or green or hell-fire-orange.
  • You’re asking for blind commitment. Don’t assume visitors know what you’re asking them to sign up for. People don’t read pages, they skim. They could’ve easily missed the part where you mention a free trial or key benefits.
  • You’re not offering any value. Asking someone to “sign up” offers no help in changing the visitor’s thinking from “Why should I?” to “I want this!”

How to get more signups from your signup button:

  • Tie it to your product. If you have a SaaS for trading bitcoins: “Start Trading Bitcoins.” If you have a marketplace for artists: “Start Selling Art.” This helps prevent the button from being overlooked.
  • Give, don’t take. “Get Access” and “Sign Up” both lead to the same thing, but one makes the visitor feel they’re getting something, while the other doesn’t.
  • Compel people to act. Use action verbs such as get, start, and try.

Examples from startups that get it right:

Startups with Good Signup Buttons

(I especially like gliffy’s “Start Drawing,” which implies how quickly you can get started, and directly relates to their product: an app for easily drawing diagrams.)

Case study: Getting 3x more clicks by changing two words.

Take this example from one of my clients. Like many software companies, Scalyr—a log aggregation and monitoring tool—asked its visitors to “sign up.” I suggested that we test a version of the button that gives and compels: “Try it Free”. Here’s what happened:

Asking visitors to Try it Free increased clicks by 212%.

Complete obliteration of the old “Sign Up” button, and a huge win for a test that took two minutes to set up.

A/B test on Scalyr resulted in 212% increase in clicks.

Now try it yourself…

Test a variation of your “sign up” button with something that givescompels, and is tied to your product. It’s one of the easiest things to test, and could have a huge effect on your conversion rates. If you follow my advice then you’re almost certain to do better than asking visitors to “sign up.”

PS — I write a post like this about once per month, about customer acquisition, conversion optimization, and web analytics. Enter your email below and I'll keep you updated about future posts.

  • Jamie Kravitz @uxinsf

    This is a good reminder to the general rule to make button labels descriptive and actionable, but also make it persuasive. Nice to see it backed up with data showing improvement as well.

  • Very good post! Thanks for examples from real life.

    As for ‘Sign up’ buttons I already started to ignore them. The company has to really get my interest to gain me as user.

  • Great post! Especially like the ones that are all about the benefit. Like Gliffy’s “Start drawing”. At (social media-as-a-service) we have tested a few. “Try It Free” gives the best result so far.

    • Grigoriy Kogan

      Thanks Jeniece. I would argue that you can do more on your homepage to get even better results. For example, it wasn’t clear to me even what you do even after skimming the entire page. You should email me. ;)

  • Great article, Grigoriy. It’s a simple concept that should honestly be obvious to people in our industry, but at times we can get so caught up in always doing things the same way, that we forget to re-evaluate our standard practices. This is a great exercise in getting back to basics and questioning the status quo. I’ll be sharing this with friends and colleagues.

    • Grigoriy Kogan

      Thanks Austin, glad you found this interesting.

      I can understand the tendency to follow the status quo… When developing a site or app, there’s simply not enough time (and/or money) to spend on meticulously thinking over every detail. Especially for startups. Unless you have all in-house staff and all the time in the world, I don’t see a way of avoiding this.

      That’s what makes conversion optimization so valuable, and necessary. There will always be room to improve signup rates (or other conversions), but you won’t know unless you test and optimize.

  • It’s clear good call-to-actions are the king in getting the responses you want for your business. I will definitely implement this on my websites. Thank you!

  • The real question is why we present logging in and signing up as different things. In either case, it’s about a visitor gaining access to a personalized experience. How about prompting a user – whether a first time or return visitor – to identify herself? She can do so with a user name and password or a social account login. If our system determines we’ve never seen her before, then we can ask her for additional information to personalize her experience going forward. In no case do we present it to the user as a “registration” or “sign up”.

    • This is an interesting theory Roger. Have you seen this done on any other websites already? I like the concept, and am curious how clearly others were able to implement this idea.

      • My team is working on it right now. Covering all the bases is a challenge, so we’ve been documenting the scenarios and edge cases as we go. (E.g. what happens if a first-time visitor logs in with her Facebook account, returns in the future, and then logs in with her Twitter account? We don’t recognize her as the person who logged in before, so we have to take special steps to “unify her identity”.)

