{"id":191,"date":"2023-02-14T21:32:36","date_gmt":"2023-02-14T21:32:36","guid":{"rendered":"https:\/\/www.ridgeline-analytics.com\/?p=191"},"modified":"2024-05-05T23:36:42","modified_gmt":"2024-05-05T23:36:42","slug":"b2b-revenue-attribution-for-sem-with-tableau-python-and-machine-learning","status":"publish","type":"post","link":"https:\/\/www.ridgeline-analytics.com\/index.php\/2023\/02\/14\/b2b-revenue-attribution-for-sem-with-tableau-python-and-machine-learning\/","title":{"rendered":"B2B Revenue Attribution for SEM with Tableau, Python and Machine Learning."},"content":{"rendered":"<p>A common challenge for B2B marketers is how to attribute revenue to a search program. Unlike typical direct marketing tactics, the path to B2B revenue often involves many channels, programs, stakeholders and decision makers.<\/p>\n<p>There&#8217;s usually not a direct path from a specific digital touch to a sale, which can make long-term SEM programs difficult to justify with revenue contribution. This leaves search marketers with but clicks, search rankings and impressions, which rarely tell a compelling story about marketing&#8217;s contribution to revenue.<\/p>\n<p>A client approached us with an interesting proposal: he wanted to demonstrate SEO&#8217;s contribution to pipeline and sales. With no direct connection between search impressions, clicks, and ultimate a sale, we had to come up with a solution that relied on statistical inference.<\/p>\n<p>Our solution was to blend a few data sources:<\/p>\n<ul>\n<li>Web traffic and conversion data from Adobe Analytics which contained a unique customer identifier on each page as well as each URL touched by that unique customer. This identifier was also passed to salesforce anytime a web conversion occurred, i.e. a form submission.<\/li>\n<li>A CSV file of managed SEO terms and preferred landing page URLs from a vendor tool.\n<\/li>\n<li>CRM data from Salesforce.com, which included a unique customer identifier along with all revenue activity. This allowed us to get a file which had a URL and the last-click revenue associated with that URL.\n<\/li>\n<li>Google Search Console data, which allowed us to view search entrances per URL.<\/li>\n<\/ul>\n<p>The basic application logic worked like this:<\/p>\n<ul>\n<li>Map all the data sources together using the unique customer key and load into a temp working source.\n<\/li>\n<li>Grab the top keywords, click and impression data for the URL via the Google Search Console REST API into another working source\n<\/li>\n<li>Classify the keywords into brand, non-brand, product, and service categories using machine learning algorithms. We used TextBlob and FuzzyWuzzy python libraries for this.\n<\/li>\n<li>Visualize and analyze the data using an analytics platorm (<a href=\"https:\/\/www.tableau.com\/\">Tableau<\/a>).<\/li>\n<\/ul>\n<p>We now could drill down into each search query and see the revenue opportunities associated with it.&nbsp; This made it easy to identify revenue opportunities in an interactive, visual way.<\/p>\n<p><\/p>\n\n\n<p><\/p>\n","protected":false},"excerpt":{"rendered":"<p>A common challenge for B2B marketers is how to attribute revenue to a search program. Unlike typical direct marketing tactics, the path to B2B revenue often involves many channels, programs, stakeholders and decision makers. There&#8217;s usually not a direct path from a specific digital touch to a sale, which can make long-term SEM programs difficult &hellip; <\/p>\n<p class=\"link-more\"><a href=\"https:\/\/www.ridgeline-analytics.com\/index.php\/2023\/02\/14\/b2b-revenue-attribution-for-sem-with-tableau-python-and-machine-learning\/\" class=\"more-link\">Continue reading<span class=\"screen-reader-text\"> &#8220;B2B Revenue Attribution for SEM with Tableau, Python and Machine Learning.&#8221;<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":[],"categories":[12,7,10],"tags":[],"_links":{"self":[{"href":"https:\/\/www.ridgeline-analytics.com\/index.php\/wp-json\/wp\/v2\/posts\/191"}],"collection":[{"href":"https:\/\/www.ridgeline-analytics.com\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.ridgeline-analytics.com\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.ridgeline-analytics.com\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.ridgeline-analytics.com\/index.php\/wp-json\/wp\/v2\/comments?post=191"}],"version-history":[{"count":4,"href":"https:\/\/www.ridgeline-analytics.com\/index.php\/wp-json\/wp\/v2\/posts\/191\/revisions"}],"predecessor-version":[{"id":329,"href":"https:\/\/www.ridgeline-analytics.com\/index.php\/wp-json\/wp\/v2\/posts\/191\/revisions\/329"}],"wp:attachment":[{"href":"https:\/\/www.ridgeline-analytics.com\/index.php\/wp-json\/wp\/v2\/media?parent=191"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.ridgeline-analytics.com\/index.php\/wp-json\/wp\/v2\/categories?post=191"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.ridgeline-analytics.com\/index.php\/wp-json\/wp\/v2\/tags?post=191"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}