Research fills in the gaps

How can you build a site for your audience if you don't understand what they want/need?

Getting Started

Chapters Three to Five were very much about you uncovering your needs in more detail and starting on the path of determining who you are building this website for. This chapter is the introduction into expanding your knowledge about your audience and moving away from gut assumptions, generalisations, and manager driven opinion. Use Research to gain knowledge and data that can drive sensible decision making.

In-depth research is a skill that can require lots of experience and resources. Our goal here is to give you the starting points that allow you to become proficient in a solid introductory research process for any project in keeping with the overall objective of this guide, which is for organisations that don't have all the resources but still need to improve their process and results. Use these options and you will be acquiring useful and intelligent information that will help you make better decisions. You can always learn more and engage specialists as your business processes mature.

User Needs

Understand the Needs of your audience

In Chapter 4 we looked at your audience and gaining an understanding of them which highlighted an approach focused on getting a better understanding of who your ideal customers are.

You can base it on existing customers whom you wish to replicate or those people you know you want as customers.

To build a website designed to not only attract your ideal clients but also to convert them requires you to understand their needs, wants, desires.

If you have learned one thing so far in this guide, it should be that 'it's all about the customer.' To make sure we can put ourselves in the minds of our clients and see it from your clients perspective we need to research more about them and how they go about finding answers to their questions and what triggers them to take action.

Research isn't just making an assumption from one view of a single piece of evidence it is about gathering as much available information on the topic you can to help guide your decisions through the rest of the process.

As Erika Hall wrote in 'Just Enough Research';

Research is simply systematic inquiry. You are solving the problem of a lack of information.

Make it practical as well. If you are starting out with a small site and don't have any or many customers, then you will probably be unable to get much existing research data but can ask prospects or ideal targets to gather their perspective. Then continue the research process post launch to understand better and improve what you built.

understand your audience

Learn from what you already have.

learn from analytics

The starting point for any research on your website development should be your current site. Instead of making snap "gut feel" decisions about what you need, learn to use data. Many reasons for a website change often come from internal boredom with the current site or frustration with how it's managed.

While it's true that having a site that's hard to administer will make life harder for you, if the site is working well for your customers you need to be very careful about why you are proposing to change it and what needs to change.

You will want to start looking through your analytics (e.g. Google Analytics or similar programs) and getting a better understanding of what your current site performance is.

If you don't have access to or are not even sure if you have Google Analytics (the most common free Analytic Tool) on your site, there is a simple check you can make.

  • Go to your website
  • In your browser, right-click on the home page
  • Choose 'View page source' - this will bring up the HTML for the page you are on.
  • Choose CTRL-F (win) or COMMAND-F (Mac) to bring up the Find panel
  • Type in:  Analytics or  UA-
  • If the code exists on your site, it will highlight some code with a value similar to this: UA-159492-3
  • If no results show up for either of those searches, then it is highly unlikely you have Google Analytics on the site

What to do if no Analytics:

  • Get in contact with whoever controls the current site content and coordinate with them to get Analytics on the site.
  • Once added you are going to need to let it gather data for a month or more. The more data, the better any observations.
  • Move on to other parts of your research and scope and come back to check your decisions when you have enough data to work with

What to do if there is Analytics:

  • Get in contact with whoever built or managed the site before you and ask who has the Analytics access.
  • Get yourself added to it as an Admin.
  • Once inside you need to start learning what data matters and what doesn't.

Just because you have Analytics installed doesn't mean it will be well configured or providing you with the best insights.

With the new goals in mind, you can start to analyse what currently assists that goal and what is missing.

Ultimately you want to know what parts of your site and from what Source (referral site or engine) lead to the most conversions that matter to your business.

The key things you will start with:

  1. Traffic sources (which ones matter most to your business currently)
  2. Content performance (what pages/content seems to hold people on your site and lead to your most conversions)
  3. Goals / Conversion (are you tracking these correctly and what does it tell you e.g. Inquiries, sales, etc.)

Understanding Analytics and the many threads of information it can provide you is beyond the scope of what our guide is here to teach. There are many articles, guides, and courses online that will help you get your analytics working correctly, inform and educate you on key metrics and help you use it in making better decisions.

Key research topics in your Analytics for your plan:

  • Gaps in your current content to what your customer analysis identifies.
  • Problematic pages that users bounce off and don't help them
  • What your goals are and which ones you can measure
  • Gaps in traffic sources

Learn about behaviour

User behaviour data on your site is the best way to win over anyone in your organisation that hasn't switched to a data-focused approach to your development. When you are looking to understand site flows for Information architecture fixes, or conversion problems, there is no better way than to watch and learn from Usability Testing Tools.

Listed below are several methods which will help you better understand how people are currently utilising your site. This type of information will help you make more informed decisions. Note that this practice should be included on your new site to ensure the new work you complete achieves the goals you set out to make. Always Testing!