        • Roger, I like how you’re thinking. We’re actually using somewhat of a similar solution in this comment section, if you really think about it. While I’ve never uploaded a profile picture to this particular site, one still displays as a result of my email address being linked to a Gravatar. It would be ideal to have a service that unifies social presences with one common account (perhaps an email address). Require their email address for sign up and through that, the “unified presence” will supply you with all of their social accounts. From that point on, they can log in / identify themselves with any of their social accounts.

          That’s distant thought, of course, and it’s best for you all to tackle the realistic hurdles right now. But I’d love to see a unified solution for one’s entire social presence. It genuinely would allow us to go from requesting users to sign up / log in to simply requesting they identify themselves. Or, in the event that they’re already logged in to another social account (perhaps in a different tab), we wouldn’t need to send them through an identification process at all. For example, if I’m already logged in to Twitter, I already have access to all of the apps that are connected to it. Now we just need to find a way to unify the networks with a common denominator. Easier said than done, but a thought provoking concept for sure.

  • Thanks for the reminder to look at every call to action to make it as compelling as possible. This was a helpful, motivating, and concise post.

    I’m curious what program you used to test the conversation rates (that’s shown in your screenshot).

    • Grigoriy Kogan

      Hey Alison, glad you found this helpful. I hope you also take away the lesson that it’s hugely beneficial to test your assumptions.

      I use Optimizely to run and analyze tests. I’ve been with them for a long time (even got certified). Some may also recommend Visual Website Optimizer as another option.

      • Grigoriy, I was in a Webinar with Dan Siroker from Optimizeley the other day and the product looked very attractive. How did you decide on Optimizeley? Would you recommend it over free alternatives?

        • Grigoriy Kogan

          Austin, I’m a big fan of Optimizely. When I was just starting out, I faced the same choice many people do: Visual Website Optimizer or Optimizely.

          I first tried out the former (VWO), but immediately ran into issues with their WYSIWYG editor while setting up an experiment. After some frustration and a lackluster response from their support, I then tried Optimizely and was able to set up my first test without a hitch. I’ve stuck with them since, and use them for all my client projects.

          Frankly I haven’t tried any free alternatives. The value I get for myself and my clients from Optimizely more than makes up for the subscription fee. For example, the successful test shown in this post resulted in more value gained than the cost of Optimizely for an entire year.

  • Thanks for the feedback, Grigoriy! This is v1 of our website and was just up a couple 3 days ago to coincide with the launch of our private beta, so has not been tested much. :) We are in the process of hooking up with Optimizely to start testing, so your feedback was much appreciated!


    • I noticed that you changed the verbiage on your site (“Get an invite” / “Sign up for free” / “Count me in”). Very nice. Being able to adapt like that will give you a competitive edge.

    • Grigoriy Kogan

      You’re welcome, Hans! Didn’t mean to pick on you guys, I just chose a handful of examples where I felt a lot was being left on the table by using the generic “Sign Up.” There’s more feedback where that came from — feel free to email me.

  • Jan

    Very useful advice for many startups – fantastic!
    Why don’t you have Twitter, Facebook and other social media buttons? It would be easier to share it!

    • Grigoriy Kogan

      Thanks Jan! I haven’t found a sharing plugin that I like — do you have any suggestion?

  • Fantastic reminder! I especially need to be aware of the point about users ignoring common elements. That’s what lead to the death of the carousel/slider (thank god).

  • Dmitry

    I would recommend you to give this WP plugin a try:


    • Grigoriy Kogan

      Спасибо Dmitry, I’ll check it out.

  • Looks good.

    Conversion rate is surely dependent on a list of factors, and ‘motivating the visitor to click’ and further proceed is one of the important jobs.

  • Goran Duskic

    Thanks for the grat article, so far it’s working for us, and we have a heatmap to prove it!

    • Grigoriy Kogan

      Nice! Thanks for sharing, Goran.

      • Goran Duskic

        Thanks! BTW, we worked on our forum last night, so if you didn’t see the screenshot yesterday when you wrote this comment, I apologize. The screenshot is now back up!

  • Sonya

    Thanks for sharing, Grigoriy! It works!