Time to go outside again, I guess. I’ll need a sun hat. Sunscreen. Maybe some new sandals? What else?

With the Recommendations AI service, I might be reminded to grab a reusable water bottle and a swimsuit. Or some after-sun aloe lotion. Good thing, cause I’ll need it.

Recommendations AI is a solution that uses machine learning to bring product recommendations to their shoppers across any catalog or client list. This service is part of our full suite of Retail solutions. When you integrate with the Retail API, you get the benefit of Google’s Product Discovery. Integrating once to reap the benefits over and over. Recommendations is the starting point, and you can easily extend into Retail Search and Vision Product Search in the upcoming future. 

The Recommendations solution is fully managed, global-scale and powered by deep learning, so you can focus on a great shopping experience and let someone else worry about the infrastructure.

Compared to baseline recommendation systems used by customers, Recommendations AI showed double digit uplift in conversion and clickthrough rates in A/B experiments controlled by the customers. You can optimize for click-through, conversion or session revenue, and fine tune the models to make sure you omit out-of-stock items or duplicates, for example.

So how does it work, and how do you get started? Read on, and we’ll walk you through the pipeline, starting with the data you already have to placement in your online store. 

Formula: Data -> Model -> Placement

You start with your catalog, the list of all the things (postcards, movies, pie recipes) that you want to show your customers. Then you ingest your PII-redacted user events -this is the historic event data like home page views, add to cart events and more along with real time user events. This user event is joined with the product catalog and items that allows us to construct the sequence of shoppers’ activity, thus being able to predict what the shopper has a high propensity to purchase next. The user events can come from both online activity across devices or offline store purchases

The recommendation model will return a list of products, which are the recommendations. The brains of the operation, if you will. This model is trained using all the data that you ingest, using the latest neural network models and techniques that Google has built expertise over the years in flagship products like Youtube and News, that allows us to uncover shopper intent,  so it can best predict the right recommendations to show to the right people.

Every model outputs a list of product identifiers, but where do they go? They go into placements, the spots, panels, carousels on your customer’s journey interacting with your brand that you’ve set aside to highlight recommendations. A model can send recommendations to one or more placements, but each placement only receives information from one recommendation model. Your pages will then need to render the products with the right images, text or other metadata, using the product ID that is returned by the model.

What do recommendations look like?

Let’s start by browsing our postcard-selling website, where I’ve been buying some vintage California postcards already. The recommendations algorithm has caught on to my interest, showing me other potential cards to purchase based on my history:

Put your data to work

To get started we need to bring your data into the recommendation model, so it can understand your customers, your inventory, and your sales patterns. 

The model takes in the product catalog you use, and metadata about those products to better understand nuances in assortment, pricing and variables like size and style. You might already have this data stored in BigQuery or Merchant Center, and hence we provide easy integrations that you can leverage to get started even faster.

As for the user events, don’t worry if you already have systems in place to capture web and mobile activity. We make it easy to bring in your real time event logs by providing seamless integrations with Google Tag Manager, Javascript pixel, or even historic events from Cloud Storage, BigQuery or using inline API or JSON, so you can immediately train the models on this imported  data. All this allows you to kickstart integrating with Recommendations AI in a matter of days.

The models then construct a sequence of activities that the user went through and joins with the products that the user engaged with. Once your data is ready to go, it takes a few days to train the model. Next onto making the data work for you.

Quickly customize your model

Setting up your own recommendations project in the console gives you the ability to choose what sort of model to train (based on what recommendations you want to generate) and your objective. Are you optimizing for click-through rate–more people click on the recommendation links or products–or for conversion rate–more people choose or buy what was suggested or revenue ?

Different models can be optimized for different optimization goals.; the GCP console explains what each one can do and how you can choose to optimize it.

Let’s unpack some of this terminology real quick.

We’ve got three model types:

• Recommended for you – Means we think these are items you’ll want to buy, based on your history; this is usually used on a home page to showcase items.
• Others you may like – Means if you’re browsing the page of a water bottle, we will recommend  alternative brands of water bottles that you may like as well as a backpack, based on your engagement  history.
• Frequently bought together – Means that when anyone buys sunscreen, we notice that they often also buy aloe lotion, so we will surface those items when someone adds any one of them to their cart.

And then we have three business objectives that the models optimize for:

• Click-through rate – How frequently did somebody click on a recommended item?
• Conversion rate– How frequently did somebody add a recommended item to their cart?
• Revenue per session – How much money did the recommendations generate for you?

