Search abandonment: the impact of Shopify and Google Cloud AI partnership
TACKLING $2 TRILLION ECOMMERCE PROBLEM: Shopify’s partnership with Google Cloud AI to provide enterprise class search and discovery tooling to enterprise brands will have game changing impact on the entire Shopify ecosystem.
Shopify’s commitment to AI and the best of breed enterprise technology stack
Shopify’s recent partnership announcement with Google Cloud AI has created quite a bit of buzz within the ecommerce industry.
This collaboration aims to address the issue of search abandonment and poor product discovery experience on online shopping platforms. By leveraging Google’s advanced machine learning algorithms and AI capabilities, Shopify hopes to create a more personalized and efficient shopping experience for its customers. This move is seen as a significant step forward in the race to improve ecommerce user experience, and many experts predict that it will lead to increased customer engagement and higher conversion rates for Shopify’s merchants.
This blog post aims to further reinforce the need for this type of solution and the need for more of these types of partnerships while adding some color to the current state of the initiative and its long term impact.
How big is the search abandonment problem
Pretty big.
According to Google Cloud-commissioned Harris Poll survey search abandonment—when a consumer looks for a product on an online store, doesn’t find what they want, and leaves—costs retailers around the world more than $2 trillion annually.
Another way to sanity check the size of the search abandonment problem is to evaluate the average Shopify add-to-cart rate, a proxy metric for consumers discovering products of interest. According to a Littledata survey of 2,029 stores in September 2022, the average add-to cart cate was 5.1%. The distribution of the add-to-cart rate metric reveals the depth of this problem:
- Top 10% of all Shopify merchants have add-to-cart rate >10.5%
- Top 20% have add-to-cart rate >8.1%
Translated, even if your brand is in the top 10% of all brands using Shopify you’ll still have 90% of your web visitors leaving your Shopify store without discovering a product of interest. Sales opportunity lost, often forever.
How it works
Based on the announcement, enterprise brands on Shopify can today access Google Cloud’s Discovery Al solutions directly through Commerce Components, Shopify’s modern, composable stack for enterprise retail.
This integration, which can now be used by Shopify merchants globally and is available in most languages, enables access to Google’s advanced search and browsing technologies so that retailers can create more fluid and fruitful shopping experiences for their customers.
Having done this type of integration between Shopify and Amazon AWS AI (Amazon Personalize) to build our Obviyo Recommend app, we can attest that this is non-trivial work that requires automated data collection, data shaping, algorithm training, data synchronization, and other tasks that require specialized Machine Learning skills.
Now that Shopify has done the heavy lifting to integrate Google Cloud AI with Shopify’s Commerce Components, store developers can easily add AI-driven features without advanced data science and machine learning expertise.
What it will do?
Shopify’s integration with Google Cloud AI creates open ended possibilities to use artificial intelligence (AI)-driven product discovery capabilities that address real-world business challenges, including:
- Search: use of advanced queries, including non-product and semantic searches, to effectively match product attributes with ecommerce content for fast, relevant product discovery.
- Preferences: the optimal ordering of products on a retailer’s ecommerce site once shoppers choose a category, like “women’s jackets” or “kitchenware.” Over time, the AI learns the preferred product ordering for each page on an ecommerce site using historical data, optimizing how and what products are shown for accuracy, relevance, and likelihood of making a sale.
- Personalization: customization of the results customers get when they search and browse retailers’ websites. The AI underpinning the personalization capability uses a customer’s behavior on an ecommerce site, such as clicks, cart, purchases, and other information, to determine shopper taste and preferences.
- Recommendations: delivering personalized recommendations at scale, making them dynamic and helpful to individual customers.
- Security and privacy: ensuring retailer data is isolated with strong access controls and is only used to deliver relevant search results on their own properties.
This initiative is at its early stages and the initial feature set is focused on search and browsing aspects of user shopping experience.
Who is this for?
Initially, Shopify Ecommerce Components and the integration with Google Cloud AI will appeal to large online brands.
These brands stand to gain significant new revenue from their existing web traffic and a high ROI on investment with this type of solution. A direct integration with Google Cloud AI will further reduce investment risk while shortening time to implementation.
Enterprise brands are now given the opportunity to develop their own custom solution and to have full control of their own data and insights, all without significant additional cost.
On the downside, these brands must account for the ongoing overhead cost of people resources needed to maintain and fully utilize this high value solution.
