Web Analytics: Funnels, Filtering and Segmenting Data

To reach a conversion goal on a website, say completing checkout or registration, a user must pass through several intermediary pages. A funnel is a graphical way of displaying the dropout rates over the series of pages through which the user must navigate to reach the goal.

The purpose of funnels is to see how efficiently your pages direct visitors to your goal. It allows a website owner to see how users leave the site and make the necessary changes to improve the conversion rate. If any page in a conversion funnel is overly complicated or not designed to be user-friendly, you will see a significant drop-off. The page can be redesigned to improve the conversion of the entire process.

In the example above, over 50% of users do not continue from the Cart to Billing and Shipping. By redesigning this process, the website would significantly improve its sales without spending more money on advertising.

Filtering and Segmenting Data

Removing Irrelevant Data by Filtering

Your reports are only as reliable as the underlying data, so it is essential to ensure that your reports include only valid data. You might, for instance, wish to exclude information from visitors to your website from within your organisation.

Filters can remove certain information from your web analytics reports by excluding IP address or reporting on specific subdomains. You could find, for example, you receive a lot of data from a particular location that never converts. This could be a bot, and by excluding this data, your metrics will be more accurate.

Segmenting Data to Analyse Trends

Web analytics packages enable users to slice data along many dimensions, including location, demographic and channel. Segments let you separate out and analyse subsets of data so you can examine and respond to the trends in your business. For example, if you find that sales have dropped from users from a particular location, you can investigate the underlying reason. Maybe shipping rates are incorrectly configured?

Google Analytics

Google Analytics is Google’s free web analytics package and has become the industry standard. It is hugely powerful, supplying website owners will a vast array of metrics about their website.

Google Analytics does not just collect data from your site. It generates reports to display that data intelligently. These reports are what you can use to gain critical insight into your site and the people who visit it. Google Analytics contains a wealth of built-in reports which you help you understand your website performance. You can also build your reports or segment built-in reports to drill down on performance.

Real-time Reports

Real-Time reports allow you to watch activity as it happens on your site. The reports are updated continuously, and each hit is recorded seconds after it occurs. For example, you can see how many people are on your site right now, which pages or events they are interacting with, and which goal conversions have occurred. Reports include:

  • Locations. The geographic location of your active users.
  • Traffic Sources. The sources that referred the users who are on your site right now.
  • Content. Pages/screens viewed during the past 30 minutes.
  • Conversions. Real-time goal completions.

Audience Reports

Audience reports give details about your visitors and their behaviour.

  • Demographics. Age and gender.
  • Interests. Shopaholics, foodies, fashionistas.
  • Location. Where visitors are located.
  • Language. Language spoken.
  • Behaviour. How often visitors come to your website.
  • Technology and Mobile. The technology used to view your website, e.g., browser type.

Acquisition Reports

These reports give information on the channels that drove traffic came visitor behaviour by channel.

  • Traffic sources. e.g., Paid Search, natural search, referral
  • Google Ads. Data from a connected Google Ads account
  • Search console. Data from a connected Google Search console related to your natural search performance.
  • Social. Traffic from social networks

For each channel, the merchants can drill down to see metrics such as bounce rate, pages per visit, revenue and conversions.

Behaviour Reports

Behaviour reports tell you about performance of content on your site.

  • Most popular pages on your site.
  • Landing pages. Most entry pages on your website
  • Exit pages. Most common points of exit
  • Site search terms. Top search terms from internal site search
  • Site speed. How fast your website loads

Conversions

These reports measure how well your site encourages users to meet predefined goals (or conversions). These can be events such as signing up for a newsletter or making a purchase.

Goals

The goals report summarises your goals, including the number of completions and the conversion rate for each. Goals are user-defined actions you want to track. You can set both micro and macro goals to understand what activities contribute to overall outcomes, for example, the various stages of the checkout process.

For your goals, you will be able to track the following metrics.

  • Goal Completions. The total number of conversions.
  • Goal Value. The monetary value of conversions
  • Goal Conversion Rate. The rate of conversions out of all sessions.
  • Abandonment Rate. The people who did not complete your step-by-step goal / # of people that started your goal.

Ecommerce

The Ecommerce reports enable you to drill down on your sales to understand which products and promotions drive performance.

