Newsletter Sign Up

Submit a Question

  • This field is for validation purposes and should be left unchanged.

Google Analytics Guide

Assessing 4-Tell Recommendations

This document provides the steps to setup Google Analytics to track usage and how to assess the effectiveness of your 4-Tell Recommendations.

Background

Product recommendations can increase sales in three key ways:

  1. Convert browsers to buyers – help customers find what they are looking for
  2. Larger orders – since customers see cross-sell and up-sell items
  3. Repeat customers – a good web experience builds customer loyalty

If you are a manufacturer, they increase sales in a fourth way:

  1. Increase sales on dealer’s websites – since customers research products on your website, find related products, and then buy them on your dealer’s websites

Google Analytics (GA) is a free web analytics solution that gives you rich insights into your website traffic and marketing effectiveness. It allows you to see and analyze your online traffic and better understand conversion. To learn more about Google Analytics, you can click on any of the following help links:

Sessions

Google Analytics displays most reports in terms of sessions. Sessions represent the number of individual visits initiated by all the users to your site. If a shopper is inactive on your site for 30 minutes or more, any future activity will be attributed to a new session (i.e. new visit). Shoppers that leave your site and return within 30 minutes will be counted as part of the original session. (Thus, session = shopper’s visit and user = shopper.)

Events

Google Analytics uses events to segment reports. We trigger an event when a shopper clicks on recommendations. This enables all reports to be viewed in terms of ‘All Visits’ versus visits that clicked on 4-Tell Recommendations (i.e. ‘4-Tell Recs’). The events also enable any reports to be segmented by recommendation type, page type, email type, product name, and recommendation position (as discussed in the Advanced Usage section below).

Influenced Revenue

Influenced revenue is the revenue for shoppers that click on 4-Tell recommendations (i.e. revenue in the ‘4-Tell Recs segment). If a visitor clicks on a recommendation and buys it (alone or with something else), the purchase counts influenced revenue. In other words, it is a conversion for a shopper that clicked on a 4-Tell recommendation. Alternatively, if a visitor clicks on a recommendation, leaves the site, then returns the next day and buys the recommended product, Google Analytics does not count that purchase as influenced revenue. Visa-versa, if a shopper clicks on a recommendation, and buys something else, Google Analytics shows that the purchase was influenced revenue. The shopper may have bought this product since the recommendations kept the shopper on the site.

We’ve done A/B Testing and a theoretical review to determine that 80% of influenced revenue is an increase in sales. In other words, 80% of the revenue attributed to 4-Tell recommendations is an increase in sales.

Google Analytics Setup

The setup described below assumes you are running Google Analytics and using a 4-Tell plugin (3dcart, Magento, Volusion, AspDotNetStorefront, Bigcommerce, Shopify, etc.) that is already linked to Google Analytics.

  • If you have not setup Google Analytics or Ecommerce Tracking on your website, please see Appendix B: Setup Google Analytics.
  • If you use a custom integration, and not our plugin, please see Appendix C: Event Tracking for Custom Integration.

Add 4-Tell to Your Google Analytics

Please add GA2@4-Tell.net to your analytics account so we can help optimize your results, create reports and share the results with you. To add 4-Tell, access User Management, which is found in through the Admin menu bar (full details in Appendix A: Add 4-Tell).

Creating Custom Segments

Please wait to setup segments in your Google Analytics account until your 4-Tell recommendations have been running for a day since it takes Google 24 hours to process the events and add ‘4TellRecs’ to your event queue. The historical data is not lost even if you have not setup segments. In other words, you can setup segments and view previous data.

Create '4-Tell Recs' Segment

The easiest way to add this segment is to click on the below link:

https://www.google.com/analytics/web/template?uid=bnj-WTbVT1-i7IMASWi1TA

In case the link doesn’t work:

  1. In the main window, do the following:
    1. Click the ‘+ Add Segment’ button

b. then the red ‘+ New Segment’ button

  1. In the Segment window, do the following:
    1. Select the Name text box in the upper left, and enter ‘4-Tell Recs’
    2. Under Advanced, select Conditions
    3. Change “Ad Content” to “Event Category” (under the Content section)
    4. Make sure the next box is set to “contains”
    5. In the text box to the right of the ‘contains’ enter the text ‘4TellRecs’. As you type, the rest of the name should be filled in for you. If the text does not appear, you need to wait 24 hours for Google to process our events.
    6. Click on the ‘Save’ button in the lower left.

Create ‘NOT 4-Tell Recs’ Segment

The easiest way to add this segment is to click on the below link:

https://www.google.com/analytics/web/template?uid=sqtOOg2rQFuj6ibTYMAe-w

In case the link doesn’t work:

  1. Click again on the down arrow (as shown in the image for step 1a above)
    1. Select the ‘4-Tell Recs’ segment you just created. To the right, click on the star and click Copy
  1. Select the Name text box in the upper left, and enter ‘NOT 4-Tell Recs’
  2. Select the dropdown menu that currently says ‘Include’, and change to ‘Exclude’
  3. Click on the ‘Save Segment’ button in the lower left.

