Out now: Value-Based Design, the definitive way to prove your design’s worth. Read it.

Form Field Analysis


Everybody knows that the fewer form fields you have, the higher your completion rate. Improving the usability of your form design is one of the best things you can do to capture leaking revenue and make for an easier experience.

You can research what to do with usability testing, but how do you measure and analyze form fields’ completion rates?

Here’s what I mean. Take a shipping address. It has a bunch of fields:

  • Name
  • Address 1
  • Address 2
  • City
  • State/Province/Prefecture/etc
  • Zip/Postal code
  • Country

Let’s take “Address 2,” every value-based design’s best friend. Here’s what the aforelinked Baymard piece has to say about “Address 2:”

Yet, during testing, users were often observed to be confused by “Address Line 2” fields, which made 30% of users come to a stop and furthermore made some question if their initial “Address Line 1” input was correct.

As a result, you should treat your address form as if it’s its own mini-funnel. The successful completion of each form field should be set up as its own individual event in Google Analytics, and you should then create a goal funnel to join each event together.

Note that the form field should be validated inline (and done right) as part of this. When a field shows as valid, it should tell the customer (by highlighting green, or removing a red highlight), as well as signify to GA that the field was successfully completed.

Form as funnel

Goal funnels should be created for each form group (e.g. address, account creation, billing information, etc). That way, you can assess the drop rates from one field to the next.

Most customers fill forms out in order. (Prove it with behavior recordings!) That means your funnel reporting will be roughly accurate – and any major drops (like, say, a 30% drop on “Address 2”) will be caught handily.

Setting up events

You’ll want to fire events for successful field completion upon successful validation – which means you’ll want to build that functionality into whatever functions you’re using to validate each form field.

Completion rates

Obviously, you want no appreciable drops between individual form fields. People who start a form should be able to finish a form. If lots of customers are stopping midway through the form, that’s a sign that you need to improve the wording, change the validation, or remove that field entirely.

Error rates

Reducing error rates is obviously a goal here as well. Individual error goals should be created, in order to determine any inline validation issues.

Again, you should add error reporting into your form validation JavaScript. If a field is filled with text, navigated away, and that text doesn’t fulfill the specifications of the form field, then you should tell the customer (by showing that it’s invalid, highlighting red, etc) and report it back to Google Analytics.


What are the biggest things that people do to improve form field completion rates, and reduce error rates?

Removing & reworking fields

Obviously, removing fields helps. You can make a data-driven case for removing fields if you analyze your forms in this way.

You can also change the form labeling, placeholder text, form field size, or sidebar tooltips.

Validation schemata

Take a phone number. In the US, your phone number looks like this:

+1 (888) 555-1212

Now, think of all the ways that people could type this:

888-555-1212 1-888-555-1212 18885551212 8885551212 888.555.1212 888 555 1212

Ideally, your form should be smart enough to strip everything that’s not a number from this and do the right thing with the phone number. Or, as people type, fill in the punctuation as needed – and then save it as just a number.

Experimenting with your validation schemata – and gracefully falling back to accommodate mobile keyboards and different treatments – is worth experimenting with, for any form fields that are failing to move people along towards conversion.

Wrapping up

The tutorials that we linked in this guide are not optional. You need to understand how to write events in Google Analytics in order to do this right. And you need to understand the mechanics of form validation – as well as the most common places that customers fail to proceed – in order to implement this sort of system.

← Back to the Blog Check Out Draft’s Store →