Skip to content

Introduction to Hello Retail

Hello Retail is a platform for adding personalization to webshops and providing AI-powered ecommerce insights. This allows the development of shop features that provide a great user experience and optimize basket size and conversion rate.

The following diagram provides an overview of the data that flows into Hello Retail from the shop and the data that can be returned to the shop.

Overview of the Hello Retail Platform

Most of the solutions can be implemented in one of the three following ways:

  1. You can use our JavaScript SDK
  2. The solutions can be implemented using our RestAPI
  3. You can use Hello Retails Managed Solutions

If you are using our RestAPI to implement the different solutions, you might still consider using the JavaScript SDK for tracking, as it automatically handles cookies and the tracking Ids. If you prefer to implement tracking serverside, we have a good guide for it.

The Managed Solutions are a simple way to get started, while still being able to customize the solutions heavily. It is built on top of the JavaScript SDK and has the templates embedded inside the MyHelloRetail platform. You can read more about it here.

Terminology and features

Below is an overview of some of the terminology used on My Hello Retail, in the documentation, and in the APIs.

Term Description
trackingUserId A string generated by Hello Retail identifying a customer on a single device. helloretail.js will automatically create a trackingUserId for every visitor, and save it in a cookie. This identifyer allows tracking a user on a single device across multiple requests.
customerId A string identifying a customer. A customer will often visit a shop on multiple devices and therefore have multiple trackingUserIds. The purpose of the customerId is to track users acrosss sessions and devices. Hello Retail will automatically assign a customerId to each visitor. If the shop has its own identifyer for a customer, for instance if they log in, it is possible to override this customerId using an API call. If the shop does not set a custom customerId then Hello Retail will try to identify customers across trackingUserIds, usually with an email address. This is often possible at checkout, where the email address is included in the conversion data.
Product A product in the shops product catalog. The product will have default attributes like title, url and price, but it can also have custom attributes suitable for the specific shop. Like for instance shoe sizes.
Content An item in one of the shop's content catalogs. Hello Retail supports four specific content catalogs; categories, brands, site pages, and blog posts. These are all identical except for their name. The content catalogs can be used in search. Like products they can also have custom attributes.
Cart An object tracking the current contents of the customer's cart. It is updated using the tracking API and is associated with the customerId. Read more
Conversion An object representing conversions/sales in the webshop. They are added to Hello Retail using the tracking API and are associated with the customerId. Conversions can also be added using order feeds to load historical data or to update already tracked conversions with further information. Read more
Page view Every time a page/product is viewed, it can be tracked as page views using the tracking API. The page views are associated with a customerId. Read more
Click When a customer clicks on a product in one of the Hello Retail solutions it is tracked using the tracking API. The content returned from the Hello Retail solution will include tracking codes that can be sent to the click tracking API. The clicks are associated with a customerId. Read more
helloretail.js Is a JavaScript that should be added to shops using Hello Retail. It provides automated tracking and it is the main way the managed features are implemented by Hello Retail. It also provides a JavaScript SDK for implementing the solutions.
My Hello Retail Is the control panel for your Hello Retail solutions and features. You access it on https://my.helloretail.com. It provides analytics, insights and controls for modifying your solution.

Solutions

The Hello Retail platform offers multiple solutions. Below follows a short description of each of them.

Product Intelligence Insights

Provides an API to get various Product Intelligence powered insights about a product. Given a product, this API will return information like:

  • What time of day this product is most often purchased
  • What discount percentage works well for this product
  • What products are often bought after this product
  • Etc.

You can read more about the API endpoint here

Product Recommendations

This solution provides personalized product recommendations. The algorithm used to recommend products can be highly customized on My Hello Retail. Some algorithms accept an input context, which affects what products will be recommended. For example, on a product page, it makes sense to show alternatives to the current product. To do this, the "alternative products" algorithm will accept the current product as an input and will use that to find the alternatives. On a cart page, it makes sense to use the current cart total as an input parameter and use that to determine what products to recommend. It often makes sense to recommend products that would bring the customer over the free shipping limit.

Product recommendations can be used through the JavaScript SDK and the Rest API, where a developer will render product recommendation boxes based on the returned product data. They can also be used in a managed form where the designs are created in the MyHelloRetail platform and injected to the site.

This solution provides an advanced search engine for ecommerce. It has features like:

  • Filtering/faceting
  • Sorting
  • Boost results based on product attributes
  • Personalized boost results based on customer preference
  • Synonyms and stopwords
  • Simultaneous search in alternative content types, like category pages
  • Initial content shown before searching
  • Zero result content, shown when no results are found
  • AI-powered synonyms
  • Redirects for quick access to important pages based on certain search terms

Search can be used through the JavaScript SDK or REST API, where you can build the UI to perform searches, apply filters, and render the result. It can also be used in a managed form, where Hello Retail designs and injects a search UI. If you choose to go with the managed templates, we have some good prebuilt templates that you can customize using Liquid. You can read more about the managed search templates here.