  • Product Performance. Product level sales data
  • Sales Performance. Transaction level data
  • Marketing. Performance of promotions and coupons

You must add the eCommerce code to the pages of your site to access this data. This sends additional transaction-level data to Google Analytics.

Multi-Channel Funnels

The customer journey to your website may often be via an indirect route. They may interact with your business on various other sites before making a purchase. For example, they may click on a paid search advert and then go to a cashback site before buying.

The Multi-Channel Funnel report investigates how different channels feed into your funnel. This includes:

  • Assisted Conversions. Where a channel was on the conversion path but was not the last step
  • Top Conversion Paths. The top routes to conversion, e.g., Paid Search > Organic Search
  • Path Length. The number of interactions in a path.

Google Optimise, A/B testing and Multi-variate testing

Once you have your website analytics package installed, you can start to understand its performance and look for ways to improve it.

A web page consists of several disparate elements, all of which work together to influence your users in ways that are not always intuitive. As a result, the only way to truly test page effectiveness is to change a page and see what happens. This involves:

  1. Analyse. Analyse your analytics data to look for underperforming areas.
  2. Hypothesis. Generate a testable hypothesis about your website. For example, if I make the buy now button green, it will increase revenue by 10%.
  3. Test. Run an A/B test or a multivariate test to test your theory.

Google Optimize

Google Optimize is a free service that allows website owners to test changes in your pages’ website content to figure out what affects conversions. You choose what parts of a page you would like to test, e.g., headline, image, promo text – and run an experiment on a part of your site traffic to decide which content your site users respond to best. Two testing options are A/B and Multivariate tests.

A/B testing

A/B testing is when you present two (or more) versions of a web page to different users. The user specifies the original page they want to test and provides alternative arrangements of that page they want to compare. Google Optimize will split the traffic between the different version of the page, enabling the user to establish which version performs best.

A/B experiments are the simpler version of testing with Google Optimize. If you have low amounts of traffic and want results fast, creating an A/B test may be the right place to start.

Multivariate testing

Multivariate testing allows the testing of multiple variables simultaneously. For example, using multivariate testing, you could select the headline, image and promotional text on a page as your page sections and generate three different versions of each one. When the experiment was running on your page, a user might see Headline A, Image B and Promotional Text C altogether, or Headline B, Image C and Promotional Text A.

Basic Web Metrics & KPIs (Key Performance Indicators)

Basic Web Metrics

These basic web performance metrics make up the basic building blocks of understanding what users are doing on your website and can be combined to create more advanced metrics (see below).

Sessions

A session is a period of user activity on a website. For example, a user may come to your site, browse some pages and then make a purchase. If they are inactive for 30 minutes, the session ends.

Note that a second session will be recorded if the user returns later that day.

Users

This is the number of visitors with at least one session on your website within a given period. If a visitor returns twice in a day, they will be recorded as one user.

Pageviews

The number of pages visited on your site in each period.

Pages per Session

The average pages viewed during a session on your website, calculated as page views divided by sessions. More pages per session mean that users are more engaged and visiting more of your site.

Average Session Duration

The average duration of visitors’ sessions. Longer sessions indicate that are more engaged.

Bounce Rate

Bounce rate is the percentage of visits where users viewed only one page and left. A high bounce rate suggests that people leave your site because they are not finding what they are looking for. You can minimise bounce rates by ensuring that landing pages are relevant to the advertising campaigns.

Percentage of New Sessions

The percentage of new visitors to your website. A successful website will have a mix of new and returning visitors.

Conversion

A conversion (a.k.a. Goal) is the completion of a user action on a website, e.g., making a purchase. The website owner defines conversion types that they wish to track.

Key Performance Indicators (KPIs)

Web analytic takes the speculation out of running a website. You can tell at once how much revenue a particular campaign, channel or promotion is generating. However, judging how well your website is performing and finding areas needing improvement requires identifying the metrics which affect your business’s performance. These are referred to as Key Performance Indicators (KPIs).

KPIs help inform website optimisation and the direction of your overall business strategy. If you are trying to decide which metrics are the most important to track for your business, here are three questions that should be asked:

  • If this metric changed, what impact would it have on the company?
  • Will improving this metric help us reach our strategic goals? What is it that your website is trying to achieve?
  • Will improving this metric also improve other metrics? For example, improving the conversion rate will also improve the cost per conversion.

