How to evaluate a transactional site

When evaluating an ecommerce site, it is important to utilize the perspective of the purchaser. The core aspects of Customer Journey success are: Reach, efficiency, and conversions. This analysis begins by looking at a site’s YoY traffic volume change, and then benchmarking the previously mentioned KPIs against similar sites.

Here are the steps we recommended taking in order to evaluate a transactional site:

  • How to measure reach
  • How to measure the site’s effectiveness or ability to convert
  • How to measure a proxy for conversions

Step 1 - Measure reach

First, enter a website under Website Analysis to analyze.

1. Navigate to Traffic and Engagement and select Monthly Visits. The monthly visits help us understand total reach of a site. This can be trended out YoY to reveal true growth or decline. Unique visitors can be used to indicate the actual reach of a website.


2. Next, go to Marketing Channels. This view reveals the paths through which people arrive at the site. Direct traffic indicates brand awareness and recognition. Organic search is traffic coming from search engines for branded or non-branded search terms. Paid search volume shows search terms that they are paying for via search engines.

3. Go to App Analysis to understand the website’s app performance. App downloads represent the number of devices the app has been downloaded on, over a specified time period. However, some people will download without actually using the app, so further investigation is required. Use this metric to evaluate the growth rate of an App. Correlate this data with events to see how it influenced app acquisition (i.e. app updates, reviews, promotions etc.).

4. By benchmarking one site against the industry or their competitors, we are able to get insights around how the site is performing relative to the market. Under Website Analysis, select a site and Compare to a competitor, then navigate to Traffic and Engagement. Evaluate and compare the quality of each site’s traffic/engagement with the site. The data can be trended over time to look for any seasonality or anomalies.

5. Competitor’s can be benchmarked under App Analysis and trended over time to see the install penetration rate between the two. Install Penetration represents the percentage of Android devices that have the application installed, out of the total Android devices in the selected market.

6. Benchmark the two sites by looking at Marketing Channels under Website Analysis. By comparing the differences in the share of Marketing Channels, we can see the differences in user acquisition strategy and brand awareness.

Step 2: Measure the site’s effectiveness

1. When analyzing a site and looking at key engagement metrics, we are able to see the effectiveness of the site and the audience’s engagement.

The above chart is an example of the calculations listed below. This was pulled for looking at the past 24 months. The metrics that were pulled included metrics such as visits and bounce rate. What this shows correlation between Visits, Non- Bounce (Unbounce) Visits, and Total Pages Visited. The Y-Axis is the percent change MoM and the X-Axis is the time frame. Notice that the total pages visited tend to diverge over certain periods. When you export to excel, you are able to run the following calculations:

  • Total pages visited : TPV=visits x average pages per visit
  • Total visit duration: total visits x average visit duration
  • Non-bounce visits: Calculating the non-bounce (non-bounce visits=1-bounce rate*total visits) visits let’s you get a better idea of the engaged audience that gets past one page. Non-bounce visits will show engaged visitors and you can trend this out to calculate total visits over time. Simply export the data and calculate the non-bounce visits.

Effectiveness can be measured by looking at App data as well to give you a holistic view of both website and app metrics.  From the App side, we can calculate the Daily Active Users (DAUs). DAU’s represent the number of unique users  who  use  the  analyzed  app  on  a  daily  basis, in  the analyzed country. Use this  metric to assess an app’s stickiness compared  to other apps.

Look at: trends in bounce rate, pages per visit, and average visit duration, over time... If there happens to be an increase in bounce rate over time, that could be a concern as the audience is less engaged with the site.

2: Benchmark these engagement metrics against competitors to spot any trends within the Traffic and Engagement section.

Look at: trends in bounce rate, pages per visit, and average visit duration, over time.. If there happens to be an increase in bounce rate over time, that could be a concern as the audience is less engaged with the site.

3. When benchmarking apps: Analyze for open rate and duration on app.

Open rate: represents the percentage of the total Android devices who have installed the app, and use the analyzed app on a daily basis, in the specified country and timeframe. Use this metric to assess an app’s stickiness compared to other apps.

Time per user: represents the average daily time users spend on the  app,  in  the  analyzed  country and timeframe. SimilarWeb measures  Usage  Time  for  apps  that  are  being  used  in  the foreground. Use this metric to evaluate advertising opportunities and assess the likelihood of in-app purchase.

Step 3: Measure a proxy for conversions

*SimilarWeb does provide conversion reports with this data, however we can see the proxy using the platform for conversions.

1. To understand volume, go to Website Analysis, Content, and then select Subdomains followed by Pages.

Subdomains: Domains under the single parent site, but start before the domain name (ie Depending on the structure of the site, it is sometimes easy to spot a secure site, which is indicative of someone going through a secure checkout process. It might not be clear for every site, so it helps to look at the Subdomain section to identify this.

Once identified, we can see the secure. domain and the share of traffic that is being sent to this subdomain. You can export this data and estimate the traffic being sent to this subdomain and what the potential conversions are.

Pages: shows what the the popular urls that are capturing traffic share for the site.* Here we can look to identify shopping carts, login pages, and even thank you/confirmation pages. This indicates the shopping cart process and what share of traffic is going to the thank you or order confirmation pages.

* The methodology for pageviews is different than how we calculate web traffic as the pages need to capture a certain threshold of traffic for us to include in this section.

2. The SimilarWeb team has rolled out Funnel Analysis where you can uncover the Conversion and Performance Metrics of key players in top industries within SimilarWeb’s platform. The conversion data helps you understand a site’s conversion efficiency and scale over time. Additionally, you can take a deep dive into a specific company, analyzing the efficiency and ROI of their marketing spend, the rate of change in converted visits, and more in a Year over Year view. Another significant value is that you can benchmark funnel analysis and user stickiness for the leading players in each industry.

For more information, click here.



Was this article helpful?
10 out of 14 found this helpful