"If we use gradual engagement, we will have more, higher quality signups" is a testable hypothesis. It has been tested, by many companies. I regret that I am not at liberty to disclose most specific results, but gradual engagement is <i>really</i> tricky to pull off well, and has often roundly failed compared to the traditional get-their-email-first signup screen. This is true even at companies which don't do anything very sophisticated with the email address once they have it, which is (IMHO) generally a mistake in the sort of markets I usually work in.<p>The one product I can talk about: Back in the day, Bingo Card Creator had one-click guest accounts. Their conversion rate was 2. Not two percent. Two. <i>Ever</i>. They were a cause of a stupendous portion of my support burden. (From the perspective of most of my users gradual engagement means "The Googles ate my work and now you have ruined the day of a room full of third graders, you monster.") The engineering to support them was fiddly, and ripping it out made the application better. (Despite several attempts to improve them I don't think I ever had near the UX work invested to make the experience not be awful. Again, gradual engagement UX is <i>quite challenging</i>. In particular, the handoff between guest accounts and "real" trial accounts is of paramount importance to my business but is meaningless to customers who have guest accounts <i>until they get to school</i>, at which point they will often discover, to their surprise, that failing to make the decision yesterday to give me their email address now means their cards are totally inaccessible. I never successfully figured out a way -- copy, design, workflow, etc -- to avoid having huge numbers of people fail at this use case.)<p>Discontinuing guest accounts increased signups of "real" accounts and also sales, if I remember correctly. You can eyeball the signup graph here <a href="http://www.bingocardcreator.com/stats/signups-per-day" rel="nofollow">http://www.bingocardcreator.com/stats/signups-per-day</a> Apologies in advance for the unclear axes -- that page hasn't had the underlying code updated in years, and I didn't even consider "Hey if I run this business for forever eventually that axis is going to get <i>crowded</i>."
This is a major pet peeve of mine with new web services. A lot of them are trying so hard to optimize the sign up that you can't even figure out what the service is and how it works without giving up your contact info.<p>So instead of giving the user a chance to be sold on what you're selling, you've just turned them away before they can even learn what your pitch is. You've killed an opportunity for word of mouth as well.<p>Maybe the idea is to select for users who are so eager to learn about your service that they'll give you their email first. Maybe that kind of customer is worth a lot more money.<p>But I think many people have marketing fatigue at this point, and are only going to sign up for things they know they're interested in. I mean, I wouldn't give my contact info to a store that doesn't even let me into the door until I fork over some info.
Quote: "99% of sites/apps/services we visit now make you register and go through an on-boarding process before getting to the meat of the product."<p>That's because signing up <i>is the product</i>. The touted "product" is a fiction, a pander to get you to sign up. Another way to say this is <i>you are the product</i>, the advertised "product" is just bait to lure you in.<p>If you haven't signed up and the company e-mails you, they're in violation of the Can-Spam Act. Once you've signed up, you become a customer, a category excluded from the sanctions of the Can-Spam Act. So getting you to sign up is not just the most important thing, <i>it is the only thing</i>.<p>I wish people who wrote articles like the linked one actually knew something -- that might make their articles worth reading.
"Stop making testable claims without evidence."<p>"Stop making absolute claims about things that may only be true sometimes."<p>"Stop assuming what frustrates you frustrates everyone."<p>"Stop thinking that other people cannot see 'obvious' design flaws like you can and didn't think through the tradeoffs involved."
This article makes points that sound fine on the surface, but ignore reality. There's a very good reason why a lot of sites do this and won't stop any time soon--it works. The article claims some effects to "conversions, usage and how people feel about your product" but is very light on the details of how it actually affects those things.<p>Getting a user to sign up facilitates a whole range of options (promotional emails being just one of them) that help drive user retention and engagement. I am not advocating making your product obscure <i>until</i> they sign up. The value proposition of your product should be clear, regardless of whether someone signs up or not. But it is not clear that making them signup before they can actually use it for themselves decreases conversion or usage.<p>The complaint made here will become ever weaker as "Sign Up with Google" and other single-sign-on services become more widespread. Signing up in those cases takes a single click, and my experience is in many cases instantly personalized with my data from other services. This might make some HN denizens cringe, but the average person seems to not mind.
