A step-by-step guide to RLSA and dynamic remarketing

We’ve all spent time window shopping and canvassing for the best price or service quality available before buying online. Whether for a small ticket item, or a once-in-a-blue-moon high-end purchase. 

This natural tendency to delay our purchase decision means it is highly important for marketers and business owners to tailor marketing strategies to ensure that potential customers and leads are carefully nurtured.

The practice of remarketing has become increasingly common over the last couple of years, and a number of leading display marketing companies have developed some pretty advanced, highly tailored solutions for their customers.  

These are already helping many businesses achieve prolonged engagement with users who have shown intent on a given site.

But over the past few weeks Google has released some extremely useful new tools which help get closer to the competition’s offering, and open up sophisticated features to advertisers – no matter what the budget size.

Today I want to introduce two Google products, talk a little about the opportunities provided, and give a straight-forward outline of how to get started.

So, first up we have a brand new acronym, RLSA. Or to give it it’s Sunday name:

Remarketing Lists for Search Ads

The RLSA feature helps you use previous site interactions to improve the relevancy, and ROI, of your search ads.  

To give an example – say a number of users arrive at your site and view category pages containing blue suede shoes – add them to the basket, but leave the site without making a purchase.  This information can be used to customise your RLSA strategy for targeting those users through the purchase funnel.

Your RLSA campaign could identify them in a number of different ways on future searches – perhaps by using more generic keywords that prove profitable in your ‘standard’ campaign, or bidding more on the keywords you already advertise with.  

The main theory here is that the user has shown intent to purchase, and although further advertising investment is required to keep them engaged, the fact that they are further through the purchase cycle and have brand awareness and engagement, increases the chance of converting the sale.

Implementation outline

1. Add the remarketing tag to your site

Navigate to the ‘Shared Library > Audiences’ section of your AdWords account, in the not-always-obvious left-hand navigation:

Audience location in AdWords

If this is your first remarketing campaign you will need to generate the Remarketing tag – on the audiences page, select the ‘Set-up remarketing’ option for now. The other option is covered later in this post (if I can entice you to continue onwards!).

You’ll be invited to email the set-up details to the relevant person. These details include a tag to be placed before the closing </body> tag on every page of the site (further technical instructions are included in the email itself).

When you have completed this step, your first ‘audience’ will have been created, called ‘All Visitors’. When talking about a remarketing ‘audience’, this just means a group of visitors who share a similar trait – such as having viewed a certain page or completed an action like ‘add to basket’.  

This audience will contain every visitor to the site (assuming all pages are tagged).

The 'All Visitors' audience

All future audiences you create will be added to the list, with the level of detail shown above.

2. Add the basic audience to an ad group

To utilise the audiences that are being created, they need to be assigned to an ad group, telling the group which set of people to specifically target via RLSA.  As an example let’s do this with the ‘All visitors’ audience. 

Create a campaign as per normal, with keywords, and ads (quite possibly a duplicate of an existing campaign you already have) – this helps separate your standard search strategy from your RLSA strategy.

If these tabs/options are still not visible, you will need to ensure the campaign type is set to ‘Search Network Only – All features’.

Hit the ‘Add remarketing’ button and you will have the option of choosing any audience.  Choose the ‘All visitors’ group and save:

This ad group is now set-up to retarget your previous visitors on the search results page.  You can tailor your strategy based on this targeting, for example, change the call-to-action in the ad copy, increase the maximum bid, or add a range of keywords not included in the standard campaign.

3. Create a more tailored audience

The real benefit of RLSA campaigns will only be seen once you start to create segments within your user base, and build strategies around them.  As an example of how to do this, let’s look at creating a remarketing group for users who have read content in the news section on the QueryClick site.

In the image below we are choosing to create a new Remarketing List, then defining the members of the list as those users who have viewed urls containing /seo-news/:

More complex segments can be achieved by creating a ‘Custom Combination’ from your audiences.  For example, you may damage the relationship with existing customers if you heavily remarket them – creating a custom combination to ‘remove’ users who have viewed the purchase confirmation page safeguards against this situation.

