Written by:
Max Fellinger
During my time as a category analyst at a major retailer I learned a lot about what drives ecommerce performance and I’m going to share that with you here. Part of my job during that time was to make changes to the site that I thought would drive year over year revenue increases for the categories I was covering and I got really good at identifying problems quickly. Whether you’re comparing year over year or week over week I found that underperformance can be attributed to one or more of the following metrics; visits, conversion rate, & average order value. Everything else was supplemental to me during that time. Let’s dive in and I’ll discuss how each of the three metrics lead me to multiple million dollar optimizations.
Visits are the lifeblood of any website let alone ecommerce sites, driving traffic is what keeps the lights on so knowing how your customers are getting to your site is important. When comparing traffic to last week or last year I like to first look at the marketing channel the visits came in from. Almost always there are insights to be made by looking at the way customers enter your site compared to a previous time period.
For example, I was working on a category that was in revenue decline year over year and noticed that organic traffic had dropped off significantly. I looked at the page names with the most entries from organic traffic* last week and compared it to the pages with the most entries last year and found that our collection page had dropped off hard. What I discovered after some research was that our collection page had lost considerable ground in SEO rank to our competitors for key terms related to the category.
SEO rank can be a tough thing to get back, but a few of the tactics we used when we noticed a drop was:
- Make sure the url of the collection page aligned to the most popular keywords.
- Refresh the SEO block at the bottom of the collection page to include the most popular keywords.
- Add popular category links as buttons to the top of the collection page, if you have a rocking chair collection these might be ‘wooden rocking chairs’, ect. based on more detailed keywords.
- Making sure filter selections update the url with descriptive keywords.
Any dimension can be used for visits but the following I find to be the most useful:
- Marketing Source – As discussed previously, knowing the UTM Source is a good starting point.
- Device – Are your customers visiting from a mobile device or a desktop? The design of content needs to reflect how your customers are shopping.
- State – If you sell seasonal products, state might be good to look at to see where in the country visits are coming from.
*This is the beauty of BrightWorks, other apps might give you the traffic by marketing source but what about layering in entry pages or device? That can only be done with BrightWorks.
Conversion rate probably sounds like an obvious one, and that’s because it is. You’ve analyzed your visitor traffic, everything looks good, traffic is growing but the revenue is still down. It’s time to give conversion rate a look to determine where customers are falling out of the purchasing process. Conversion rate can be influenced by any number of factors and many of the dimensions mentioned above are a good place to start.
A lot of the not so secret sauce of conversion rate optimization is getting a customer to view the right product as quickly as possible. The more clicks a customer has to make to get to a product detail page the more opportunities you’re giving them to exit the site.
One of the first things that I’m going to compare is the conversion rate differences by device type. Customers shop very differently on a phone vs a laptop or desktop, depending on a shops sku count, mobile customers are much more likely to utilize the search bar while desktop shoppers are more likely to navigate to product. Knowing the nuances can help you optimize the path that is struggling to convert.
In a real world example, we were seeing a drop in conversion rate for mobile customers in the christmas tree category for the first week in November. Not a good time of year to see a drop in this category as you could imagine… I started to look into how customers shopped the category this year versus last year and noticed the conversion rate for customers who had used our internal search bar and searched ‘artificial christmas trees’ had really deteriorated despite the same amount of searches occuring. I then tried for myself and noticed that the search results were only returning about 10% of our assortment, leaving customers with far fewer options than our competitors.
There’s a million ways to affect conversion rate, but following the data is always the right choice. Here are a few more of my favorite dimensions to use with conversion rate:
- Device – Customer journeys look very different on different devices, they’re going to convert differently as well. Use device as a starting point to find outliers.
- Marketing Source – The traffic mix of your visitors will influence the conversion rate. Typically affiliate traffic is going to convert higher than normal while google shopping ads will usually be lower.
- Search Term – Monitoring search conversion rates is important, especially for growing assortments.
Average order value is the last of the big 3 and often times the least likely to be the reason for a decline or spike in revenue. If you’ve made it past visits and conversion rate and still don’t have a reason for a decline in revenue, AOV must have the answer (seriously, it’s the math). It’s mathematically impossible to have visits, conversion rate, and average order value all go up and still have a decline in revenue.
Shopify merchants are experts in increasing a visitors AOV with attachment of supplemental items, I see it all the time when shopping. There are a ton of apps that help merchants build bundles and special offers for bulk discounts which can drive higher AOV. The caution with only looking at AOV is that it can be easy to increase the AOV at the expense of conversion rate, use caution and always balance the two.
One of the ways to improve AOV is to monitor the attachment rates for like items. Lackluster attach rates for a category might require a refresh of the ‘better together’ recommendations or it might be time to expand your assortment with compliment items. Average order value may also be improved by implementing the good, better, best assortment model. Offering products in three pricing tiers can drive customers up the quality ladder in search for better products. It might sound counterintuitive, but adding a lesser item to the assortment can drive AOV and conversion rate.
Average order value can best be utilized with the following dimensions:
- Device – Customers on desktop devices are more likely to have a higher AOV and feel more comfortable making big ticket purchases.
- Vendor – You might have different brands or suppliers that provide products at different price points. Knowing which vendors convert and drive AOV would be worth featuring on the homepage.
Summary
We’ve covered what I like to call the BIG 3 in ecommerce analytics, your starting point for optimizing your shopify site for world dominance! If you’re feeling energized and ready to take the next step we’d love if you gave our shopify app a shot, found HERE. I took all the best features that I used in Adobe Analytics for at my enterprise ecommerce job and brought them to Shopify. Our app offers advanced web analytics that you can’t get from any other app, these are the same analytics the big guys use to drive higher online sales.
Still have questions and want to get started? Contact [email protected] and we’ll be happy to work with you to get started.