Usability Tests:

1 Heatmaps


Heatmaps are graphical depictions of users using a page on your site. They typically will highlight Clicks, Movement, and Scrolling on a given page. From reviewing heatmaps, you can determine patterns on how current users are utilising the site. Many heatmap tools will also separate these maps into device specific maps.

2 Screen Recordings

Screen Recordings show real users on your site including where they click and the pages they visit. These recordings allow you to sit as if over the shoulder of a user and experience what they are experiencing. What it can't tell you is their intent or thinking, but it shows a lot of what they do to help you build useful knowledge of how your current site gets used.

3 Form Analytics

Form Analytics are particularly helpful in highlighting problems with forms on your site. You get to see which fields cause users to slow down or reconsider their engagement and where they drop off in a form. This type of data which is a must when improving your existing site and even more so when you design up the new forms for your new goals in your new site.

4 Funnel Analysis

Funnel Analysis works like Form Analytics and Form analysis in Google Analytics and is very useful in understanding how users travel through your site on their pathway to completing a key goal. Analysis like this is invaluable in understanding e-commerce sites and where site users drop out of your intended pathway to a sale/conversion. The analysis will show you how many people are leaving your funnel and at what point. With linked recordings and heatmaps you can understand much more about why they are leaving this process which will help you design a better path.

5 User Testing

User Testing is the step from hidden passive testing to more active forms of testing. You can utilise online resources or bring in people to use your site to complete several predetermined tasks and observe what they do, hear what they are thinking and expressing and seek to understand how they interpret the content, signals and options on your site. With the ability to get the actual feedback you get third party validation of what is happening and more knowledge to fix it.

The first four items in this list you can get using online app providers such as Crazy Egg, Inspectlet, Hotjar or others. You would sign up and pay to use their tools, have a small piece of Javascript code added to your site by your current development team and then within minutes you will have data to start analysing and visualising.

It's incredibly powerful, and an important start to learning more about what is currently happening and once you have rolled out your new project part of your ongoing regular analysis of what 's going on on the site.

Ask Questions

One of the oldest and most accurate ways to gather useful information is to ask questions of real people. Interviews.

When you have data that helps you spot patterns or highlights areas you don't understand what your clients might want or need, or more importantly why they are behaving in a certain way on your existing site, you can ask people directly.

There are multiple ways to get direct feedback the simplest is to get on the phone and talk to someone, or visit them in their environment and find out what matters most to them.

This process of direct feedback can be used in smaller ongoing changes as well as in research for a new project.

Assuming you modified a client payment screen for invoices from your company, you could ensure you get copied into all notifications and then contact each client to get immediate feedback on the changes.

When you do a very direct survey with a client, make sure you stick to the primary elements you want feedback on. In this example you could ask:

  • Did you notice the payment pages had changed?
  • Were the options easy to understand?
  • Do you think this page is better than the previous one (assuming they have used the previous one)?
  • Why?

Make these calls within minutes of the payment arriving, so it was top of mind for the user. You will get direct and priceless feedback from them, and it appears you care about what you are offering as well.

If conducting less specific research in interviews, you will ask questions that don't direct the answers but seek to gather honest answers to the type of problems they have and have them outline what a solution would look like to them.

Instead of asking 'Do you use the local Snap Fitness Centre?' you could be asking 'What types of exercise do you do?', 'Is any of that in an organised environment?'.

In this form of Qualitative Research, you are aiming to uncover information that provides insight into bigger questions than how an existing customer used a process.

Not all scenarios would make such direct surveying practical and with transactions occurring at any time of the day and across many time zones would be impractical.

Using Web-based tools to ask questions.

You can ask visitors to your site questions using a variety of methods.

Most polling/survey tools for web studies will provide one of these options:

  • Link to a questionnaire
  • Pop-Up panel or box that has either the question or a call to action to participate
  • Sidebar or Footer style direct questions

There are other types of options but utilising one of these will give you enough information to assist.
The types of questioning you can ask depends on the information you are trying to gather.

Some tools like Hotjar will allow you to poll a single question to specific pages. You can ask people on a product page whether or not the information on page answers the questions they have about your product, and if not what would they like added to the page.

If you don't currently have the audience you are after; you can contact other site owners and look to conduct a survey of their users/clients who are your ideal market segment.

ask questions

Turning Research into Useful Information

Now that you have gathered a lot of ideas, information and data; it's time to start planning how to deliver it in a useful, user-friendly method that will help meet your goals.

Moving from assumptions to compiling common themes into stories and answers about your ideal audience and clients. You will be able to complete more detailed personas and have a list of common problems, possible conceptual solutions and lists of ideas that need refining into something practical for the actual development phases.

