Retargeting
Retargeting is a form of online advertising that delivers ads to people who have previously interacted with a business, website, app, product, or piece of content. Ads may reach someone who viewed a particular offer, added a product to their cart, started completing a form, or performed another defined action.
The purpose of retargeting is to remind people about an offer and encourage them to return, resume an interrupted process, or take the next step. Because the audience already knows the brand, the message can be more relevant than an advertisement shown to entirely new prospects.
How does retargeting work?
Retargeting uses information about a person’s previous interactions. This information may come from a website, mobile app, social media platform, product catalog, or customer list.
A typical retargeting process involves:
- a person visiting a website or completing another type of interaction;
- recording an event such as viewing a product or adding it to a cart;
- assigning the person to an appropriate audience;
- launching a campaign for that audience;
- showing a relevant advertisement through a supported channel;
- measuring return visits and completed conversions.
Advertising tags, cookies, app identifiers, server-side events, and information provided directly by customers may all be used to record activity. Available data depends on consent choices, privacy settings, browser restrictions, and the advertising platform’s ability to match the person with the information provided.
Retargeting vs. remarketing
The terms retargeting and remarketing are often used interchangeably. Both describe efforts to reconnect with people who have previously interacted with a business.
Under a narrower definition, retargeting primarily refers to paid advertising based on a person’s previous activity on a website, in an app, or within another digital environment. Remarketing may have a broader meaning that includes email, notifications, marketing automation, customer communication, and campaigns using data from a CRM system.
There is no single definition used by every platform, agency, and industry publication. In practice, it is more important to define:
- the data source;
- the conditions for joining the audience;
- the advertising channel;
- the membership duration;
- the campaign objective;
- the method used to measure results.
What data can retargeting use?
Retargeting audiences can be created from several types of signals. Common examples include:
- pages visited;
- products or services viewed;
- products added to a cart;
- checkout activity;
- forms that a person started completing;
- catalogs or resources downloaded;
- videos watched;
- interactions with a profile or advertisement;
- mobile app activity;
- previous purchases;
- customer status recorded in a CRM;
- membership in a customer or subscriber list.
Recorded information should not automatically be used for advertising personalization. Businesses must consider applicable laws, information provided to users, consent choices, platform policies, and restrictions involving sensitive data.
Types of retargeting
Retargeting can be categorized by its data source, advertising channel, or the way the message is personalized.
Common types include:
- website visitor retargeting;
- mobile app retargeting;
- dynamic retargeting;
- social media retargeting;
- display network retargeting;
- search retargeting;
- video retargeting;
- customer list retargeting;
- sequential retargeting;
- cross-channel retargeting.
One person may encounter several forms of retargeting. Someone might first see a product advertisement on social media, later receive a video ad, and eventually see a message related to an order they previously started.
Website visitor retargeting
Website retargeting reaches people who visited a site and met specified conditions. Instead of using one campaign for every visitor, a business can create segments that reflect different behaviors.
Example audiences include people who:
- visited any page on the website;
- viewed a particular service;
- browsed a specific product category;
- viewed a product without adding it to the cart;
- added a product to the cart without starting checkout;
- started checkout without placing an order;
- visited a pricing page;
- downloaded a resource without submitting an inquiry;
- purchased within a specified period;
- did not complete the primary conversion.
A person who read one educational article usually has a different level of intent than someone who visited a service page several times and started completing a contact form. Audience segmentation should reflect these differences.
Dynamic retargeting
Dynamic retargeting automatically adapts an advertisement to products or offers that a person previously viewed. The system uses a catalog or feed containing structured information about the available inventory.
A dynamic advertisement may display:
- an image of the viewed product;
- the product or service name;
- the current price;
- availability information;
- similar options;
- complementary products;
- a direct link to the relevant page.
Product identifiers must remain consistent across the website, advertising events, and catalog. If the website sends an identifier that does not match the item stored in the feed, the platform may be unable to recognize the viewed product or may display an incorrect offer.
Dynamic retargeting is not limited to online stores. A catalog can also contain hotels, real estate, vehicles, flights, courses, job listings, and other structured service offerings.
Sequential retargeting
Sequential retargeting presents advertising messages in a defined order. Instead of repeatedly showing the same creative, a business can gradually provide information that supports the buying decision.
An example sequence may include:
- a reminder about the viewed product;
- an explanation of its main benefits;
- customer reviews or examples of completed work;
- information about shipping, warranties, or returns;
- an invitation to schedule a consultation or return to the cart.
Each stage should account for the person’s subsequent actions. After a purchase, ads encouraging the customer to complete an order should stop. The customer may instead enter a separate campaign involving a complementary product or post-purchase support.
Retargeting in Google Ads
Google Ads can use information about people who previously visited a website, used an app, watched video content, or were matched through customer data.
Google currently uses the term “your data segments” for many audiences that were previously described as remarketing lists. Depending on the supported campaign type, these segments may be used for targeting, performance observation, exclusions, or providing automated systems with signals about valuable audiences.
The available options depend on the campaign type, data source, audience size, and current platform policies. Not every audience can be used in every campaign.
