Upstream Analytics: Know the URLs that people visit before converting

 In Behavioral Analysis, Consumer Behavior, Featured, Product News, Traffic Sources, Uncategorized

Do you want to know the URLs that people visit before converting on your website? Or, the URLs that people visit before converting on your competitor’s website? If so, then you want Upstream Analytics.

To start using Upstream Analytics, you define an online action that you’re interested in, such as a purchase or a subscription on a specific domain. The domain can be one that you own, a competitor’s, or another one of interest.

From there, Jumpshot provides you with aggregate counts of every URL that your audience visited prior to taking the designated action. These URLs go back in time over a lookback period that you set, such as 7, 14, or 21 days. This results in a collection of URLs that you can use to analyze where your audience has been, what kind of content they are consuming, and what sites are engaging them. This data will help you make better decisions about how to reach, influence, and engage with them online.

Here’s a few questions and answers about Upstream Analytics that will get you thinking about how to use it for your business.

What online actions should be investigated?

Typical web analytics solutions like Google and Adobe provide referring URL data for your own websites, but they do not provide information about where people were further upstream. Nor do they provide access to this data for other companies. The true power of Upstream Analytics data comes alive when you look at all the activities happening online before a specific user behavior.

Start with purchases. Every company can learn something by looking at the upstream URLs prior to a purchase.

To illustrate this, Jumpshot recently evaluated the online behavior prior to a purchase on and also on

For Virgin America, Jumpshot found that 23% of its online purchasers visited a competitor’s airline site prior to purchasing, 18% visited an online travel agency (OTA) site, and 9% visited a meta search site such as KAYAK or trivago. Virgin America can now use this insight to refine its competitive positioning, foster deeper relationships with prominent OTA sites, and increase brand exposure on popular meta search sites.



For Groupon, Jumpshot found that 41% of its online purchasers visited a social media site prior to purchasing, 19% visited YouTube, and 13% visited a coupon site. Groupon can now dive deeper into the upstream data to find out which social media sites and personalities have the greatest potential to influence its online purchasers. And, it can establish bigger ties with influential coupon sites.



Other online actions to consider investigating include cancellations, category visits, brand searches on Google, and on-site searches. You can look at these events for your own company or you can gain competitive intelligence by looking at these for any competitor.

Cancellations, for example, are interesting because the upstream behavior from a cancellation may indicate why people cancelled. It may also provide a clue into how to win them back (if these were your cancellations) or how to win them over (if these were your competitors’ cancellations).

What audience segments should be studied?

The audience segments will depend heavily upon the action that you’re looking at. For example, there may be a demographic audience that is rising in importance for your business, such as Millennials. Jumpshot can show you all the upstream URLs visited by Millennials that converted. You could then compare the upstream activities of Millennials who convert on your site to the upstream activities of Millennials who convert on a competitor’s site in order to discover new opportunities or partnerships where your competition is having success.

You can also use your CRM data to create your audience segments. For example, if you’re studying category visits on your site, you can compare the upstream activity of customers to the upstream activity of non-customers. Similarly, you could study the upstream differences between customers who checkout with Visa versus customers who checkout with Mastercard. Or, you could study heavy spenders versus light spenders.

How should a massive collection of URLs be analyzed?

You can extract valuable information from Upstream Analytics by looking at the data in the following ways:

  1. Look at the top 10 domains and the URLs within those domains.
    This provides a snapshot of the most popular upstream sites and the internal destinations on those sites that make them popular with your audience.
  2. Look at the top URLs overall.
    This shows you the specific content, regardless of domain, that your audience has most commonly viewed. It could be articles, category landing pages, social media profiles, video URLs, etc. You’ll get a quick sense of what’s popular with your audience.
  3. Filter your view of top URLs overall.
    Some URLs provide better insight than others. If you filter out login pages, home pages, and pages without a descriptive URL, then identifying what’s popular with your audience will be easier.

Things to look for with Upstream Analytics

Using any of the three steps above, look for people, companies, and content that your audience is gravitating towards. Then, think about ways to associate your brand, products, and services with these people, companies, and domains. Try looking for:

  • Brands: Look for familiar and unfamiliar brands in the upstream data. Perhaps there are brands in other industries that have a high correlation with your audience segment that you wouldn’t have thought of previously. If so, you could potentially work with them in a joint partnership. For example, if Virgin America converters show a strong affinity for visiting Dwell Magazine, maybe Virgin America could look to offer discounted or free subscriptions to their Elevate members.
  • Influencers or celebrities: Identify the presence of influencers in the upstream data that you’re not yet collaborating with already. Or even better, identify those that you are working with and look to expand that relationship.
  • Publishers: Recognize publisher sites that you may be able to add to your programmatic campaigns or develop more unique sponsorships if they over-index in popularity with your segments of choice. Beyond advertising, you may be able to identify domains for your PR team to reach out to as well.
  • Affiliates: Identify affiliates in your competitor’s upstream data that are not in your program. Reach out to them and add them to your program to increase your reach.

As you can imagine, Upstream Analytics can inspire a number of business development and marketing ideas, which align directly with the interests of your audience.

Bottom line: When setting the direction for things like co-marketing, influencer outreach, media planning, and affiliate relationships, get the data to make the right decisions. Jumpshot’s Upstream Analytics can provide data to help you formulate a strategy to expand your reach and grow your business. You choose the action that you want to look at, you select the audience that you want to understand better, and then we give you an enormous quantity of data to study. You’re just one request away from getting the data you need.

Recommended Posts