Ecommerce companies have many options at their disposal to understand who’s visiting their site, how marketing is performing, and how to recalibrate it to drive traffic and conversions. Many brands will start by looking at Google Analytics. Others may use data from ad platforms such as Google, Facebook, and TikTok. Still others will look into independent attribution platforms such as AdAmplify’s Dimensions platform.
What are the pros and cons of each of these approaches? Let’s break them down to help ecommerce brands understand which resources they should use to analyze their audience and convert attention into customer dollars.
Google Analytics
When you’re trying to understand who’s visiting your site, how to find more visitors, and how to convert them into customers, Google Analytics is a natural first stop. It’s free and easily accessible. The way it works is that you give Google the ability to track user interactions on your site by adding a snippet of Google code to it. Google then provides reports based on an analysis of visitors so that you can understand traffic patterns.
But as with most free tools, not to mention tools provided by a search advertising behemoth with incentives to overweight its own role in customer journeys, there are downsides to relying on Google Analytics. Here are some cons:
- Google provides reports based on an analysis of visitors, not customers, so its data does not help you understand the behavior of specific customers. It’s hard to distinguish between visitors and customers, and even if you can spot customers, you don’t have rich data on each of them
- One size fits all. Google determines what it will report on
- Google is not always accurate
- Google Analytics decides on the attribution models it uses. After GA’s deprecation of both Linear and Position-Based multi-touch attribution models, and First-Click, Analytics now supports only Last Non-Direct Click and their data-driven model.
- Google defaults to a 30-day conversion lookback window, but the site admin may select 60 or 90 days. This restriction (also seen in various ad platforms) limits and will distort attribution modeling for longer-term sales cycles or for breaking out initial and subsequent purchases
In short, Google Analytics is a good starting point for brands wading into their visitor data. It’s free and accessible. But it doesn’t provide much control or granularity. Its lookback window is limited, especially for brands with longer sales cycles. And it doesn’t provide recommendations to guide future marketing efforts.
Ad Platforms: Google, Facebook, and TikTok
The most popular ad platforms also offer their own attribution data so that brands can understand how effective their marketing investments are and calibrate spend on an ongoing basis.
This data can also be tempting. If you’re already using these platforms to advertise, their performance data is just a couple of clicks away. But there are downsides to letting a company that wants you to spend more ad dollars call its own balls and strikes.
Here are some of the risks to know when using ad platforms’ attribution data:
- The more ad platforms that are used, the more time is required to analyze and compile their data into comparative reporting. With each platform’s setup criteria (e.g, lookback windows) and reporting being unique, and with varying policies in place from platform to platform on taking credit for a purchase, it is hard for marketers to create a holistic, accurate picture of marketing performance based on each ad platform’s data
- Look-back windows that define the timeframe of touchpoints taken into account in a user’s journey to conversion are typically restricted (e.g., for a maximum of 90 days), with defaults set to 30 days for Google Ads and a maximum of 7 days for Facebook. These incompatible look-back windows make it impossible to compare Facebook conversion reporting to Google’s conversion reporting. For companies with longer user journeys than 30 to 90 days, this means interactions prior to the lookback window are no longer taken into account
- The platform has a vested interest in showing its performance in the best possible light (for example, Google Ads will take full credit within the defined lookback window for any conversion that a Google ad was part of)
- Google Ads is moving increasingly towards restricting users’ ability to configure campaigns. Google’s “Performance Max” campaigns are now the default, and they provide users hardly any control beyond setting a budget
While it’s fair to use Google to analyze Google ad performance or Facebook to do the same on its platform, you have to question the neutrality of an attribution product that has a vested interest in supporting its own performance. Also, it’s challenging for a brand advertising on multiple platforms to piece together the data. You may want an attribution platform that pieces together the data for you and gives you more flexibility for configuring features like a global lookback window.
Third-Party Attribution Platform: AdAmplify’s Dimensions
An independent attribution platform like Dimensions has three key strengths: objectivity, optionality, and recommendations.
