Which Search Engine’s Organic Traffic Was More Lucrative for a Shopify Site? – A Mini SEO Case Study Comparing Sales Per Session from Google, Bing, DuckDuckGo, Ecosia, Brave, and Yahoo!

By Ethan Lazuk

This mini case study shows how organic traffic from search engines like Bing and Brave were more lucrative than other search engines on a sales per session basis for an ecommerce website in Shopify, according to data derived from Shopify Marketing Reports.

It’s a small sample size, and I’m only raising a question here, not providing an answer.

That said, Fabrice Canel of Bing recently shared some data that I found interesting, particularly the part cut off from the tweet excerpt below:

“Additionally, it’s important to remember that Bing powers several other search engines, including DuckDuckGo, Yahoo, Ask, Ecosia, Swisscows and more surfaces as Windows, Copilot and more. If you sum up all, the impact is even more significant. https://comscore.com/Insights/Blog/Generative-AI-Search

This isn’t just a matter of click volume; it’s about the business value derived from those clicks, which correlates well with searchers satisfaction. SEOs who monitor business conversion rates per 1000 users find Bing’s traffic to be particularly valuable. Happy SEOs tracking business conversion = happy “Bing” searcher = happy Fabrice.”

Indeed @lilyraynyc, Bing usage extends beyond what many search engine optimizers (SEOs) may realize. If you zoom in, there have been notable trends to consider as the market share of desktop search engines in the United States since 2019 https://t.co/kcJ0sXX572

Additionally,… https://t.co/MMTrN4d6tO

— Fabrice Canel (@facan) April 2, 2024

The point about “business conversions” got my attention. So I decided to revisit this case study and give it a quick refresh.

The data is from August of 2023 (before generative AI really took off in Search), but I think it gets at an interesting topic as we enter more of the AI era:

How should we rethink and track the value of organic search visits and SEO’s contribution to overall marketing and business goals?

Back story on this mini case study:

Back in August of 2023, I was working with several ecommerce clients who’s sites were built with Shopify, so I took more interest in organic search revenue reported in Shopify, per its marketing reports.

One thing I noticed is that some websites I was working with received notable levels of sales attributed to organic sessions from customers who used search engines other than Google.

These included Bing, Brave, DuckDuckGo, Yahoo!, and Ecosia.

I decided to dig in and explore the user journeys on those search engines and its impact on the ecommerce data I was seeing.

Specifically, I was curious how the level of sales a website received from a particular search engine’s users compared to its level of organic traffic from that search engine.

In other words, which search engine’s had the more lucrative traffic, based on revenue attributed to organic search in Shopify?

As it turned it, there was a noticeable difference, particularly separating Google and Ecosia from the rest of the bunch in terms of having lower sales per session, while Bing and Brave stood out for having higher sales per session.

Mini case study

Shopify’s Marketing reports are a great compliment to GA4 ecommerce reporting. They provide information about conversions related to different marketing channels.

Shopify defines conversions in the report as “when an online store visitor becomes a paying customer.”

In other words, unlike GA4 events, Shopify conversions are related to revenue:

“The reports can outline which of your marketing channels are resulting in the highest number of customers and the customers of the highest value, based on their interactions with your online store.”

– Shopify Help Center, Marketing reports

Here are some other key vocabulary to know:

• Referring Channel: Touchpoint along the path to completing a marketing goal, like a sale; in this case a search engine
• Organic: A type of online surface that a customer was on before landing on the store; others being Paid or other
• Sales: Total sales from all channels after discounts, plus returns (not including tax or shipping)
• Sessions: A continuous set of online store activity where each visit is less than 30 minutes apart; unlike sales, sessions aren’t affected by the attribution model
• Last non-direct click attribution model: Direct traffic is ignored and 100% of credit goes to the last channel clicked
• Any click attribution model: 100% of the credit goes to each channel clicked

The main Shopify report I used for this case study is the sales attributed to marketing report, which I filtered for organic search.

Here are the comparisons of data using different attribution models.

Last non-direct click

This screenshot below shows 30 days of channel performance data for a Shopify store using the last non-direct click attribution model with filters setting the referring channel as a search engine and the traffic source as organic.

Total sales for all 6,064 organic search sessions was $15,247.23.

That’s an average of $2.52 in sales for every organic search session.

Compare that to each search engine individually:

• Google: $2.15 per session
• Bing: $8.07 per session
• DuckDuckGo: $4.01 per session
• Brave: $4.40 per session
• Yahoo!: $5.76 per session
• Ecosia: $2.06 per session

Bing organic traffic was the most lucrative for this website at $8.07 in sales for every session from Bing search, accounting for 3.2% of total organic sessions.

Yahoo!, Brave, and DuckDuckGo contributed an average of $4.73 per session, accounting for 8.7% of total organic sessions.

Ecosia was the least lucrative at $2.06 per session, but also the least used with just 0.3% of total organic sessions.

Google was the second least lucrative at $2.15 per session.

Any click

Any click tends to attribute the largest amount of sales to organic search of any attribution model in Shopify because it equally credits all touchpoints along the messy ecommerce user journey.

Below is a screenshot of the same site using the same 30-day timeframe:

Total sales for all organic search using the any-click attribution model is $22,175.41.

That’s an average of $3.66 in sales for each organic search session.

