<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Matthew Michaelson on Medium]]></title>
        <description><![CDATA[Stories by Matthew Michaelson on Medium]]></description>
        <link>https://medium.com/@matthewmichaelson?source=rss-44cc563f4c3f------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*psSbNy6UhdEZW4WX.</url>
            <title>Stories by Matthew Michaelson on Medium</title>
            <link>https://medium.com/@matthewmichaelson?source=rss-44cc563f4c3f------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sun, 22 Nov 2026 10:18:08 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@matthewmichaelson/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[The Truth About Retargeting and How to Think More Strategically About It.]]></title>
            <link>https://medium.com/@matthewmichaelson/the-truth-about-retargeting-and-how-to-think-more-strategically-about-it-a323fc3c36c7?source=rss-44cc563f4c3f------2</link>
            <guid isPermaLink="false">https://medium.com/p/a323fc3c36c7</guid>
            <category><![CDATA[advertising]]></category>
            <category><![CDATA[facebook-marketing]]></category>
            <category><![CDATA[growth-marketing]]></category>
            <category><![CDATA[retargeting]]></category>
            <category><![CDATA[growth-hacking]]></category>
            <dc:creator><![CDATA[Matthew Michaelson]]></dc:creator>
            <pubDate>Sat, 09 Jul 2016 13:51:05 GMT</pubDate>
            <atom:updated>2016-07-09T14:40:54.823Z</atom:updated>
            <content:encoded><![CDATA[<p>Retargeting, the act of serving ads to recent site visitors, yields great return on investment. That’s what we’re told. And your agency is probably showing you fantastic numbers. I’ve heard far too many account managers tell clients that they’ve “returned 10x on retargeting ad spend,” The truth is, the numbers aren’t as good as they seem.</p><p>To begin to approximate what the true return on retargeting ad spend is, let’s think about how to structure those retargeting campaigns and why. There are two characteristics of your site visitors that determine the effectiveness of a retargeting campaign (ignoring the ad creative). The first is how recently they visited the website and the second is how much intent they signaled while on the site.</p><p>To Let’s take a few examples. If someone came to your homepage and didn’t visit any other pages a month ago, they probably aren’t very interested in what you’re offering and showing them an ad now might be fruitless. On the other hand, if someone came to your site, visited five pages, added a product to their cart yesterday, showing them an ad might be pretty lucrative. Marketers know they only have to push them a little bit further and make sure they don’t forget the product/service/company in order to convert the prospective customer. Maybe they offer them free shipping.</p><p>The architecture of your retargeting campaigns should take into consideration both of these characteristics. Take a look at the graph below. On the y axis we have demonstrated intent, with visitors who demonstrated high intent at the bottom and low intent at the top. On the x axis we have recent visitors starting at 0 days, going outward day by day. Each of the points on the graph represents a site visitor.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*4E1SOhqC8-iCa5Mf9Sz9iA.jpeg" /></figure><p>Now let’s mark this chart up with different targeting groups. We might have one ad set (to borrow Facebook terminology but same concept applies to any other medium) that targets people who added to cart (a strong intent signal) in the past 15 days and excludes anyone outside of those 15 days. We could have another that targets people who added to cart from 15–30 days and another for folks who didn’t add to cart but did make it to a product page in those same time periods. As we break these into smaller and smaller pieces we’re able to control exactly who sees ads, what they see, and how much they see it.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pRVkEEBvp1O3U1I59__0Zg.jpeg" /></figure><p>The closer these ad sets and the visitors they represent, are to the intersection of the x and y axis (high intent, very recent) the higher the return on ad spend is expected to be. By breaking these ad sets into different groups, we can decide which group of people we’d like to target based on the expected return of serving them ads and which people are better left unserved so we can allocate budget in places with higher expected return.</p><p>The just visited high intent group looks like the best performer and we’re seeing the most conversions come in as a result of those ads. However (big however) aren’t those people the ones who were most likely to come back and make a purchase anyways? The visitors who made it to a product page but failed to add anything to their cart were probably less likely to come back anyways and make a purchase, so perhaps the true return on ad spend was greater with this group. Perhaps the visitors who didn’t even make it to a product page are the ones who really need to see a compelling message to re-engage their interest. If you’re buying in to this logic, you can see that we really have no idea where our retargeting dollars are most efficiently spent.</p><p>There’s a solution to this problem. In order to approximate our true return on ad spend we can take a portion of our site visitors and serve them no retargeting ads at all. This is called a hold-out test. We can then take a look and see how many of those people who were not served ads ended up coming back and making a purchase anyways and compare it to the population that was served ads. The delta is the impact created by those retargeting ads.</p><p>If you aren’t in a position to run these tests, keep a close eye on attribution windows (the amount of time that can elapse since a prospective customer saw an ad before a conversion can no longer be taken credit for by that ad) and choose a conservative time frame. If you choose something like like 1 day, and not the 30 days your agency is likely using, you can be more confident that the conversions you are reporting were actually a result of the ads you served. This may be overly conservative and you’ll need to keep that in mind as well.</p><p>The truth is, the retargeting metrics that your agency pedals out are crap. They barely mean anything. There are far too many considerations to solely consider the number of people who saw an ad after coming to your site and made a purchase. If you want to maximize your marketing budgets, run some hold-out tests and get closer to discovering the true return on ad spend. Don’t trust data without context. Data &gt; conventional wisdom.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a323fc3c36c7" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>