Is a confirm email address field still considered a best practice?
Having an input field to confirm an email address is pretty standard. However, I would like to know if using a confirm email address field still considered a best practice.
Confirm emailinput controls account for a majority of the form abandonment for a client of mine. There are a variety of different reasons for it but i'm offering the following suggestions.
- Remove the confirm email address. My rationale is that no one actually double types so if they've copied an pasted from the email input control they'll both match but you can't guarantee either is right.
Personally i'd like them to remove it. I think they are pointless and the form is very long as it is but i need evidence not just my opinion.
I'd be very interested in knowing what - Did you remove the extra field? Did it affect the form abandonment? After all, the only way to determine of the problem was successfully solved is to test...
Ideally it will autocomplete anyway https://developer.mozilla.org/en/docs/Web/HTML/Element/Input#attr-autocomplete
I’d recommend not including a ‘confirm email’. I agree that many people copy and paste rather than re-typing the email address from the original ‘email address’ field (defeating the purpose of the confirmation field).
Some related research/findings/best-practices:
- email confirmation was mandatory in only 18% of sites […] large websites such as Facebook, LinkedIn, Stumbleupon and Twitter don’t require password confirmation.
- ‘designers tend to remove all unnecessary details and distractions which don’t help the user to complete the form’. More detailed analysis showed a trend of using very few mandatory fields – more than 50% of forms used at most 5 mandatory fields, while optional fields were often avoided. This can be useful to you when deciding on required fields.
2) Luke Wroblewski’s ‘Inline Validation in Web Forms’ is also a great resource for findings and best practice recommendations re web forms design
i'll add some technical stuff:
name the field "email" -> browsers recognize the field and provide you with your email addresses you complete in those fields
put the type of the field "email". some devices (like iphone) use a different keyboard when dealing with this type of inputs, i read a very good article on this but i can't find it.
Is this the article that you are referring to? http://diveintohtml5.org/forms.html
Excellent advice - not sure why this wasn't upvoted - until now :) The 'type="email"' is actually from HTML5; obviously some vendors are already recognising it.
Indeed, mobile Safari (i.e. iPhone/iPad) supports field type specific keyboard layouts and provides other UI data entry UX goodness. Not sure if other WebKit-based mobile browsers do too yet (e.g. Android default browser) but they will in the future...
In my opinion, the "confirm email address" field is just another barrier of entry for your registration.
My theory on this and from personal experience: People type their email all the time, if there's a mistake in what they just typed, they notice it rapidly.
Here's an interesting registration flow to think about:
- User registers
- User is automatically logged in after registration and the confirmation email is sent simultaneously
- You add a site-wide notification to tell the user to check his inbox for the confirmation email and if he didn't receive it, he may have mistyped it. Perhaps also explain how important this information is for your site.
Here's a decent discussion on IxDA: Asking twice for email address in a form
I agree with what you've written but is there any evidence to suggest it's not just our (Experienced web users) preference?
Here's a hypothetical to consider. Let's say your signup page gets 500,000 visitors and a 20% conversion for 100,000 signups a year.
Without an email confirmation field, reasonable estimates say 0.4% (or 400) of these users enter the wrong email address. With email confirmation, a generous estimate says that drops to 0.1%. So the net loss from removing the confirmation field is 0.3% (300 users).
You only have to improve your conversion rate to 20.06% to break even on that loss.
Now if we say email confirmation constitutes one of five required fields on the average signup form, removing that field means reducing the user's work by 20%, and it's hard to imagine that would improve conversions by less than 0.06%.
There is also the opportunity that users who enter the wrong email address will realize it later and correct it, or simply create a new account.
With these things in mind, I would favor forms without email confirmation.
I've seen this question brought up in a number of places. After a lot of comments and iterations at one such forum (can't remember where), the best practice i found was:
. Ask the user to enter their email address . At the end of the form show the email address again with a message saying: Email confirmation will be sent to: [email protected] (change)
This gives the user an option to change the email address if the one entered isn't right. Asking users to enter the same information twice is not quite helpful and more often than not, users like me just copy and paste the email address from the first box.
Is this the article you mentioned? Solving the repeat email address issue - http://tinyurl.com/mlekdl
I'd also vote for asking it only once.
If your really need to, you can confirm the email in other ways as well, ie:
See a good article on exploring email inputs from Luke Wroblewski (also the source of the image above)
I'll be saying something that won't be popular. Like most answers here we have deployed our iPhone app without a second confirmation email field.
But know we are seeing many email bounces. Users write Gamil, instead of gmail. Or Yaoo instead of yahoo.
We are using Amazon SES and our bounce rate is already at 1.25%. This is what Amazon says about it:
Your current bounce rate is 1.25%. This is measured over the last 2,378 eligible emails you sent, spanning over approximately the last 88.4 days. We expect our senders' bounce rates to remain below 5%. Senders with a bounce rate exceeding 10% risk a sending Pause. Learn more.
We have gone up from 0 to 1.25% in less than a week. By that rate we reach 10% and our entire registration comes to an halt, just because of a slight inconvenience in the UI. Lesson learnt, we are going back to introduce the confirmation field again to catch bad emails in the first place.
Suggestion in a related vein (once is enough for e-mail address entry, agreed) - find a way to float an enlarged version of the text entry field when an onFocus() occurs so the user can easily see the contents of the box while typing and identify errors which would be slightly less obvious at 12 points.
Consider the users (both internal and web-facing) and weigh the inconvenience against the repairs required to fix typos.
A position for the keeping of confirmation:
Customers normally find us and come to us; they don't browse by. So they will put in a little bit of effort to stay. Our customer base isn't leading edge tech savvy so they are not understanding of how badly things go wrong when they typo their email address. Then they get quite frustrated and blame us for their mistake.
My store uses Magento Commerce and everything is keyed on the email address. I'm currently adding a second field for confirmation. At least when someone registers with a typo email address (and places an order) we can search on their name and find the typo.