Here are averages at an individual search engine level:

• Google: $3.30 per session
• Bing: $9.12 per session
• DuckDuckGo: $4.83 per session
• Brave: $7.10 per session
• Yahoo!: $7.91 per session
• Ecosia: $2.06 per session

Bing’s organic search traffic was again the most lucrative at $9.12 per session.

Yahoo! stayed the second most lucrative at $7.91 per session, while Brave was still third at $7.10 per session.

Next to Ecosia (which stayed the same), Google was again the second least lucrative at $3.30 per session.

Comparing last non-direct click and any click

Since we used two attribution models, let’s compare how much using the any click model impacted the reported sales per session compared to the last non-direct click:

• Google: $2.15 vs. $3.30 = +$1.15 per session
• Bing: $8.07 vs. $9.12 = +$1.05 per session
• DuckDuckGo: $4.01 vs. $4.83 = +$0.82 per session
• Brave: $4.40 vs. $7.10 = +$2.70 per session
• Yahoo!: $5.76 vs. $7.91 = +$2.15 per session
• Ecosia: $2.06 (no change)

Here we see that the last non-direct click and any click attribution models were closest for DuckDuckGo, Google, and Bing, with a variance of of $0.82-$1.15.

However, for Brave and Yahoo!, the variance was $2.15-$2.70.

This suggests that the nature of user journeys incorporating the different search engines may have nuances to investigate further.

Why would sales per session vary across search engines?

One possibility why search engines had different sales per session numbers could be the percentage of ads on them.

I showed in a traffic fluctuations case study how the presence of ads for branded product terms, for example, can suppress organic traffic levels.

Another possibility could be the characteristics of the search engines’ audiences.

I gave ChatGPT with Link Reader the following prompt:

“How might you explain the different types of audiences using these search engines for organic search: Google, Bing, DuckDuckGo, Brave, Yahoo!, and Ecosia, based on their respective about pages: https://about.google/, https://learn.microsoft.com/en-us/microsoftsearch/overview-microsoft-search-bing, https://duckduckgo.com/about, https://brave.com/learn/, https://www.yahooinc.com/about, https://ecosia.helpscoutdocs.com/article/406-about-ecosia.”

Caveat: Keep in mind that LLMs can hallucinate and have other limitations, as shown in this article about ChatGPT SEO prompts by Shawn Anderson.

Here was ChatGPT’s response:

“Summary of Different Search Engine Audiences Based on Their About Pages

Google

• Audience: General public, businesses, and developers.
• Focus: Providing useful and relevant information through search, as well as various other services like Maps, YouTube, and Google Drive.
• Unique Selling Points: Speed, relevance, and a broad ecosystem of services.

Bing (Microsoft Search)

• Audience: Organizations, work or school users, and the general public.
• Focus: Combines capabilities of Microsoft Search and Bing to provide relevant results from an organization and the web.
• Unique Selling Points: Enhanced privacy and security measures, multiple entry points for search, and tailored experiences for work and school users.

DuckDuckGo

• Audience: Privacy-conscious individuals.
• Focus: Online privacy and data protection.
• Unique Selling Points: Does not track users, offers a suite of privacy tools like Web Tracking Protection and Email Protection.

Brave

• Audience: Users seeking more security and faster web experiences.
• Focus: Online security, ad-blocking, and faster web browsing.
• Unique Selling Points: Blocks ads by default, focuses on privacy and performance.

Yahoo!

• Audience: General public with a focus on sports, finance, mail, and more.
• Focus: Media, technology, and business platforms.
• Unique Selling Points: A diverse range of services and a focus on mental health awareness and support.

Ecosia

• Audience: Environmentally-conscious individuals.
• Focus: Planting trees and protecting the environment.
• Unique Selling Points: Dedicates 100% of profits to planting trees, operates its own solar plants, and is privacy-friendly.”

– ChatGPT response to prompt shown above, pulled on 9/23/23.

Aside from ads or audience differences, there could be differences in the search results themselves.

Perhaps different audiences were shown different product display pages or even page types (like category pages) across similar queries.

Another thought is maybe Google’s Surfaces Across Google (SAG) and non-SAG organic results contributed to variance in the data.

For what it’s worth, I searched the name of the brand as a navigational query on desktop for all of the different search engines, including Google (traditional and SGE), Bing (traditional and Copilot), Brave, Ecosia, Yahoo!, and DuckDuckGo.

Google and Ecosia were the only SERPs that pushed the first organic result for the website below the fold with ads (i.e., the user had to scroll to see an organic result).

These search engines were also the least lucrative for organic traffic on a sales-per-session basis in this case study.

Again, this only involved one Shopify store, so it’s a tiny sample size.

This is just food for thought.

“There There”

This case study showed that for an individual Shopify store, Bing and Brave were the most lucrative on a sales per session basis over a 30 day period.

This might be because their SERPs showed fewer ads compared to Google and Ecosia, the least lucrative and most ad heavy.

Then again, I don’t think we can draw too large of conclusions from this. It’s more so meant to raise the suggestion that how we think about the value of traffic should evolve.

Especially nowadays in 2024 with more generative AI features in Search (and tangential too it) than we had in August of 2023 when this data was gathered.

While I’m not able to continue this particular case study, I do plan to explore this topic in greater depth in the future.

Till then, enjoy the vibes:

Thanks for reading. Happy optimizing! 🙂