How long should your web forms be to optimize conversion? We take a look at the data, the practicalities and share general principles for you to determine the correct length of your form.
When it comes to forms and how well they convert, there is a long standing and, so far unresolved, discussion about what the ideal length a form should be. Too long and your user will lose the will to live (in most cases), but too short and you might miss out on capturing valuable information and leaving your users with a feeling of uncertainty. You may well have seen headlines and clickbait that suggest X number of fields convert the best, but is it really that simple? Is there such a sweet spot? Typically, the truthful answer here is “it depends”. We outline how to analyse what the right length is for your form below but, in essence, the general principles are:
As clichéd as it may sound, when it comes to forms and how long they are it’s not really about how much you’ve got, but more to do with what you do with it. In fact, our own data here at Zuko suggests that the number of form fields may not impact conversion as much as you might think. Some have found that reducing form fields leads to a drop in conversions and others, that increasing the number of fields leads to an uplift in conversions and resulted in better quality leads. Huh?
If you’ve taken the time to read the reports above, you’ll see that for all of the experiments described, there were much bigger things to consider beyond the simple numerical value of how many inputs there were for a user to complete. But just in case you TL;DR’d, here are some general principles to consider when planning the length of your form to maximize conversion rates.
One of the first and foremost things that you must do is closely evaluate each and every form field that you are asking your users to fill out and decide whether or not that information request is actually something that you really need in order to get a form completion / conversion. Ask yourself whether it’s crucial, or just nice to have. If it falls into the former category then figure out if there is a more optimal way to ask for it. For example, could you just ask for a name, rather than first name AND last name? If the information request is more of a “nice to have” then make sure it’s always an optional choice and substantiate the request for it. Remember, sometimes you have to point out the obvious and be explicit in why a particular piece of information might be helpful if the user gives it to you, even if it is optional. A great example here is phone numbers. Along with dates of birth, phone numbers are closely protected information so there has to be a good enough reason to hand that over. An example might be asking for a phone number to receive tracking notifications on a valuable purchase.
Sometimes you can gather extra information from your users if they believe it will be helpful for them to give it to you, even if you keep it optional. For example, perhaps if the user provides you with their date of birth they’ll receive extra discounts, or a birthday surprise - something of sufficient value to the user for them to justify handing over the information (just make sure you follow through!).
Another approach is to try and understand what information you absolutely must have now to complete a given task and what information you could potentially ask further down the line once the initial hurdle of completing the form has been jumped over. For example, an account registration form may only ask for the basics in order to create an account and once that has been done, providing notifiers to the user in the account area that information needs to be completed at some point could work well. Another use case may be asking an account holder for more detailed information during particular critical pathways, but only when it’s applicable to do so rather than asking for everything straight away.
An offshoot of this is also providing your users a mechanism to come back to a partially completed form to complete it when they’re ready to do so. This is especially important for forms that are long or complex just by the very nature of the information they’re trying to gather (think financial forms such as car insurance quotes as a good example, or usually anything government related). Users should be able to return to a form that retains information that they have already entered, allow them to review and edit previously entered information and continue to complete the form at their discretion.
This should be self explanatory, but the amount of effort required to fill in your form and indeed, the types of information requests being made, need to equate to the perceived value of the thing that the user gets once the form has been completed.
If you’re presenting a user a form to sign up to a newsletter on a website they’ve never visited before and landed just seconds ago, then a) question why you’re asking for this information now, stop it and work on your timings to trigger it and b) make sure you are only asking for the basics - email address should be enough!
Let's look at another extreme. If you’re presenting a form to the user that will guarantee a quick and easy way to get competitive insurance quotes that will save them 100’s of pounds, then suddenly the value of completing that form potentially outweighs the effort of physically doing it and reduces barriers to providing more sensitive information (as the expectation is that this sort of information is actually needed to get a quote).
Is this ringing any bells? If you’ve heard of Fogg’s Behaviour Model, then you’re on the right track.
What Dr Fogg is basically saying is that the more motivated someone is by an end result, the more steps or hoops they’ll jump through to get there. You only have to look at TV shows like “I’m a Celebrity… Get me out of here”, “Love Island” and “Big Brother” to see this sort of behaviour in its full glory (or horror, you decide).
You can also see this phenomenon in Zuko’s conversion benchmarking data. Information collected across thousands of forms shows that forms on local government websites have crazily high conversion rates (85%) despite having the largest amount of inputs required to complete (71 on average). What is going on there? Users are highly motivated to complete the form; they probably really need the outcome a completed form will give them (housing application, tax rebate, etc) and, just as importantly, where else would they go for it? The local government monopoly on the required service ups their motivation enough for them to overcome any horrible, overlong form UX thrown at them.
It’s a classic quote, “quality over quantity” but from the conception of your form it is a question that you need to ask yourself before even starting to think about what information you will request from your user. If we think back to the Marketing Experiments case study, not only did they manage to increase the conversion rate of their form by introducing more fields to be completed by the user, but the quality of their leads also increased because they were able to validate them as they came in.
How did they accomplish this? By ensuring that the questions the form asked balanced with the value to the user and the cost of giving it the information. They realised that forms are not just forms, but are the opportunity to set up a conversation with your potential lead that can then be continued in the future. That, the right “ask” at the right time can imply value to your users and encourage them to continue. And, that your users will personify forms (tone, voice, personality) - they see forms as more than a “data transaction”, it’s a whole conversation and you (the person who designs the form) must guide that conversation.