It can cost real money foregone as the $300 million button article showed <a href="http://www.uie.com/articles/three_hund_million_button/" rel="nofollow">http://www.uie.com/articles/three_hund_million_button/</a><p>On the weekend I decided to try some boots from Zappos. They were bought by Amazon 4 years ago, and have "legendary" customer service. Not in my experience. You can login with your Amazon account, but then it asks for your name. Strange. I go to checkout and they want my billing name, street, city, zip, state, same for shipping, card number, expiry, cvc and who knows what else. Turns out that they aren't integrated with Amazon in any meaningful way.<p>I sent them an email about it - maybe I'd missed something. I got some nonsense explanation about it being for my security. Then they started spamming me about the abandoned shopping cart. At this point I discovered that each cs rep has some "humourous" boilerplate about how they are going to help you, before doing something completely unhelpful. They also don't keep track of replies (an id in the subject or just looking at the in-reply-to header would work) so each one starts a new ticket where a different rep doesn't look at the history and does a completely different unhelpful thing. I've now asked 3 times that they delete my account.
Sometimes, yes. But there are a lot of services where you <i>do</i> need an account to try it at all -- how are you going to try out Mint, or Duolingo, or Path, or OkCupid, without creating an account first?<p>Obviously you can create "dummy accounts", but they often won't actually give a decent idea of the site's experience (browsing profiles on OkCupid doesn't give you anywhere near the experience of having people message you), and then how is the person going to convert their dummy account into a real one later on, if they never even put in their e-mail or password?<p>There are certainly plenty of times when sites go overboard in asking too much of you up-front, but it's not always the case.
Another potential alternative: local accounts.<p>I worked on a site that's very privacy-oriented and users were still asking for accounts. We didn't want to store their information on our servers. Instead, we use localstorage to keep their settings and a list of things they've uploaded. They don't even have to sign in, and they still get the same experience. Try it: <a href="https://mediacru.sh" rel="nofollow">https://mediacru.sh</a><p>It obviously won't work for everyone, but if you just want to offer users a means to keep track of what they've been doing on your site, consider going locally.
I don't mind that. What bothers me is not knowing what a service actually does. Either no summary on their website, or what is much more common -- in follow-up e-mails.<p>I sign up for a service, or an invite for a service, or even just for a launch announcement, and weeks or months later get an e-mail that makes no mention of what is being touted.<p>But it's a plague in general. It's very common that Wikipedia has better descriptions of companies or software projects, even their commercial offering than the official website.<p>Try looking up which language some piece of software is written in. Often, googling it + github and then clicking on the repo breakdown is the easiest way to find out.
This was a major reason for creating my board game web site, BreakBase ( <a href="http://www.breakbase.com" rel="nofollow">http://www.breakbase.com</a> ). All I wanted to do was play a board game against my friend, and not have to worry about all of the nuances of making an account. Just share a link and play a game. Making an account is an after-thought, if you like using the service.<p>edit: hyperlinks, how do they work?
Users expect to register. I've seen UX lab studies testing it, without registration users get <i>really</i> confused and don't understand what's happening with their data.<p>Conversion can actually drop heavily with non-registration approaches.<p>Obviously it depends on the app, one where you only need to use it once you can and probably should design your UX to not need it. But if your app actually needs to maintain user state over multiple sessions then creating a user makes a lot of sense.<p>(Although you can build up user profile over time rather than requiring it upfront)
We tried this with the web app launcher at <a href="https://starthq.com" rel="nofollow">https://starthq.com</a>. Visitors could create their app launcher before signing up.<p>It did not work as expected as users were confused and few completed the process. Right now we ask for the email up front, but don't demand that the email is verified by clicking the link we send them. This works much better and we're seeing more than two thirds of the users coming back after the initial visit.
I very often turn away from sites that ask me to sign up. It doesn't matter how interested I am, if you ask me to sign up without allowing me to see what I came to see, I will go away and forget I ever heard about you.
So true; imgur should be mentioned as a service that offers great value without asking you to sign up.<p>I started urgeous.com with the same idea: blogging without signup. It got zero traction and was/is maybe too complex to use as it is but I still believe in the idea! ;-)
As a marketer, I see the temptation. Whether you like it or not, the things that you sign up for are more sticky, regardless of how good the product is.<p>It takes a leap of faith as a marketer to let you see a product without getting a way to get you back first. What marketers need to realize is that if you force people to sign up first they will leave in droves before they have any idea what you've created.<p>I think the optimal landing page is "create an account" with a way to bypass and "see it first."