Dimensions only has a vested interest in helping you get the best possible marketing results; it isn’t incentivized to overstate the performance of any given channel. It comes with far more granularity and optionality than Google Analytics and ad platforms, enabling you to view ad performance via multiple attribution models and by focusing on different metrics.
Finally, to simplify all that data, Dimensions uses machine learning to forecast future revenue by channel and to surface key recommendations for future marketing campaigns. This means you get not just a holistic view of marketing performance but also holistic guidance on how to recalibrate marketing to improve performance.
Here’s a more detailed breakdown of Dimensions’ capabilities:
- Users are tracked at a named level in the Dimensions database, so retargeting each purchaser or sorting purchasers into user-defined cohorts is a standard feature
- The look-back window is client-configurable back to 600 days, which is advantageous for sites with longer sales cycles or seasonal cycles and for tying subsequent purchases back to the initial channel or campaign
- Attribution data is maintained for 2 years, with an option to extend it for additional years
- Dimensions supports multiple standard attribution models (first-click, last-click, last non-direct click, linear, non-direct linear, and its own Machine Learning model), with a reporting option to compare any two models. Using a number of models can be important in getting different views of channel and campaign performance (for example, Dimensions’ first click model breaks out total purchases by initial purchases and subsequent purchases so you get a fuller picture of the long term value of bringing new purchasers through the various channels and campaigns you’re using).
- User purchases are broken down in each attribution model by initial purchase and subsequent purchase (revenue, number of customers, and number of orders) for channels, campaigns, and ad creative. Average order value and gross margin at every level helps marketers readily understand each channel and campaign’s ability to attract customers who repurchase and the profitability of each channel and campaign
- User journeys that show each named user, their last and cumulative purchases, number of days between orders, and graphically renders each user journey with all touchpoints with their applicable channel and campaign
- Summary reporting on the number of customers who have purchased once, twice, etc. plus the mean number of days between orders for a site. This allows the site to develop a unique cadence for campaigns retargeting customers
- Trends and Key Metrics Dashboard highlights overall performance for marketers and executives
- Saves time for marketers/business owners whose sites are using multiple marketing channels by collecting and analyzing the essential data in a single platform
Machine Learning Powers the Latest Version of Dimensions
Dimensions’ proprietary Machine Learning engine provides marketers with a unique set of features that highlights each channel’s ability to generate future initial and subsequent purchases. This ability to forecast lift is unique to Dimensions. Here’s a brief summary these capabilities:
- Custom Graphical Linear Model that shows the current weight (contribution to revenue) for each channel being used by the site
- Based on an analysis of the user behavior of visitors coming from each channel, Dimensions Attribution calculates the Probability of Purchase for each Channel and compares it to the Site’s overall probability of purchase. This tool provides a graphical analysis of channel performance and, by overlaying each channel’s weight, which are performing well, which are underperforming, and which offer opportunities to explore underutilized channels that have a high probability of purchase
- Predictive forecasting of revenue from initial and subsequent purchases that could be generated over time from driving additional traffic to the site overall, or from each Channel
- Recommendations of where there’s opportunity to scale conversions and revenue
Key Points When Choosing an Attribution Solution
Google Analytics has its place in providing visitors insights for marketers, such as demographics, devices, and operating systems, that are typically not included in an attribution platform. And using ad platforms’ own measurement data is convenient. But these options provide you little control, they’re hard to piece together to create a holistic picture, and they have a vested interest in overstating their own performance.
Dimensions provides another layer of actionable data, with a focus on doing the heavy lifting in attribution — the analysis of marketing performance. Using machine learning, Dimensions provides deeper insight into your marketing channels and campaigns by analyzing and reporting on what is working and what’s not. Most importantly, it highlights where there are opportunities to more reliably scale your business.
If you want to talk about using Dimensions to understand your audience and how to transform more of them into repeat customers, book a demo today.