Some of the key metrics which you should consider for your business’s KPIs are listed below.

Conversion Rate

A conversion can take different forms, depending on the purpose of a website. On most retail websites, a conversion is usually a sale. However, on other sites, a conversion might be an interaction where a user leaves some information, for example, completing a registration form. Conversion rate is the number of conversions (i.e. visitors taking the action you wanted them to take) divided by the total number of visitors.

Conversion rate (%) = (completed conversions / number of visitors) x 100

A better conversion rate can have a dramatic effect on the profitability of your site. According to Wordstream, the average conversion rate is 2.35%, but some businesses achieve above 10%.

Average Order Value (AOV)

The Average Order Value measures how much your customers typically spend on a single order. Hopefully, when people visit your website, they buy more than one item or choose more expensive over cheaper items. The average order value measures the success of cross-selling tactics on the website.

Average order value = total order value/number of orders

Shopping Cart Abandonment

Shopping cart abandonment refers to users adding items to their shopping basket but leaving the site without paying for the goods. This is a widespread occurrence online, with some website reporting as much as 50% of shopping carts abandoned. Abandonment happens for a variety of reasons. Users may be bored with the number of screens in the shopping checkout or visit the checkout to find the shipping costs. Many of the reasons for checkout abandonment boil down to poor usability and navigation.

With abandonment rates so high, reducing the percentage of drop out can improve the return on investment from online advertising and increase sales without increasing spending.

Shopping cart abandonment (%) = (number of completed carts / carts loaded) x 100

Cost Per Acquisition

Cost per Acquisition is the amount you typically spend to bring in a new customer. The acquisition cost calculation determines the value of a given promotional effort over a given period. If £2500 spent on 250 new customers, then the Cost per Acquisition is £10.

Acquisition cost = acquisition or promotional cost/Number of new customers

The amount you are happy to spend on a new customer is decided by your Customer Lifetime Value calculation.

Customer Lifetime Value (CLV)

Customer Lifetime Value is the revenue a customer will generate for your business over the course of your relationship with them.

Customer Lifetime Value = Customer Value x Average Customer Lifespan

Where Customer Value = Average purchase value / average purchase frequency rate

Cost per Conversion

The Cost per Conversion can be used to compare the price of different types of advertising or promotion. The metric focuses on results and not simply costs. A campaign may have a high absolute cost but a low cost per conversion, making it a cost-effective campaign. Conversely, a second campaign may have a limited budget but a high cost per conversion, making it expensive.

Cost per conversion = Costs/number of sales

Return on Investment (ROI)

Return on investment is a metric used to measure an investment’s effectiveness, be it a new computer system or online marketing campaign. Return on investment is given as a percentage:

Return on investment (ROI) = (Return/investment) x 100

For example, an online seller of electronics creates a PPC campaign to sell iPads. Each sale of an iPad generates a profit of £50. The PPC campaign costs £2000 and makes 200 sales.

Return on investment = (100 x 200/2000) x 100 = 500%

Benefits of Web Analytics

Web analytics is the measurement of visitors’ behaviour on a website, giving website owners vital information on who is visiting a website and what they are doing. If appropriately managed, this information can inform marketing activities and website design, leading to increased traffic and higher sales. Many businesses, however, do not measure the performance of their website in even the most basic sense.

The web allows tracking online customers and sales with accuracy and ease not possible with traditional marketing methods. Offline, it is not easy to get an accurate picture of how effective an advertising campaign has been. As the famous American businessperson John Wanamaker once said, ‘Half the money I spend on advertising is wasted; the trouble is I don’t know which half.’

However, the picture is different online as visits and sales can be tracked with great accuracy. With the correct setup, you can know where your visitors came from and how much they spent. For example, if a user clicks on a Google Search Ad and arrives at your shop, you can know the advert they clicked and the search term they used. By aggregating this data, different online promotion strategies can be measured and compared.

Web analytics is all about numbers, and they can seem daunting to some people. However, the process need not be complicated and can help build a more profitable website with a bit of time and effort.

Benefits of Web Analytics

Web analytics can give business owners essential metrics about the performance of their website. These metrics can tell merchants:

  • Who their customers are?
  • How they find your business
  • What they do when they get to your website

From within these metrics, businesses can choose Key Performance Indicators (KPIs) against which to measure their website’s performance. These are the metrics that have the most impact on your business’s bottom line.