In this way you can add and remove users based on whole range of url/referrer combinations until you get segments you feel will be effective additions to your marketing strategy.  

Now you’ve been armed with the basic instructions for getting all set-up, it’s over to you to determine the best use-case for your business goals.

Dynamic remarketing

Similar to RLSA, dynamic remarketing is all about engaging further with users based on the content they’ve viewed on site.  

However, the main differences here are:

  • this is more traditional remarketing territory, where you will be reaching users via the Google Display Network
  • the targeting groups are created dynamically based on your Google Merchant Centre (GMC)

In short, Google will combine visitor data with information (prices, images and descriptions) from your GMC to create highly targeted (e.g. product specific) image ads.

Features of these ads include:

  • Optimised layouts: using performance data to determine which layout to use.
  • Product recommendations: showing users how products relate to one another.
  • Customisation: tailoring logos, text and colours to ensure all ads fit with brand guidelines.

This tool has been in beta for a while, and through a combination of ease of use and early performance, is proving to be a very valuable addition to AdWords.

Implementation outline

Lets have a look at the basic steps required for set-up:

1. Google Merchant Centre

Ensure you have a product feed from your site live within GMC (if not, then you can read all about how to get started here).

Link your AdWords account in GMC.  In the Settings area, select AdWords and just input your AdWords ID number.

2. Create a campaign in AdWords

Create a new campaign as you normally would, but when selecting the ‘Type’ make sure you choose Remarketing.

In the section below, you will be given the option to use your GMC set-up to tailor the ads.  Assuming this is the first campaign of this type in your account, the option to ‘Set up dynamic remarketing’ is given – follow this link through.

You will now be taken step by step through the process of getting the custom tags onto your site.

  1. Sending the code: whether it’s yourself or a developer making the changes, just add in the relevant email address..
  2. Starter remarketing lists are automatically created for you.  It’s worth noting that information such as buyers and abandoners will come from page-type data passed through the tags.
  3. Continue back to the campaign and set-up your budgets as normal.
  4. Follow the instructions provided on screen to create an ad group which contains relevant remarketing lists (audiences).
  5. Use the handy ad generator to create the customised adverts. There are many ready made templates, so it’s definitely worth spending some time reviewing options and playing around with variations.

3. Add remarketing tag and custom parameters to your site

Most of the nitty-gritty development details are contained within the email sent to the address added in the above steps (and here also), so we won’t linger here for too long – only just to give an overview of the tags themselves.  

The tag should be placed on every page of the site, and is similar to a standard remarketing tag, but with three custom parameters which will allow ads to be matched from the product feed.  This part of the code is:

<script type=”text/javascript”>

var google_tag_params = {

ecomm_prodid: ‘REPLACE_WITH_VALUE’,

ecomm_pagetype: ‘REPLACE_WITH_VALUE’,

ecomm_totalvalue: ‘REPLACE_WITH_VALUE’,



So, all we need the site to do is pass back:

  1. Product ID
  2. Page type (home, category, cart etc) – every page must be categorized.
  3. Total value for product pages, the product price; for cart pages, the basket amount

Once the developers have  tested the implementation using the Chrome ‘Tag Assistant’ extension recommended in the set-up email, you should be ready to go ahead and create the ads via the ‘wizard’ provided by Google.


In over eight years of managing AdWords campaigns, I’ve not known a year in which so many major updates have been rolled out.  

There was a bit of kickback with the roll-out of enhanced campaigns due to the removal of some targeting features, but now we are seeing a good number of really big improvements which provide opportunities to get far more in-depth and strategic when advertising with AdWords.  

RLSA and Dynamic remarketing are two really useful additions, that once you are familiar with, will potentially become staples in your PPC toolbox.

Via Econsultancy

Copenhagen INK

Lars M. B. Anthonisen is Global Account Lead @ Google. Previously, he held various digital marketing positions at media companies across Europe and Asia including Regional Digital Director at MediaCom APAC, CMO at Adform and Digital Manager at Universal McCann Worldwide.

You may also like...