LR: When it comes to social data, I am curious how sellers are responding to that sort of data when it comes to commerce. And do you feel like additional data from the customer is something that sellers are becoming more excited about?
And one of the ways that we try to connect customers and sellers on the Amazon detail page, is a new feature we introduced earlier this year called Ask, A-S-K, and you could see now, but if you go down some of our detail pages, we do something really interesting which is that we allow customers to ask a question about a product.
And we serve those questions up in an anonymous way to both the sellers of the items and also customers who previously purchased that item. I think that a lot of sellers want, I am imagining, their product to be discoverable.
How do you balance that need for discovery with also having millions of sellers on the platform? PF: Well, one of the things that our sellers love about selling on Amazon is they are really in complete control of their business. So when it comes to these data-driven recommendations we allow them to opt in and opt out. We make over 50 different recommendations today and they can choose which recommendations we service up to them based on the ones that are most important to them or most important for their product.
We challenge ourselves and we measure how useful they are. We know the majority of our sellers actually use our recommendation today, because we measure and track that.
Then we also track how much improvement to their business did they get from using our recommendation and we hold ourselves accountable. LR: What would you say the most popular tool that sellers using when it comes to personalizion?
PF: The No. So that still remains by far the most popular type of recommendation we make today, because you could imagine for sellers who either sell a lot of products in total or who are trying to manage their inventory through different types of seasonality or who are also trying to manage their inventory across multiple marketplaces.
Many of our largest sellers sell on other marketplaces in addition to Amazon. So they have a really big challenge to keep up with the demands of managing their inventory well. And so we have millions of unique products at Amazon today and yet, I can tell you we have an opportunity to add millions more. If we could help sellers to serve customers better, our customers will be happy, sellers will get to grow their business and of course that creates a great Amazon Marketplace.
And I think this is a game-changer, Leena, because if you think about the history of business, the only way you could experience it geographically was to maybe go plant the flag and open up a new office and add lots of capital and lots of overhead trying to figure out how to serve a new country. Being able to reach 10 countries on the Amazon Marketplace alone, plus customers from all over the world who shop those 10 marketplaces, is becoming a bigger and bigger opportunity for sellers.
Personalization and recommendations are creating significant impact in terms of the customer engagement and improving the overall lifetime value of your customers. For online shoppers, they are more likely to shop on a website that offers and makes personalized recommendations for them.
At first glance, matching users to items may sound like a simple problem to solve. However, the task of developing an efficient recommender system is extremely challenging and complex. Building, optimizing and deploying real-time personalization today requires specialized expertise in analytics, applied machine learning, software engineering, and systems operations. Few organizations have the knowledge, skills, budget and experience to overcome these challenges, and they often end up either abandoning the idea of using recommendation or build under-performing models.
For over 20 years, Amazon has built recommender systems at scale, integrating personalized recommendations across the buying experience — from product discovery to checkout. Amazon has made incredible personalization advances with its artificial intelligence, machine learning and predictive analytics to help all AWS customers do the same.
Amazon has recently launched Amazon Personalize which is a fully-managed service that puts personalization and recommendation in the hands of developers with little or no machine learning experience. Amazon Personalize allows the customers to create private, customized personalization recommendations that is built off of the customer data. With Amazon Personalize, you provide the unique signals in your activity data page views, signups, purchases, and so forth along with optional customer demographic information age, location, etc.
You then provide the inventory of the items you want to recommend, such as articles, products, videos, or music as an example. Then, entirely under the covers, Amazon Personalize will process and examine the data, identify what is meaningful, select the right algorithms, and train and optimize a personalization model that is customized for your data, and accessible via an API. All data analyzed by Amazon Personalize is kept private and secure and only used for your customized recommendations.
The resulting models are yours and yours alone. With a single API call, you can make recommendations for your users and personalize the customer experience, driving more engagement, higher conversion, and increased performance on marketing campaigns.
We at Capgemini are leveraging Amazon Personalize for our customers to automate and accelerate their machine learning development and drive more effective personalization at scale.
Amazon Personalize will process and examine your data, identify what is meaningful, allow you to pick a machine learning algorithm, and train and optimize a custom model based on your data.
This article will walk you through the process of configuring Amazon Personalize and integrating it into your Braze environment using Connected Content. This is done using a hands-on workshop that will walk you through all the steps required to deploy and train Amazon Personalize solutions, and then to integrate these solutions into a Braze email campaign using Connected Content.
These examples are deployed in a fully-functional example eCommerce site called the Retail Demo Store. You can use this reference architecture implementation as an outline to implement Amazon Personalize in your own environment. You will need to clone the Retail Demo Store repo and follow the steps outlined below to deploy the workshop environment to your AWS account.
An AWS account is required to complete the workshop and to run the integration code. Braze will send emails to your users based on their behavior or based on attributes of their user profiles. This data can be used to identify users and to build user profiles that can be used to determine when to send a message or email.
This event data flow will happen in parallel to the same behavioral event data being sent to Amazon Personalize. In this workshop, the demo store uses Amplify to send events to Personalize as shown in the diagram. This data is used to train a recommendations model that can in turn be used by Braze Connected Content to personalize content to users of your mobile and web applications when your Braze campaign runs.
Braze Connected Content will be able to get these recommendations via a recommendation service running in AWS. The Retail Demo Store workshop shows an example recommendation service deployment. In a deployment scenario in your own infrastructure, you will need to deploy a similar service in order to get items from your own catalog service. This list does not represent all the possible regions where you can deploy the project, only the ones that are currently configured for deployment with the Retail Demo Store.
Accept all the default parameter values for the template. The deployment of all the project resources should take minutes. Before you can provide personalized product recommendations, you first need to train the machine learning models and provision the inference endpoints that will allow you to get recommendations from Amazon Personalize.
The CloudFormation template deployed in Step 1 includes an Amazon SageMaker notebook instance that provides a Jupyter notebook with detailed step-by-step instructions. With the Amazon Personalize solutions and campaigns in place, your instance of the Retail Demo Store is ready to provide recommendations to your email campaigns. In Step 1 you also deployed the environment that contains the Retail Demo Store web application, and all associated services, including the recommendation service that you will need to integrate your email campaigns with Braze using Connected Content which will use the Amazon Personalize campaigns you deployed in Step 2.
Message Personalization. Contextual Location. Neura Actions. Dynamic Content. Movable Ink. Amazon Personalize.
0コメント