How to Show Segments in Reports

Click again on the down arrow to show the segments you just created (1a above.) ‘All Visits’ is clicked on by default. Click on the two segments you just created:  ‘4-Tell Recs’, and ‘NOT 4-Tell Recs’, and then click on the ‘Apply’ button. Now, reports show analytics for ‘All Visits’, visits that clicked on 4-Tell Recs, and visits that did not click on 4-Tell Recs.

NOTE: The reports list the segments in the order that you clicked on them but once selected, you can drag the boxes at the top to rearrange the order.

Web, Email, and Mobile Segments

You probably notice we have additional segments in the above list. The segments you created include anyone who clicked on a 4-Tell recommendations. You can separate Web from Email. If you have a mobile template or responsive design, you can use Google Analytics > Audience > Mobile reports to compare the results.

As background, visitors that click on 4-Tell on the website trigger the event ‘4TellRecsWeb’, and visitors that click on an email trigger the event ‘4TellRecsEmail’. (Our events were changed in 2/2015. If you want web only for before that date, the new segment must have ‘Exactly Match’ rather than ‘Containing’ in the match parameter – see step 2d.)

Google Tag Manager

If  you use Google Tag Manager on your site, our solution automatically sends a ‘4Tell’ tag event to Tag Manager. Tag Manager needs to be setup to forward that event to Google Analytics, as described below.

In Google Tag Manager, you have two steps.

First, you setup a trigger that is triggered on the ‘4Tell’ event (see screenshot to the right). You select ‘Triggers’ in the left menu, and then the ‘New’ button at the top of the frame. You select a Custom event, have it fire on ‘4Tell’, and name it ‘4Tell’.

Second, you create a tag that is fired by the ‘4Tell’ trigger, and the tag forwards the event information to Google Analytics (see screenshot below).

You select ‘Tags’ in the left menu, and then the ‘New’ Button at the top of the frame. You select Google Analytics, Universal analytics (unless you are still using Classic Analytics), and Track type of Event. Then, fill in the event boxes by selecting the button to the right and choosing ‘New Variable…’ from the bottom of the dropdown list. In the ‘New Variable’ box, select Data Layer variable, enter the variable name as eventCategory, eventAction, eventLabel, and eventValue to match the tracking parameter, select Version 2, select ‘Create Variable’ button, and finally give the variable the name eventCategory, eventAction, eventLabel and eventValue to match, and Save. You don’t enter the brackets {{}} since Google automatically enters those for you. Leave Non-Interaction Hit as False.

When done, select ‘Save Tag’, and select Publish. The final result should look like the screenshot below.

Key Performance Indicators (KPI)

Google Analytics reports show four key performance indicators (KPIs) to help you assess the effectiveness of implementing 4-Tell’s recommendations:

  1. Percentage of revenue influenced by shoppers who click on recommendations
  2. Increase in conversion for shoppers who click on recommendations
  3. Percentage of product/page views is from shoppers who clicked on recommendations
  4. Increase in products/pages viewed per visit by shoppers who click on recommendations

Analyzing the Key Indicators

You access your account from www.google.com/analytics/, and click on an account on the left side. This shows the ‘Standard Reports’ window. Select the Segments drop-down menu (down arrow), select ‘All Visits’ ‘4-Tell Recs’ and ‘NOT 4-Tell. Don’t forget to hit the Apply button.

Now, view the different analytics reports. The two key one’s we focus on are Conversions>Ecommerce>Overview and Audience>Overview. The comparison of ‘All Visits’ versus ‘4-Tell Recs’ (i.e. visitors who clicked on 4-Tell recommendations) versus ‘NOT 4-Tell Recs’ is shown throughout.

Revenue & Conversion KPIs

To find the first two KPIs, in the menu of the left near the bottom, you view Conversions > Ecommerce > Overview. This report shows the Conversion Rate and Revenue for all visits and visits that clicked on 4-Tell recommendations (shown below with arrows).

Page views & Pages/Products per Visit KPIs

To find the third and fourth KPIs, you view Audience > Overview on the left, and look at the Page views stat and pages (i.e. products) per visit stat (shown below with the arrows).

Why These Key Indicators

Now that you understand the power of Google Analytics, you can quickly be overwhelmed with all the different ways you can slice and dice the data being captured from your website. To help you get started, we’ve identified four key indicators to best analyze the effectiveness of implementing 4-Tell recommendations. Below find out why we chose these specific KPIs:

1. Influenced Revenue = Percentage of revenue from shoppers who click on 4-Tell

This is a logical first indicator and a great first step to see how much of your topline revenue is influenced by recommendations. 80% of the revenue influenced by recommendations is an increase in sales. (See Influenced Revenue section above for details.)