Pages

This solution provides a way to implement category, brand or landing pages with a filtering/faceting UI and where the ordering of products can be adjusted from My Hello Retail, and personalized to each customer. It has features like:

  • Filtering/faceting
  • Sorting
  • Boost results based on product attributes
  • Advanced analytics
  • Overrides so you can set up boosting for individual pages
  • Personalized boost results based on customer preference
  • Fixed products for specific categories
  • Flexibility for the developer on how products are categorized

Pages can be used through the REST API or JavaScript SDK where a developer will build a UI for showing products and applying filters. It can also be used in a managed form where Hello Retail designs and injects a pages UI. If you go with the managed template, you will get a template that has all the logic for filters, sorting, and more, and it can be customized using the Liquid template language. You can read more about this option here.

Newsletter Content

This solution provides a way of adding personalized product recommendations to emails sent through ESPs.

It works by adding dynamic images to the email. By adding the recipient's email address to the image URL using merge tags, each individual recipient has unique image URLs. Hello Retail then generates images with, for instance, the product title, image, and price inside the image.

The design of the generated images is configured on My Hello Retail, and the products recommended to the recipients are found using the same recommendation engine as the one used in "Product Recommendations."

Data Synchronization

The platform is powered by the data it gets from the webshops. This makes configuring good and accurate data synchronization an important first step when setting up a shop with Hello Retail.

Hello Retail broadly works with two types of data:

  1. Inventory Data
  2. Customer Behavior Data

Notice: Hello Retail provides extensions, plugins, and API integrations with many shop platforms. This means that in most cases, a developer will not need to do much work to get the data synchronization working for a shop. The description below serves as an overview of what data can be synchronized with Hello Retail, but it does not mean that the integration has to be done from scratch.

You can find our list of platform integrations and guides here.

Inventory Data

The primary inventory data is the catalog of products that are sold in the shop. Hello Retail needs this data to be able to: Perform searches on the text of the products Provide filtering options based on the various attributes of the products Return product data as part of the response to various API calls, like search Analyze the product catalog to extract insights in combination with the customer behavior data The product catalog is not a static data set. New products appear, old ones are removed, and attributes like prices are changed. It is therefore important that Hello Retail's view of the product catalog is continuously kept up to date. Hello Retail provides two different ways of achieving this.

Product Feeds

There are two ways of providing product data to Hello Retail, you can either provide a product feed (JSON, XML, CSV) that we can parse, or you can use our API to push the products to us.

If you use one of our integrations, they will automatically expose a product feed which can be digested easily through our feed system.

For our feed system we have built a rather comprehensive ETL that can modify the data coming from the feed. More information about this option can be found here.

If you want to push data to our system, you can find more information about it here.

There are a few required properties for a product:

  • URL (to the product)
  • Title
  • Image (url)
  • Price

You can, however, always supply many more fields. You can read more about the data for products here.

Content Feeds

Beyond the product catalog, Hello Retail also supports supplementary data types. These are similar to the product catalog in many ways, but can only be used in search. This allows synchronizing a list of category pages, content pages, blog posts, and brand pages to Hello Retail. Each of those catalogs can be searched through the search API, which is typically used to guide users to those types of pages when they search for something that is not a product.

The content can be synchronized to Hello Retail through feeds, where Hello Retail sends a periodic HTTP request to fetch an up-to-date version of the content. You can read more about content feeds here.

Customer Behavior Data

The platform also uses customer behavior data. This data is provided by calling the various tracking API endpoints. Hello Retail collects the following data types:

  • Product Views: The customer navigated to the product page
  • Clicks: The customer clicked something in one of Hello Retail's features
  • Carts: The current cart contents of the customer
  • Conversions: The customer completed a purchase
  • Searches: The customer searched. This is only required if you have purchased the search solution.

The data is also used to show analytics about how often your customers interact with Hello Retail features and how often they end up buying something based on the products shown by Hello Retail. Specifically for search, this also provides analytics about what the customer searched for and how often they clicked the results.

So, throughout all our solutions, we use the combination of this behavioral data and Product Intelligence to identify what products a user is most likely to buy. The benefit is that you have a coherent personalization experience throughout Hello Retail Solutions.

Conversion Feeds

Both the historic and current conversion data are important to understand customer behavior on your site. Therefore, it is also possible to provide conversion data through an order feed, just like the ones used for products and content. This is a good way to import historic conversions but also useful to ensure that Hello Retail sees all conversions, even if the customer has a browser that blocks the tracking calls.

You can read more about order feeds here