We implemented this with Decal about year ago with an online tour that requires no signup[1]. We initially had an online tour which not only required signup but verified your email.<p>Not only did this mean fewer people would try it out, but it meant we got a great deal of disposable emails.<p>Our motivation was obvious: we wanted the free tour to get leads. Each time someone created an account to test, some resources were consumed so we needed to make sure we got a "high value" lead with a valid email.<p>When I watched Kevin Hale's Mixergy interview he talks about the fact that when they first launched the idea of Wufoo on their blog particletree they didn't even have a backend. It just demonstrated to people what it was like to use.<p>We were inspired by that when creating our own tour and we created a way to communicate the benefits our product offers for both deployment and end user interface in a frontend application that requires no signup or account creation.<p>Interestingly, not only do we get a better quality of email now, but about 50% of people who take the tour, put their email address in even though we only ask for it <i>at the end</i>.<p>[1]<a href="http://www.decalcms.com/" rel="nofollow">http://www.decalcms.com/</a>
If you are not making sign-up mandatory, you would not be able to validate the pain point of the problem. People normally would give you their email address(or even pay through their credit card) if you are solving burning problem for them. For self funded start-ups, this may not work.<p>However, this technique discussed in the article may work for consumer start-ups where you think of making money later and your present problem is to get millions of users.I assume you have sufficient runway(could be in the form of venture funding) to follow this approach.
Like nearly everyone else, I <i>despise</i> going through the registration process on every site I try. But certain applications simply don't work without registration. Actually, lots of applications don't. I recently started working on a tiny web app (<a href="http://alexreidy.me/apps/WhereIsMyComputer" rel="nofollow">http://alexreidy.me/apps/WhereIsMyComputer</a>) that lets you find your computer if it's lost or stolen, and by its very nature it's quite useless if you can't log in to see your device's location (and it's absurdly revealing if you alternatively display a page with thousands of names and corresponding coordinates). Since many sites simply wouldn't work without registration, I propose that we make registration more painless (maybe get rid of the whole email requirement and instead delete accounts that don't verify with email if they are clearly spammers or inactive) or, when a registration-free trial is not viable due to the nature of the application, we could design sites with a demo feature or demo video.
I suppose for me there are several categories.<p>1) If other data or significant user input of any sort is required or possible get a log in so that progress can be recovered. Shopping is probably an exception, I don't generally want an account just to buy the thing.<p>2) If it is a webservice or API or something similar I won't sign up generally until I have seen a) the pricing b) some documentation and c) had the option to review the terms of service. Going back to an earlier post today about improper use of Google's maps API. Don't hide your sales information (and that includes technical documentation) behind a sign in without a good reason.<p>3) In some cases I understand the the email address and permission to send me further marketing is the price of access to some information. I understand this and if the offer is good enough I may expect although I will probably decline four out of five times.
Umm.. you are missing a huge part of the psychology behind incremental buy-in. Think about it a bit. If someone let you test drive a Ferrari for a year would you ever be inclined to buy it? Why right. But if you had 20 friends tell you something is amazing and you should buy it and you will be amazed, chances are you will be intrigued and much more likely to put your money down. SaaS/cloud products are much harder to sell IMO than downloadable software too. We used to use the analogy that if someone is downloading your software we are getting an invitation into someone's front door to sell them on the product's virtues. With SaaS, the sales scenario is kin to trying to sell via product display in a crowded mall. Sensory overload with too many different options to distract someone's attention.
Well, I'm building a dating site as a hobby project. I could add a "Just try without an account" option but then, there's still a rather long form to fill to decide who the user should be connected with. Removing only two fields doesn't seem like a very good idea in this case.
Instead of abandoning the registration, maybe it could be streamlined to make it easier for the user.<p>In many cases it would be enough to just ask for user's email address. Giving just some email address would allow me in to service and I could start using it. On the background the system would send me the standard welcome message via email, but that would not require any immediate action from my behalf.<p>If I decided to become a regular user, I could then setup my password on the service. Maybe the system could point out in the UI that I'm not yet fully registered.<p>In case I forgot the whole thing and tried to use the same email to log on the service next week, the service would remind me that this email is already in use and ask me to go through the verification process to setup a password.