Deliver anywhere along the journey

Now that you’re all set up in the Retail AI console, you can test out the recommendations in the console, even before you deploy to production.

You can integrate Recommendations into your frontend by calling the Predict APIt. The placements of recommendations will report data back into the dashboard and you can analyze and measure success for future iterations. 

On top of that you can use the recommendations for other parts of your customer’s journey. Email promotions, storefront kiosks, display ads or follow-up notifications can include recommendations based on past activity and cart contents. The model gives you useful product recommendations for a wide variety of touchpoints and steps in the purchasing process.

More best practices, and guides, are available inside our documentation.

How to get started

Training your own models can be tedious, time-consuming, and expensive. On top of that it requires deeper data science expertise to set up. Let us do it instead!

You can see how IKEA Retail uses Recommendations AI in this recent talk and blog from the Google Cloud Retail Summit..

To get started today you’ll need to make a Cloud project and enable the Retail API, which then allows you to access all the recommendation tools in one menu. Bring in your catalog and purchasing data, define a placement or two, and you can start putting recommendations on your site in a matter of days.

In the last two decades, BHI, once a small construction company in Vernal, Utah, has expanded its operations to dozens of industries in over 25 states. But while the company grew, its technology trailed behind. File sharing and emails both ran off a single server. Editing a document was a grueling undertaking that required an employee to VPN in, patiently download the file, edit it, then upload it again—all from a job site with a limited internet connection. Acknowledging the need for better technology, BHI adopted Google Workspace and AppSheet and transformed its business to be more innovative, productive, and profitable.

Rebuilding collaboration and productivity

BHI first looked to replace its archaic email and file-sharing systems. They found that Google Workspace—as a proven and popular collaboration solution for remote teams and a familiar, preferred platform that most employees use in their personal lives—was the clear choice. BHI migrated its 500 employees to Google Workspace in less than a week. Productivity immediately soared, since employees could collaborate more easily with Google Workspace’s products and access them from any device, from anywhere—even at the construction sites.

For instance, the contract writing process drastically improved with Google Workspace. Version control had always been a struggle—as employees circulated a draft contract from one computer to the next, versions became unruly and randomized, untracked changes popped up, and files went missing. With Google Workspace, the pre-construction, legal, and finance teams can all edit and track changes of a contract from wherever they work. Version control has become a thing of the past, as everyone is always working on the latest version.

By adopting Google Workspace, BHI decentralized its operations and began collaborating in real time across over 150 job sites. Management and IT gained new insights into challenges that they may never have uncovered nor confronted without a cloud-based productivity tool. The IT team also saw an opportunity to overhaul the methods of their deskless workforce, which accounts for 90% of the company. This deskless majority needed to connect directly to the data that influenced, impacted, and comprised their work. They also required more streamlined processes. AppSheet provided the answer to the workforce’s needs, enabling IT to create applications that simplified key tasks so mobile workers could get their jobs done faster and better.

For example, one BHI client required extensive inspections multiple times a day. Before using AppSheet, a BHI inspector would perform the inspection, return to a computer, record the results from paper to a spreadsheet, copy and paste them into a formatted report in a Google Docs, convert the Doc to a PDF, and email the final inspection report to the customer. This manual process took two hours to complete. Now, using AppSheet, the inspector inputs the results on their mobile device while performing the inspection. As soon as the device connects to the internet, it automatically updates the database with the inspector’s inputs, adds the inspection results to a preformatted report built-in Google Docs, and emails the final report as a PDF to the customer. With AppSheet, BHI shortened this inspection process from two hours to six minutes, saving over $50,000 a month in labor. “With Google Workspace and AppSheet, BHI has turned into a digital company with real-time data allowing for real-time decisions, collaboration, and transparency in a decentralized environment,” says Johnny Hacking, Director of IT.

An easier software choice for faster results

BHI also found that building apps with AppSheet is much easier than using other solutions, such as developing with low- or full-code tools or purchasing third-party products. Because AppSheet is a true no-code platform, anyone can learn to build and maintain apps. Since adopting AppSheet, the BHI IT team has become so proficient with the platform that they have eliminated the company’s previous backlog of software needs. Now, when a new site opens up, IT proactively works with the site managers to understand its unique requirements in order to build a custom app. Start to finish, scoping, and building this site-specific app typically takes less than a week—compared to months with traditional development platforms. 

IT team members aren’t the only BHI employees building apps. As Hacking explains, “We have people that aren’t in IT, and when they have an idea we say hey, log into AppSheet, build it out, and then we’ll help out with some final touches on it.” 