Retargeting in Meta Ads
Meta allows advertisers to create Custom Audiences from website activity, app events, catalog interactions, customer lists, and engagement across Meta technologies.
Retargeting through Facebook Ads may use:
- events recorded by the Meta Pixel;
- events sent through the Conversions API;
- activity within an app;
- interactions with a company profile;
- video engagement;
- lead form activity;
- interactions with advertisements;
- products viewed in a catalog;
- an appropriately prepared customer list.
Sending events through a server-side integration may improve data quality and measurement continuity, but it does not bypass privacy requirements or user choices.
Retargeting for e-commerce
In e-commerce, retargeting can support customers at several stages of the purchase process. A campaign may remind someone about a viewed product, an abandoned cart, or a payment process they started.
Common scenarios include:
- advertising a viewed product;
- presenting similar products;
- reminding shoppers about items in their carts;
- reaching people who started checkout;
- announcing that a product is back in stock;
- promoting complementary products;
- encouraging another purchase of a consumable item;
- reactivating inactive customers;
- excluding buyers from ads for a product they already purchased.
Performance also depends on catalog accuracy, pricing, availability, image quality, shipping terms, and the experience after a person clicks the ad. Retargeting cannot repair a broken shopping cart, a failed payment process, or an unclear offer.
Retargeting for services and B2B customers
Service businesses and B2B companies can also use retargeting. These purchases often involve longer decision-making processes, several website visits, consultations, offer comparisons, and approval from other people within an organization.
An advertisement may remind prospects about:
- a particular service;
- scheduling a consultation;
- a catalog or report they downloaded;
- a webinar;
- a case study;
- an application deadline;
- requesting an estimate;
- a solution to a problem they previously researched.
In B2B marketing, the number of submitted forms may not be enough to evaluate campaign performance. Lead quality, proposals, closed contracts, and customer value should also be considered. Data from later stages of the sales process may come from a CRM system.
How should retargeting audiences be segmented?
Segmentation allows a business to adapt its advertising message, budget, and campaign duration to each person’s previous behavior.
Audiences may be segmented by:
- pages visited;
- products and categories viewed;
- events completed;
- stage in the purchase process;
- time since the last visit;
- number of visits;
- cart value;
- purchase history;
- customer type;
- content engagement;
- original traffic source;
- customer value.
An audience should not be so broad that it combines people with entirely different needs. It should also not be so small that the platform cannot deliver the campaign effectively or repeatedly shows ads to the same limited group.
Audience membership duration
Membership duration determines how long a person remains in an audience after completing a specified action. It should reflect the length of the decision-making process and the nature of the offer.
A person who abandoned a cart containing an everyday product may be most valuable during the first few days. Purchasing a vehicle, business software, or an expensive professional service may take several weeks or months.
A business can create separate audiences for recent visitors, people who visited several weeks ago, and customers who have been inactive for months. Each group can receive a different message.
How should a retargeting ad be tailored?
An advertisement should reflect the person’s previous activity and provide a specific reason to return. A generic message asking the person to visit again may not provide enough value.
A retargeting ad may:
- remind the person about a viewed product or service;
- explain an important benefit;
- address a common customer concern;
- present a review or completed project;
- explain shipping, return, or warranty terms;
- show similar or complementary products;
- invite the prospect to schedule a consultation;
- communicate a genuine promotion or deadline;
- lead to content that supports the decision.
The call to action should clearly explain the next step. The advertisement must also match the page that appears after the click.
Where should a retargeting ad send users?
The homepage is rarely the best destination for every retargeting advertisement. A more relevant destination may be a product page, shopping cart, booking form, service page, comparison page, or dedicated landing page.
The destination should:
- match the advertisement;
- allow the person to resume the process;
- work correctly on mobile devices;
- load quickly;
- display current prices and terms;
- make the next action clear;
- avoid forcing users to repeat unnecessary steps;
- remain accessible to people using different interaction methods.
Retargeting ad frequency
Frequency indicates how many times an individual sees an advertisement during a specified period. Excessive delivery may cause ad fatigue, annoyance, and negative perceptions of the brand.
Warning signs may include:
- rapidly increasing frequency;
- a declining click-through rate;
- a rising cost per result;
- negative user reactions;
- using the same creative for an extended period;
- continuing to reach people after their need has ended.
There is no single correct frequency limit for every campaign. Audience size, channel, buying cycle, number of available creatives, and the intensity of other marketing communication all matter.
Exclusions in retargeting
Exclusions prevent ads from reaching people for whom the message is no longer appropriate. They can reduce unnecessary spending and improve the audience’s experience.
A campaign may exclude:
- people who completed the promoted conversion;
- buyers of the advertised product;
- customers who already have the advertised service;
- users who opted out of communication;
- company employees;
- people outside the service area;
- users assigned to another campaign;
- prospects who have entered a later stage of the sales process.
A purchase does not need to end all advertising communication. The customer may enter a separate campaign for a complementary product, service renewal, or the next stage of the relationship.