Website Analytics can help an eCommerce website improve in three principal areas:

Online Promotion

As covered in this book, there are many different methods of promotion available for promoting a website. Web analytics can help identify which ones work best for your business by allowing the measurement and comparison of different sales channels. The data collected can help answer several questions about your marketing:

  • Which marketing channels refer visitors to your site?
  • How much traffic do they drive?
  • What products do customer buy?
  • What do customers spend?
  • Where do your site visitors live?
  • What are the demographics of your audience?
  • What is the conversion by channel?

The answers to these questions will enable you to focus your marketing efforts, showing you the channels to invest in and which to abandon. They can also help you define and understand your target audience.

Website Design and Performance

Online promotion gets people to a site, but it is what they do when they get there which matters. The navigation and usability of a website can have a massive effect on its performance. Installing tracking software such as Google Analytics allows the analysis of users’ navigation path through the website, outlining usability issues. Insights include:

  • Exit points. Are users leaving the site at particular points before completing a conversion?  For example, when shipping is quoted.
  • Shopping cart abandonment. What percentage of your items are added to a cart but never bought?
  • Bounce rates. The percentage of website visitors who leave the site after viewing only one page. This indicates that the site does not meet user expectations.
  • Page traffic. Which pages on my site get the most traffic?
  • Site speed. How quickly do my site’s pages load?

These insights give you concrete ways of improving your site performance. For example, you could learn that certain products are getting lots of traffic but few conversions. By enhancing page element (e.g., description, images and prices), you can run comparative tests to learn which features improve performance.

Search Engine Optimisation

Analytics insights also provide valuable insights for improving your search engine optimisation. The Web analytics can help answer questions about how users find and interact with your site including:

  • How much traffic is from natural search?
  • Which landing pages get the most traffic?
  • Popular search queries driving traffic to your site.

Using this information, you can see what content on your site attracts search engine traffic and the keywords driving traffic.  You can use this information to discover where you are performing poorly and create higher quality content.

The benefits of web analytics

Web analytics can give business owners essential metrics about the performance of their website. These metrics can tell merchants:

  • Who their customers are?
  • How they find your business
  • What they do when they get to your website

From within these metrics, businesses can choose Key Performance Indicators (KPIs) against which to measure their website’s performance. These are the metrics that have the most impact on your business’s bottom line.

Website Analytics can help an eCommerce website improve in three principal areas:

Online Promotion

As covered in this book, there are many different methods of promotion available for promoting a website. Web analytics can help identify which ones work best for your business by allowing the measurement and comparison of different sales channels. The data collected can help answer several questions about your marketing:

  • Which marketing channels refer visitors to your site?
  • How much traffic do they drive?
  • What products do customer buy?
  • What do customers spend?
  • Where do your site visitors live?
  • What are the demographics of your audience?
  • What is the conversion by channel?

The answers to these questions will enable you to focus your marketing efforts, showing you the channels to invest in and which to abandon. They can also help you define and understand your target audience.

Website Design and Performance

Online promotion gets people to a site, but it is what they do when they get there which matters. The navigation and usability of a website can have a massive effect on its performance. Installing tracking software such as Google Analytics allows the analysis of users’ navigation path through the website, outlining usability issues. Insights include:

  • Exit points. Are users leaving the site at particular points before completing a conversion?  For example, when shipping is quoted.
  • Shopping cart abandonment. What percentage of your items are added to a cart but never bought?
  • Bounce rates. The percentage of website visitors who leave the site after viewing only one page. This indicates that the site does not meet user expectations.
  • Page traffic. Which pages on my site get the most traffic?
  • Site speed. How quickly do my site’s pages load?

These insights give you concrete ways of improving your site performance. For example, you could learn that certain products are getting lots of traffic but few conversions. By enhancing page element (e.g., description, images and prices), you can run comparative tests to learn which features improve performance.

Search Engine Optimisation

Analytics insights also provide valuable insights for improving your search engine optimisation. The Web analytics can help answer questions about how users find and interact with your site including:

  • How much traffic is from natural search?
  • Which landing pages get the most traffic?
  • Popular search queries driving traffic to your site.

Using this information, you can see what content on your site attracts search engine traffic and the keywords driving traffic.  You can use this information to discover where you are performing poorly and create higher quality content.