Empowering employees outside of IT to build AppSheet apps has benefited teams across the company. For example, one team on a solar farm construction project needed to perform inspections of an array of specialized technical equipment with which IT was not familiar. In this case, the solar team, being the subject matter experts, simply built the app according to their specifications. All IT had to do was jump in at the end to help finalize the app—the final product was completely customized to the solar team’s very specific needs. This highlights the benefits of no-code development: on-the-ground experts build and prototype job-specific apps and IT puts the finishing touches on the functionality, all while keeping guardrails in place.

Using Google Workspace and AppSheet also simplifies BHI’s reliance on third-party software. “Our company divisions are so diverse that they each require different software,” says Hacking. “Plus, if we buy a third-party software, we still have to customize it by getting other software, because they don’t play well together. By replacing this software with AppSheet apps, we save a lot of money. And it’s easy to integrate those apps because on the back end, they’re all connected to Google Sheets.” 

For the few third-party software products that BHI hasn’t been able to replace, they’ve used Apps Scripts to import the software’s data into Google Sheets, so that their data all lives on one platform. Overall, BHI has been able to free up 10% of its total IT spend by replacing third-party software with Google Workspace and AppSheet apps.

Company culture transformed by technology

“Because we’ve used Google Workspace and AppSheet and it’s been so successful, people are very willing to use them,” Hacking says. “And for a construction company to gain so much trust in these platforms in such a short period of time—that’s huge. Now we’re at the point where employees are asking, ‘What else can we do?’”

Adopting Google Workspace and AppSheet has not only made BHI more productive, but also transformed the company culture. Employees feel empowered to be more innovative and are constantly seeking ways to improve. Simplifying employees’ work has led to a direct boost in morale. 

The role of IT has also transformed. As Hacking explains, “In just three years of using these technologies, IT has gone from being support overhead to being proactively brought to the table to take part in the business strategy.” Increased IT visibility and culpability has helped solve business problems and led to increased profitability. 

BHI continues to find new ways to leverage Google Workspace and AppSheet to improve business. In response to COVID, they created a timesheet app that allows employees to clock in without touching a device. Most meetings are now held virtually using Google Meet. Fleet maintenance and repairs are all managed with apps. And, of course, the company has better data and insights than ever. As Hacking puts it, “We’re continuing to stay on the cusp of innovation because of Google Workspace and AppSheet.”

The Google Maps Platform Android SDK supports extensions for reactive programming, which helps you write code to handle asynchronous operations.

Write reactive and responsive mapping applications with Google Maps Platform

In mobile apps, asynchronous events can happen at any point in time: user touch events, waiting for network calls to complete, or receiving push notifications, to name a few. As an app developer, accounting for these events and composing them with other asynchronous events can be challenging. Reactive programming is an alternative to passing callbacks for different events and helps simplify the process of working with asynchronous events. In reactive programming, events are modeled as a stream, emitting items over time.

https://youtube.com/watch?v=1TmJvOZfBVQ%3Fenablejsapi%3D1%26

There are two libraries you can use to write reactive code: Kotlin Flows and RxJava. The next two videos show you how to use each one.

Writing reactive and responsive mapping applications using Kotlin

If you’re a Kotlin developer looking to use Coroutines and Flows, the KTX library allows you to use Kotlin Flows to receive events. The Maps KTX library includes extension functions that return Kotlin Flow objects, so you can listen to events in a reactive manner. Unlike suspending functions, which return a single value, Kotlin Flows can return several values over time. For example, you can use a Flow to receive camera event changes over time. To get started using Kotlin Flows in your app, include the Maps KTX library in the dependencies section in your build.gradle file.

https://youtube.com/watch?v=Cotx1ZmEYg8%3Fenablejsapi%3D1%26

Creating reactive maps on Android with RxJava

Integrate Google Maps Platform SDKs with popular Android library RxJava. RxJava is the Java implementation of Reactive Extensions, which is a library for composing asynchronous and event-based programs using observable sequences. One thing RxJava does is allows you to convert callback-based asynchronous code into a chain of transformations. Learn how this works and the other ways you can use RxJava in the third video in this series.

https://youtube.com/watch?v=AgGE7fAMrdA%3Fenablejsapi%3D1%26

​​We hope you learn more about reactive programming concepts to help you build responsive and reactive mobile apps in this three-part YouTube series. Have ideas for helpful videos you’d like to see on our channel? Leave a comment on any of our videos. And don’t forget to subscribe to our YouTube channel for the latest updates, tutorials, customer stories, and more.

For more information on Google Maps Platform, visit our website.