Retargeting vs. new customer acquisition
Retargeting reaches people who have already interacted with a business. Prospecting campaigns are designed to find new audiences.
An audience modeled after existing customers is not a retargeting audience if it contains people who have never interacted with the brand. Information about valuable customers may still help the advertising system identify new prospects with similar characteristics.
A business should not dedicate its entire budget to retargeting. If acquisition campaigns stop bringing new people into the funnel, the number of users available for retargeting will eventually decrease.
How is retargeting implemented?
The implementation process depends on the website, advertising platforms, and planned campaign scenarios. It may involve:
- defining the campaign objective;
- selecting the behaviors that should create audiences;
- implementing the required tags or integrations;
- configuring consent management;
- preparing a catalog or feed;
- ensuring identifier consistency;
- sending the required event parameters;
- creating audiences and exclusions;
- setting membership durations;
- preparing advertising creatives;
- selecting destination pages;
- configuring conversion tracking;
- testing the implementation before launch.
Testing should cover more than the presence of an advertising tag. Events need to fire at the correct time, avoid duplicate recording, send accurate parameters, and respect the person’s consent choices.
Retargeting, cookies, and privacy
Retargeting may involve storing identifiers, observing user activity, building advertising profiles, and transferring information to third-party platforms. Businesses must consider the laws that apply in each advertising market and the policies of the systems they use.
In the European Union, cookies and similar technologies used for behavioral advertising generally require prior user consent. Privacy information should clearly explain the purposes, tools, types of data involved, and the method for withdrawing consent.
An appropriate implementation may require:
- blocking relevant technologies until consent is obtained;
- recording and respecting user choices;
- making consent withdrawal easy;
- limiting the amount of information collected;
- setting appropriate data retention periods;
- controlling access to audiences and customer lists;
- updating privacy and cookie disclosures;
- following restrictions involving sensitive data;
- regularly reviewing vendor configurations.
A server-side integration does not eliminate consent and data protection obligations. It changes how information is transmitted, but it does not mean that data can be collected or used without evaluating its source, purpose, and lawful use.
How is retargeting performance measured?
The relevant metrics should reflect the campaign objective. Evaluation may include:
- reach;
- impressions;
- frequency;
- click-through rate;
- cost per click;
- conversion volume and conversion rate;
- cost per conversion;
- revenue and profit margin;
- average order value;
- ROAS;
- ROI;
- recovered carts;
- lead quality;
- customer value;
- assisted conversions;
- view-through conversions.
Results should be evaluated separately for different audience segments. A campaign targeting cart abandoners may perform very differently from one targeting every visitor to a company’s blog.
Retargeting and conversion attribution
Retargeting audiences already know the business and are often closer to making a decision than new users. A campaign may therefore report strong results even when the advertisement was not the only reason for a purchase.
Before converting, a person may have arrived through organic search, paid advertising, email, social media, and direct visits. Assigning all value to the last click can overlook earlier stages of the journey.
View-through conversions require particular caution. Seeing an advertisement before purchasing does not prove that the transaction would not have occurred without retargeting.
The campaign’s incremental impact can be evaluated more accurately through control groups, experiments, and conversion lift analysis. The availability of these methods depends on the platform, campaign type, and audience size.
Limitations of retargeting
Advertising audiences do not include every person who completed a particular action. Their size and quality may be affected by:
- lack of consent for advertising cookies;
- user privacy settings;
- script and ad blocking;
- browser and operating system restrictions;
- people using multiple devices;
- an inability to match the available data;
- incorrect event configuration;
- insufficient audience size;
- a short membership duration;
- platform policies affecting particular industries or data categories.
The number of people reported by an advertising platform may therefore differ from the number of users meeting similar conditions in an analytics system.
Common retargeting mistakes
Common problems include:
- showing the same advertisement to every website visitor;
- failing to distinguish between stages of the purchase process;
- not excluding people who already converted;
- advertising unavailable products;
- using inconsistent identifiers across the website and catalog;
- allowing frequency to become excessive;
- using one creative for too long;
- setting an inappropriate membership duration;
- sending every ad click to the homepage;
- creating a mismatch between the advertisement and destination page;
- building every campaign around a discount;
- recording events incorrectly or more than once;
- activating advertising tags without accounting for consent;
- sending information that the platform should not receive;
- evaluating results only through last-click attribution;
- treating retargeting as a substitute for new customer acquisition.
When should a business use retargeting?
Retargeting can be particularly useful when:
- customers rarely purchase during their first visit;
- the decision-making process takes time;
- users frequently abandon carts or forms;
- the offer requires comparison and additional information;
- the website generates enough traffic;
- the company has a valuable customer database;
- repeat purchases are possible;
- the business sells complementary products or services;
- event and conversion tracking works correctly.
Retargeting may be less valuable when the audience is extremely small, purchases almost always happen immediately, additional communication provides no benefit, or legal and platform restrictions prevent appropriate use of the available data.
Effective retargeting is not about following a person everywhere with the same advertisement. It should deliver a useful and relevant message at a time when it can help the person resume an interrupted process or make an informed decision.