Sorting your Data

After completing your research, you will need to sort the data you have gathered so it can be used effectively. One of the easiest and most common ways to sort this disparate but related data into useful information is to compile an Affinity Diagram.

An Affinity Diagram is simply data listed in themes or groups.

The offline method for creating Affinity Diagrams is to use Post-It notes and a whiteboard. You write down every idea/piece of data onto individual Post-It notes and then these can be visually sorted into 'columns' or groups on the whiteboard.

There are numerous tools you can use online to make an affinity diagram; the easiest is potentially Trello, which is the tool used in the screen shots below. In Trello Terms a 'List' = 'Group/Theme' and a 'Tile' = 'Card/Post-It'.

In each instance, you are trying to create functional groupings that will help you understand what you are trying to design better.

One Diagram might focus on the user Observations and Thinking whereas another might be more concerned with usability and visual design.

You will use the information gathered in numerous ways. Some of it will help you build better Personas, others will stand out as obvious content categories, all of it should help you determine what you need to bring together for your new site to help your audience get a better result from what you provide them.

Using Trello as an example.

Gather all your Notes and summaries from interviews, polls, user surveys, your internal customer, and sales teams and have it handy to group together. You can do this as a group (ideally) or if you are the sole resource on your own.

  • Create a new Trello Board.
  • Create a first List (Column) Called Legend which you can use to help identify what colours you are using.
  • Create individual tiles for types of data or interviews, etc. Give each a unique colour
  • Create a List to add tiles (I called this one Ungrouped in this instance)
  • Start adding information you have gathered from your interviews. Try to keep the title as succinct as possible. If ncessary add more data inside the tile
  • Add colour labels to each tile to represent where this observation or comment was gathered from. It's easy to see common items this way

Once you have added a good number of tiles, the list will become a bit unmanageable, and you will want to add your groups so you can drag tiles into them and start the process of grouping.

You may already have groups in your mind from your reviews of the data, or quite possibly they will stand our from the data you are adding in.

Note: If doing this with Post-It Notes offline it's very much the same process.

In our example of a Legal Company with data gathered from interviewing Young couples and families about Wills, you could end up with Group Headings like:

Lack of Knowledge, Fear of Legal Process/Lawyers, Money

You can add these as Lists in Trello.


Then it's simply a matter of dragging related tiles into the most appropriate column.


Using Affinity Diagrams/Tools is going to help you create useful information lists that help you solve issues and plan better as you progress to the next stages.

Not all data you gather might fit neatly into a list or group but still is crucial. An observations document that records items of high relevance to your project will allow you not to forget comments or observations of behaviour that will impact design or other functional items.

For example noticing that everyone gets a very slow page load on a particular part of an existing site (or all of it) should get you to record 'current site is very slow and leads to user frustration or abandonment.' You would then add to your design scope 'Site needs to load fast (under X seconds) on any (or listed) devices.'

From Data to Design

data to design

The next four chapters (Site Maps, Page Scoping, Wireframing, and Content) are listed in the order that seems to make the most sense in a linear process.

In reality, the process is more concurrent than linear, where elements you work on affect items in another part of this process and you need to adjust in each area.

It doesn't mean it's the perfect flow, but there is a degree of logic to it.

Where I struggle most in laying down a definitive method is the 'chicken before the egg' issue of content.

The debate is always "how can you plan a page if you don't know what the content will be" verse "How can you write all the content if you don't know how it will be presented."

Traditionally you would be presented with some design layouts and asked to gather content, or you draft up content for the designer who then wedges that into the theme they have already chosen for your site.

Neither of these methods create the right result but in the absence of a better method what else can you do?

Depending on each project I work almost in parallel through the next four steps. With the core content needs recognised but not written, I understand the topics, the goal and aim for that content and what its purpose is.

I would then step through Chapter 7 (Sitemap) and Chapter 8 (Scope) to define everything about the needed pages. I would typically then have enough to request the initial content because the scope would have outlined content needs pretty well.

If everyone understands that nothing is final at this point, it's ok. Once the wireframing has been done and approved (Chapter 9) when we get to Chapter 10 (Content is King), we should be able to guide the final content drafts with specific visual elements needed.

Thus, if we want a call out panel on a particular piece of content, that might not be written yet, but the bulk of the page content is. So now we can request from the content team that particular element with more accurate requirements e.g. a suggested link, purpose, etc.

This is true for visual content requirements as well. We discuss this more in Chapter 10.

Key Takeaway: Moving forward each of the following chapters exist in conjunction with the next, and you will likely be updating in each of the areas as you refine your scope.

BUY the entire guide as an ebook or paperback


Buy the Guide


Available on Amazon, Kobo, iBooks, Barnes & Noble and other great bookstores

Free access to the Website Planning Guide Facebook Group with any version you purchase