2013 and we still haven't really gotten to a good, universal single-sign on system.<p>OpenID seemed like it was going to get us there (at least to me), but nowadays many sites moved away in favour of Google or Facebook or Twitter specific logins, often asking for a little more rights than I'm comfortable giving them.<p>Still, anything is better than signing up to a site with a username and password, and receiving a mail with your account info in plain text. Since the LinkedIn and Twitter hacks, I've lost faith in the backend developers to treat the storage of account info with the respect they deserve. (at least use a hash and a unique salt per user...)
Great example of try first, signup later that I coincidentally just discovered is Litmus' email rendering preview system - <a href="https://litmus.com/email-testing" rel="nofollow">https://litmus.com/email-testing</a><p>They go quite far with it, which was really nice for my specific use case, allowing me to share the results with my team before proceeding with registration.<p>Now it begs the question whether or not I'll come back later to signup for the service, but regardless they've left a fantastic impression on me as a potential future user.
I suspect we don't hate signing up as much as the OP posits. I particularly dislike the shopping sites such as One Kings Lane that required singnup/login just to browse, but apparently OKL is doing something right since it I believe it's a "leader".<p>On the contrary, Stripe's guest access always struck me as very odd, and unlikely to be copied. For payments, I'd rather go ahead and signup to get my sandbox account.
I think Apple's app review guidelines help somewhat in this regard. They specifically forbid apps from requiring registration in order to work. I have had several apps rejected for this reason, and am now careful to always offer a path into apps that can be taken without signing up for an account.<p>Of course, they seem to enforce this irregularly, but at least their policy indicates they seem get it.
If I recall, Pandora has a relatively gentle onboarding. You can start a new station without being logged in and it's not until you interact with the site further (to rate, skip, etc) that they start pester you to sign up. I don't know how their conversion rates are, but I appreciated being allowed to at least get a brief feel for it before handing over my contact info.
This is the reason I started working on Streme (<a href="http://streme.co" rel="nofollow">http://streme.co</a>). I just wanted to be able to collaborate on a list of links (for sharing music recommendations with my brother) without needing to register for something and then also without having to convince him to register as well.
I understand the motivation for promiscuous early adopters but what's the motivation for the business?<p>If your business implement this strategy, you're relinquishing (in exchange of nothing) to the benefit of user retention created by the loss aversion bias and sense of belonging that the "harder" conversion has generated.
Slows development end of story. Since the dawn of the library people liked to be able to browse freely!!! meaning they could pick up and put down knowledge at will. Think to the shitty D&D movie and how the thieves want to sneak into the mages horde of knowledge and what a shitty hitty world that was
Count me as the opposite.<p>I like sites that have lots of screenshots and emails educating me on how something would work before I have to bother using it.<p>I have seen a lot of sites that let me "try before I sign up". More often than not, this gets me caught in the weeds of the product and it feels like work and I get turned off.
Totally agree with this article! There are very few instances where a signup is mandatory to "test the waters". At the very least you should be able to observe(not participate) an app/site/whatever without having to give credentials.<p>Every time I make something new, whether for myself or for work, conversion is the very first though. How can I eliminate text inputs, make less required, or eliminate them all together until the user feels it is worth their time to actually sign up? This is the question everyone creating these sites should ask.<p>It's incredibly frustrating to make it half way through the hoops on some promising website or app only to be confronted with the "sign up now" to get what you actually came here for in the first place and wasted 30 minutes creating, oh and where's your credit card? You can shove that right up your ass, i'm out, oh and pissed.<p>How this works is beyond me. I'd rather spend a week, month, year whatever making something better for myself than be bullied into signing up for your crap app.
Yep, I completely agree. We actually did exactly this (just last week) and created a QR for a live demo on our landing page (www.infoduce.com). This way people can see more than just a screenshot before having to "sign up."
What can I say, it's a good point.<p>500px does this nicely by giving you an anon user account on your first visit, and you can use the site in full that way. If you prefer to "sign up" and keep your stuff, it can be done.
The best way to stop making people sign up is to have a universal authentication method <i>that is trusted and used by everyone</i>.<p>I don't believe we'll see a good one in our lifetime.<p>But, that would stop an involved sign-up process.
Users are currency in the business of the web, and until putting a 5+ digit number of signups in your deck is ineffective, we'll optimize for the collection of email addresses over engagement.
I hate creating new accounts with passwords, etc but I don't mind "sign in with facebook/google/something most people already have".