The sort of thinking above doesn’t just relate to specific lead generation forms. You can also set up simple conversations in all sorts of scenarios, even eCommerce checkouts. For example, you could just ask for a “shipping address”, or turn it more conversational by asking “Where would you like us to ship your new shoes to?”. By reminding the user about the shoes you’re increasing the perceived value of the question and they may well be more inclined to enter the data required (even if your address lookup is shonky).
One thing that will impact your forms, no matter how many fields you do or don’t have in it, is how usable it is. A good place to start finding out how to do this would be to take a look at some of the advice we have on the Zuko blog, or check out our optimisation guides, too. But, since you’re here already, these are the biggest things to look out for:
Before you can even start to optimise a form for usability you need to understand what fields are working and are not working. Not to blow our own trumpet, but one of the best ways to do this is through Zuko data! It can tell you all sorts of useful information, such as:
You can get a free trial to put your form through its paces so you can get started on making it better. There’s nothing to lose, and we won’t even ask you for credit card details either.
Once you’ve figured out what fields need attention, your second port of call is removing any unnecessary distractions from the page. This doesn’t mean removing everything except the form, but keeping elements that are useful and add to the perceived value of providing information and removing those that add nothing of value at all.
Always provide any instructions that are helpful towards form completion. This can start before the user has even seen a form as given in this example for the UK Government online passport renewal service:
Instructions should also be given in the form itself, especially for fields that you know might have a high abandon rate. We’ll use the passport service as an example again, this time for when it comes to uploading your own digital photo to the website:
Not only do they give very clear instructions as to the criteria photos must meet, they also assist the user at the point the photo is to be uploaded:
All field labels should be clear and describe what information is to be entered and indicate clearly whether it’s mandatory or optional. Labels can also be used to provide helpful information when specific formats of data are to be entered:
Ideally, error handling should be performed in-line and not on “submit”. That means checking for potential problems with the data entered immediately as the user leaves a form field and presenting any errors and solutions to fix the problem that may be present in the moment, rather than waiting until later (post submit).
Error messages should be clearly visible, provide enough visual identifiers (not just the colour red!) that it is noticed by the user and articulates in plain language what the problem is and how it can be rectified. The same can be said for indicating when a user has completed a task successfully - let them know!
The Gov passport service doesn’t follow this advice, lets try and break the date of birth field and see what the user journey is like:
No inline validation to check DD MM YYYY (all of which are not valid)
Errors shown on submit. Error message only relates to Month. Error highlighting only for Day and Month, not year.
If you fix MM and submit, it tells you DD is incorrect. Still no indication YYYY is out of range
If you fix DD, MM and submit it then tells you YYYY is out of range but still highlights DD and MM as wrong when they’re right
The error handling here is poor. Not only are error messages only shown on submit, there are also problems with the logic used to check if fields have been corrected and now validate. This approach means that there are more steps to take to get a valid date of birth entered, whereas if inline validation had been used when the form was first presented the issues could have been fixed much quicker.
Forms, especially if they are longer in nature, or require complex data to be entered should be ordered and structured in a way that it not only makes logical sense to the user but also cuts the perceived effort time to complete it. Chunking related information together in steps can be helpful here as well as providing an indication as to what steps have been completed and those that remain.
Consideration should be given to which form elements you use and how that affects the ability for a user to navigate your form (tabbing) quickly and with ease and even to the positioning of those elements within your form.
You may also give some consideration to using smart form defaults to decrease the amount of effort your user has to put in.
Hopefully by now you’ve got some good ideas to figure out what the optimal length of a form should be and understand that it is entirely dependent on the purpose of the form and a balancing act between the users perceived value of the outcome versus the cost of filling it in. However, how can you be sure which number of fields is best for your form? The answer is to test and learn using data, ideally through an A/B test process.
If you’ve sufficient conversions and traffic to your form page you can run an A/B test to give you quantifiable and statistical findings between your control form and a variation of it that you want to try out (for example, your current form and one with some of the fields removed). Just make sure you’re tracking all the metrics that give you the insights you’ll need. For example, it’s probably not just enough to look at how many form submissions you got. You’ll need to dig deeper. How many of the submissions resulted in X (where X is a specific goal such as a final sale)? Or, perhaps you’ll investigate the overall completion time of a form, or the performance of a particular field in relation to abandon rates. The choice is yours, just make sure you’re tracking the data that helps you make better decisions depending on the overall goal you’re aiming for.
Sometimes you can’t run an a/b test for a number of reasons - insufficient traffic or conversions, technical complexity -v- ROI, maturity of optimisation program, stakeholders, hippos, insufficient resources, lack of research, ethical considerations, legal implications - and so the list goes on. Don’t fret though, not all is lost.
You can still compare what happens if a change is made and had it not have been made in the first place over a series of time instead. First, line up the KPI’s that you want to change and then ensure everything related to that KPI is identified, tracked and measured correctly. Then roll out your changes and see if those changes impact your KPI’s in the way that you’d hoped.
As noted earlier, it is too simplistic to give a ‘one size fits all’ number. The audiences and objectives of every form and checkout are different so it is going to depend. You are better off following the principles we have outlined to come to the right answer for you: consider the user and the businesses’ needs / expectations, be brutal in stripping out unnecessary fields, don’t include a question for information that you can get later instead, make sure you think about the users motivation (don’t have long forms for low motivation outcomes) and continually test and refine to see what works best for your form and its audience.
For more form design help take a look at our Big Guide to Form Optimization and Analytics.
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