.svg){kind=small}
We see thousands of online forms every year here at Zuko. This means that we have a huge database of form user behaviour that we can use to identify and analyse common trends to help the UX community improve the form experience for their website visitors.
We’ve previously released form conversion benchmark data comparing forms in different industries, but we do also get asked if we have benchmarks for individual fields. Although we firmly believe that every form is unique with its own quirks and improvements that can be made, there are common building blocks that form the basis of most forms. These are the questions that we consistently get asked when we try to fill in any form.
This is the first article in a series where we examine the aggregated performance of the most common form fields - name, address, email, phone number, postcode and password - to see which ones, in general, are the most problematic for users. We also then look at the types of issues that users may have with the most difficult fields and how you might be able to solve them.
This article focuses on the aggregated data for each field as a whole. Future articles will look at more segmented data (completion status, device type, etc) to see if we can surface more granular insights that may be of interest to the community.
As a starting point, we extracted data from 1,362 different forms in the Zuko database over a 12 month period. We then isolated the 6 most prevalent fields which gave us the below sample sizes (i.e. forms that contained this field).
From there, we aggregated the data and calculated an unweighted mean for the key metrics - abandon rate, time spent + fields returned - to allow comparison. We used an unweighted calculation to prevent high traffic forms from dominating the results.
As a countercheck, we also calculated the standard deviation of the abandon rate plus the median abandon rate in case the data was unusually spread.
The key metrics for each field are given in the table below:
The most obvious standout from the data is that the password field clearly looks to be the one that users struggle with most. It has the highest mean (and median) abandon rate, people have to return to the field to correct it more than any other, and users spend more time on it than all other fields than the address one (which typically spans multiple lines of text input).
Email Address and Phone Number follow Password as the fields with the highest abandonment rates (although the dropout rates are substantially lower).
The second, and less obvious, observation is that the standard deviation for all metrics is higher than the mean abandonment rate indicating that the data is likely to be highly skewed. Also, in all cases, the median abandonment rate is much lower than the mean which implies that field abandonment rates for most forms fall into the lower range with a tail of poorly performing fields dragging up the mean. If we look at the histograms for the password and email fields we can see that this is the case (Note the x-axis scale should be multiplied by 100 to get the abandonment rate as a percentage).
What does this pattern imply? Practically, it means that most forms that use these fields have an “acceptable” abandonment rate for them (i.e. at the lower end). Likely, they have implemented a treatment that minimises friction for the user - the questions are relatively simple after all (who forgets their own name!). However, there is also a long tail of forms who don’t seem to get it right. They are using a UI pattern that frustrates the user and causes them to drop out. Let this be a salutary warning - you can mess up even the seemingly simple fields on your form and cause unnecessary user abandonment if you are not careful.
Whilst every form is unique, we do see certain issues regularly crop up for some of these fields so we thought we would share our experience here. If you have a high abandonment rate for these particular fields you can use this advice as a starting point to develop your improvement hypotheses.
Who doesn’t think that creating passwords is a giant PITA? Either you make your password so difficult to remember so you forget it the next time you try to login. Or you use the same password every time, compromising your security.
We’ve discussed passwords previously, showing that most users have to return to the PW field at least one, a source of aggravation. If your form analytics data is showing that your password field is causing friction for your users, check if you are doing any of these things:
Over-restrictive stipulations - the biggest cause of password frustration is the fact that every site seems to have different requirements on special characters, numbers, capital letters, etc. You want your users to have secure passwords but you also want them to remember those passwords. Forcing them to create a 14 character password with multiple special requirements is a guarantee that they won’t be able to log in the next time they come back.
The important thing to remember is that password security is determined more by length than complexity. As this below chart from Statistica shows, adding one character to a password increases the theoretical ‘brute force’ cracking time by much more than adding one layer of complexity. For instance, if you have a 10 character password you can increase cracking time to 5 years by adding a single character and ‘only’ 7 months by adding a greater level of complexity.
What does this mean? Essentially, you don’t need to be as restrictive as you thought. We generally recommend following the UK Information Commission Office’s guidelines:
Unhelpful error messages - If your password error messages are unclear, accusative or unhelpful you are probably causing frustration and abandonment. Make sure any error messages state exactly what the issue is and what the user needs to do to fix it.
Not using inline validation - Inline validation is one of the most powerful tools you have to improve user experience on forms. The definitive study on this showed an average 22% completion uplift after implementing this technology. You should be firing any error messages as soon as the user focuses out of the field. Don’t wait until they click submit before telling them they’ve inputted incorrectly.
Forcing them to confirm their password - Resist this temptation. You may believe that the reduction in future password errors is worth it but we have never found it to be the case. The dropout from being unable to successfully confirm the password will likely outweigh the potential benefits.
And just to show we practise what we preach, we’ve removed the confirm password requirement from our own forms. We’ve even got a case study to show the effect of removing the confirm password requirement - a 56% increase in conversions.
Not allowing unmasking - If you don’t allow unmasking of the password, you’re going to have a lot of user friction. Forced masking is completely unnecessary as well - we’re not in the 1990’s anymore where people will be dialling in from an internet cafe! Most users will probably be on a mobile device so they can easily move to a secure place to avoid prying eyes.
Email comes in second after password in terms of average abandon rate. Why do people struggle with something so simple? In our experience there are generally two potential problems that businesses inflict on their users:
Mobile UI - Always make sure that your HTML is set to ‘Email’ rather than ‘Text’. This will mean that the ‘@’ sign will show when the mobile keyboard is revealed and the user won’t have to go hunting for it.
Confirm Email - As with password, if you ask users to confirm their email address you are going to get friction and dropout. Most users just copy and paste their email anyway so you aren’t going to cut down on input errors anyway. It’s much simpler to remove the requirement to confirm and see your abandonment rate decrease.
Phone number is comparable to email in terms of user frustration and abandonment. Why is this? Two main reasons:
Users hate giving you their phone number - It’s a sad truth but users are very suspicious of you asking for their number. They fear being spammed by sales calls so would rather drop out of the process than hand it over. Baymard Institute research showed that unexplained phone fields were a direct cause of abandonment. You are better off not asking for it at all. If you really must ask then explain why you need it. A simple line saying that you need their number in case there are any issues with delivery will do wonders for the completion rate on your phone field.
Format confusion - More than any field, phone numbers have a cornucopia of ways you can potentially enter the information. Do you add a ‘+’? The country code? How about spaces or dashes? <HEAD EXPLODES EMOJI>
Make it easy for the user. Do the formatting for them so they don’t need to think. Ideally a free text entry box where they can enter the number as they usually would and you do the formatting at the back end; either through some smart algorithm or, more simply, by getting your call centre operatives to interpret it themselves when they need to call. Humans are quite good at that - who knew!
We’ll be back with more data analysis of individual form fields but if your appetite has been whetted and you want more form tips then check out our Big Guide to Form Analytics and Optimization.
Zuko is the most powerful form analytics platform available on the market. Find out how to improve your form and checkout conversion by taking a product tour.
PRODUCT TOUR