In the long run, we expect that Shopify will further embed Google Cloud AI capabilities into its platform making it possible for masses of small to medium-sized brands to benefit from its capabilities.
How much it will cost?
When it comes to Google Cloud retail pricing ecommerce teams may have some difficulty determining the exact cost of the service for their brand.
Google Cloud’s pricing is usage based, with each service component priced separately.
For example, search and browse queries in Retail Search are charged at $2.50 per 1000 requests. There’s no charge for importing or managing user events or catalog information.
When it comes to Recommendation AI charges the math gets more complicated.
Here is a copy of Google’s own example of how charges will be calculated:
Example B
This example illustrates a lower volume use case.
In this example a retailer makes 10,000,000 prediction requests a month and trains a single model per day, which, by default, automatically retrains once per day. This amounts to about 150 node hours of model training per month. The model’s quarterly tuning accrued about 90 node hours per tuning; to find the cost per month, we’ll use the monthly average, 30 node hours.
Let’s calculate the price for one month of usage. Because this retailer’s number of prediction requests this month doesn’t exceed 20,000,000, requests will all be charged at the first tier of pricing, $0.27 per 1000 requests.
-
- 10,000,000 predictions = 10,000,000 predictions / 1000 * $0.27 = $2,700
To calculate the cost of training and tuning:
-
- Training charge = 150 node hours * $2.50 = $375
- Tuning charge = 30 node hours * $2.50 = $75
The total cost of predictions, training, and tuning in the given month is $3,150.
It would be fantastic if Shopify could negotiate a master license agreement for Google Cloud AI to simplify pricing and potentially enable pass-through billing.
The impact on Shopify community
Shopify’s partnership with Google Cloud AI will have a long term impact on the Shopify ecosystem:
Merchants: the integration with Google Cloud AI creates a widespread awareness about search abandonment and poor product discovery experience and the value of AI-driven solutions. It will ultimately increase demand for AI-powered solutions, which will benefit merchants and the rest of the Shopify universe.
System Integrators: AI-powered components represent both the challenge and opportunity for system integrators. Challenge, because they will need to acquire additional skills and resources needed for these types of projects. We expect that most of them will take a ‘wait-and-see’ attitude until they detect more significant market demand for such projects.
App Developers: the existing search and personalization software developers will not be initially impacted by this new development since their applications are mostly used by mid-to-small organizations. However, in the long run they will need to redefine their place in this emerging ‘best of breed technology’ architecture.
Why we’ve written this post?
Why is Obviyo, a software company that performed a similar integration between Shopify and Amazon AWS AI to develop its Obviyo Recommend app, praising what Shopify has done with Google Cloud AI, even though Shopify’s additional offering is potentially competing with its own solution?
It’s because what Shopify has done is an example of creating a ‘bigger pie’ for everybody in the Shopify ecosystem.
Our common problem is not too much competition, but rather a lack of demand for these advanced technology solutions.
Shopify merchants are not, as a rule, technology experts. They’re not waking up in the morning thinking about technology strategy, but rather are spending every waking moment thinking about their products and how to retail them.
By partnering with Google Cloud AI Shopify is actually driving a demand for the use of AI-driven solutions. Since no single solution can satisfy all market needs, they are indirectly creating market opportunities for companies like ours.
Obviyo’s strategy?
When two years ago we decided to partner with Amazon’s AWS AI team we did so driven by a desire to provide a best of breed personalization solution to Shopify merchants. Notably, this is the same AI that powers Amazon’s market place.
We concluded that big players, like Google or Amazon, would dominate the AI arena and inevitably commoditize this technology stack.
That’s why we decided to focus our energy on the data side and ease of use.
Obviyo Recommend is a comprehensive ecommerce product recommendation and personalization solution provided as a collection of independent micro-apps called Growth Bots that are super easy to understand and activate on any Shopify store.
Here is an example of one such Growth Bot (mouse over to learn more or view demo):
Product Add-On
PLAY: Reduce number of steps in the buying process by offering additional products that can be added to online shopping cart when clicking on ‘add-to-cart’ button.
WHY IT WORKS: Buying products individually involves more decision making, and more steps. Whereas, the product add-on growth bot allows a customer to buy multiple products together.
PLACEMENT: Product page
Conclusion
Shopify’s integration with Google Cloud AI to help merchants address search abandonment and poor product discovery experiences has the potential of propelling the adoption of AI technology in the Shopify world. It creates awareness about new revenue growth opportunities which will drive more demand for AI-driven applications and agency integration services.