For a detailed analysis of sales, A/B testing is needed. In A/B testing, some visitors see recommendations (group A) and some visitors don’t (group B), and the results are compared. If you are interested, we offer Professional Services and are happy to setup this test for a minimal fee.

Note: You should not compare revenue before recommendations and after recommendations since seasonal trends can influence sales, such as an increase before holidays, decrease after holidays or during the summer, and so on. In addition, promotions or price changes can have short term affects that mask the effect of recommendations.

2. Increase in conversion for shoppers who click on recommendations

We know you spend a tremendous amount of time and money driving people to your website. Anything that can help convert more browsers to buyers will ultimately increase the return on investment. This metric shows definitively how recommendations are working to increase your conversion rate.

This increase in conversion provides a multiplier effect on marketing budgets, especially search engine marketing (SEM). Many retailers see a 25% increase in overall conversion, which means that your marketing dollars are create 1.25 times more revenue for your store.

3. Percentage of page views is from shoppers who clicked on recommendations

You can view the percentage of shoppers that click on recommendations, but we don’t believe this to be as useful an assessment as page views. We know shoppers that click on recommendations view more pages, which really means they view more products. By showing them relevant products, shoppers tend to want to see more, all of which ultimately lead them to buy more. This is why we believe that the number of page views by shoppers that click on recommendations is a more accurate reflection of the effect of recommendations.

4. Pages (i.e. Products) viewed per visit for shoppers who click on recommendations

A basic retailing tenet is the more products a shopper views, the more likely they are to buy something. This also holds true in the online world, so this one makes our top four KPI list too.

Along with viewing more products, a shopper stays on your site for a longer time. This actually impacts your overall online marketing strategy. We believe Google measures time on site as part of its page rank for search order. Thus, if people spend longer on your site, you may be ranked higher in the organic search results.

5. Average Order Value (Be Careful)

You may be wondering why we are not suggesting evaluating the average order value (AOV) as a key indicator. Of course this is an important metric. We analyzed 89 of our clients over a 4 month window: 2 slow months and 2 peak months. On average, these clients see an increase of 2% in their AOV.

Unlike conversion, which always increases, AOV can increase or decrease for these customers. This happens since some shoppers who would have bought nothing now buy something with recommendations. As such, you’ll have some additional smaller orders pulling AOV down, and some add-on products pushing AOV up, potentially cancelling each other out and therefore not dramatically changing the AOV. As such, we caution you to not over analyze the AOV.

Advanced Usage for Web & Email

You can use Google Analytics to create advanced reports so that you can determine the effect of different page types, different recommendation types, e .g. cross-sell on product detail pages, similar on the product detail pages, and cross-sell in the cart, and the position of the recommendation in the block. We provide a brief description below, and Google Analytics has extensive help.

Event Category = 4TellRecsWeb/Email

In our plugin, we setup ‘Event Category’ as ‘4TellRecsWeb’ or ‘4TellRecsEmail’. Using these parameters is discussed above.

Event Action = Email/Page Type & Recommendation Type

We’ve also setup ‘Event Actions’ to show page type and recommendation type, or email type and recommendation type. When a shopper clicks on a recommendation on a web page, some example event actions are ‘ProductDetail-CrossSell’, ‘Home-Personal’, and ‘Category-CategoryTopSellers’. When a shopper clicks on a recommendation in an email, some example event actions are ‘OrderConfirmation-CrossSell’, ‘Newsletter-Personal’, and ‘AbandonedCart-Similar’. The full list of web page types, email types and recommendation types are listed at the end of this section.

You create a custom segment as described in the Creating Custom Segments section above, except you use ‘Event Action’ rather than an ‘Event Category’, and then choose that the action contains the page type and/or recommendation type, or email type and/or recommendation type. You can select several segments to create a report that compares several recommendation types.

Web Page Types

The web page types are:

ProductDetail, QuickCart, AddToCart, Category, Home, Search and Admin.

Email Types

The email types are:

OrderConfirmation, OrderStatus, ShippingConfirmation, OrderPartiallyShipped, OrderRMA, OrderReturned, ProductReview, OrderBackordered, OrderHold, OrderCancelled, OrderStatus, AbandonedCart, BetterPrice, EmailFriend, InStock, GiftCertificate, Newsletter, Other, RatingRequest, ReorderNotice, WelcomeNotice, and DailyDeal.

Recommendation Types

The recommendation types are:

CrossSell, Similar, Personal, TopSellers, and CategoryTopSellers.

Event Label = Product Names

You can track which products are clicked as recommendations using ‘Event Labels’. In our plugin, we’ve already setup the product name as the event label. You can create a segment where you add ‘Event Label’, then select the product name and can view reports segmented by that product (or a list of products using the ‘or’ function in segments).

Event Value = Position

You can track by the position or position range of the recommendation using ‘Event Value’. The integer value of the position is set as the Event Value, and custom segments can be setup using event value in a similar method as previously described.

Leave a Reply

Your email address will not be published.