Background

At IKEA we have multiple places in our customer journey in various channels where different kinds of personalization can deliver a superior customer experience. Product recommendations in the shopping basket, content recommendations in editorial sections, inspirational recommendations on product pages and more. After a while in the broader “recommendations” team there was a decision to split the team to have one sub-team focused on product recommendations. The pandemic altered customer behavior and needs as well. At that inflection point we decided to change our way of working and dive head-first into a more scientific approach to handle the operational complexities of delivering high quality product recommendations at scale. We deemed this necessary to improve our level of personalization and to have a holistic understanding of our customers.

Data Driven Decisions

The first step was to radically improve our ability to get high-quality quantitative information to understand how our ‘recommendation’ solutions affected personalization. We did this through high volume A/B testing on customer behaviour and after initial experimentation, we had a few key learnings:  

1. The mix of both UX and algorithms are really important for a cohesive customer experience. 
2. The quality of personalization can’t be measured in silos. Statistical significance  can be attained  by testing several groups of recommendations at once.

Once we came up with a solid framework for gathering data and acknowledged how little we knew about our customers, we were able to explore an incredible number of creative options – nothing was off the table. This was a very humbling experience, in that it opened up new perspectives for personalization, a more curious and less confined way of thinking. We learned to trust the data because it might show you things you don’t expect. 

Experimentation and Learning Framework

Our teams created ways to quickly deploy experimental modifications to our existing solution. This enabled experimentation in the front-end with the user experience, including details in headings and images. This also covered tweaks in the backend with anything from detailed manual additions or removals of recommendations to mixing and matching of various algorithms both home grown and from Recommendations AI.

This flexibility came with an overhead–more complexity and cost relative to directly retrieving recommendations from Recommendations AI. However, the benefit was that we were no longer dependent on manual evaluation of what made for a good recommendation system. We aligned on a data-driven and qualitative approach to provisioning recommendations and significantly accelerated our experimentation timeline. Together with optimization of the CI/CD pipeline this enabled the team to take an idea or hypothesis from inception to A/B testing with customers in less than half an hour.

Recommendations AI Experiments

Our team’s infrastructure was already running on GCP and when we received early access to Recommendations AI, the requirements to get started were minimal and that allowed us to start with initial tests requiring minimal effort and investment.

We started with a few use-cases and identified places where our existing recommendation algorithms needed improvement or complementary recommendations. We also explored additional ways where more useful information could be presented to the customers through personalized recommendations. 

Recommendations AI Model Combinations

While Recommendations AI might be considered a simple API to get a set of product recommendations, as we dove deeper into the solution it became apparent that it could be tweaked in several different ways to offer many fine tuning configurations to meet business goals. While too much fine tuning and customization could lead to subpar performance, in general we found that it was a great strategy to give us several versions of ML powered recommendations to work with. The further you personalize the experience, the more options you have to likely pick the best one for the customer.

Recommendations AI models like  ‘Recommended for you’, ‘Frequently Bought Together’ and ‘Others you may like’; are coupled with business goals like optimizing for conversion rate, click through rate and revenue. We experimented with many different model combinations and custom rules. All this was easily configurable right in the GCP console. One of the simplest custom configurations we used was to only recommend items that were in stock, and when items were out of stock we looked at similar items that were available to augment the experience. 

Collaboration with Google

Our collaboration with Google Cloud accelerated our learning process during experimentation. We worked closely together early in the product development. Additionally, their model provided flexibility to change direction and allow for more options than we had previously. Ultimately, this provided us a way to drastically improve our time to market with a product that produced tremendous results that we could not have accomplished on our own.

Results and Takeaways

With more personalized and real-time recommendations available we saw great success. We were able to increase the number of relevant recommendations displayed on a page by +400%. To accommodate the wider repertoire of recommendations we had to change the user experience. For example, in some places we had horizontally scrolling displays of product recommendations which were much easier for customers to use.

Another consequence of displaying more personalized recommendations was tangible improvement to conversion rate and average order value. Recommendations AI algorithms helped customers in two ways: 

1. Customers were able to find products that they liked quickly and establish their preferred choice among other options more quickly as well, giving them confidence to make a purchase through much fewer clicks. Even though we previously already had well tuned recommendations of several types, with Recommendations AI we measured +30% improvement in click through rates. 
2. Average order value saw a +2% surge with numerous examples of how Recommendations AI could help customers find both attractive and directly complementary products, expanding the customer purchase from a single product to an entire home furnishing solution.

As a direct effect of having stronger business results, the team started exploring more places in the customer journey where our growing buffet of recommendations could be used. We’d start with an initial experiment to answer if displaying recommendations in the specific context made sense at all. Frequently the data that emerged from these experiments prodded us to iterate further on what additional types of recommendations would be most appropriate to show to the customer as the customer’s behaviour evolved. Today, most of IKEA’s site recommendations are powered by Recommendations AI.

One key takeaway is that for some types of personalized recommendations there are benefits to using advanced algorithms that require a lot of high level data science and engineering competence to build since they outperform simplistic approaches. In some places, simplistic approaches work very well and in others the right decision is to not have product recommendations at all. For an effective use of product recommendations you need to have all the above options and the ability to tell when to use which one.

Next steps

When working with something so tightly related to customer experience, there is a constant change in user behaviour and new learnings to observe and adapt to. Product recommendations are rarely the main stand alone experience and frequently something that is used to help and enhance an experience. We see a lot of value in having a large toolbox of possible options and a team with a relentless focus on collaboration to improve the customer experience. We’re working directly with the Recommendations AI team and experimenting with several new features that we’re excited about. 

In the future we see opportunities of improving the customer journey through a more visual experience that inspires the customer rather than relying on customers to use their imagination to visualize groups of products together. Vision Product Search provides that and is something we’re looking into deploying next. We’ll be sharing more about our journey with Recommendations AI at the Google Cloud Retail Summit session ‘IKEA’s Approach to Building a Powerful Recommendations Engine’ on July 27th 2021.

Best wishes to all developers from the IKEA product recommendations team & the Google Recommendations AI team!

In the past year, Google has worked to surface timely and reliable health information, amplify public health campaigns, and help nonprofits get urgent support to Indians in need. Now, we are continuing to focus on helping India’s businesses accelerate their digital transformation, deepening our commitment to India’s digitization and economic recovery. To support customers and the public sector in India and across Asia Pacific, we’re excited to announce that our new Google Cloud region in Delhi National Capital Region (NCR) is now open. 

Designed to help both Indian and global companies alike build highly available applications for their customers, the Delhi NCR region is our second Google Cloud region in India and 10th to open in Asia Pacific. 

What customers and partners are saying

Navigating this past year has been a challenge for companies as they grapple with changing customers demands and economic uncertainty. Technology has played a critical role, and we’ve been fortunate to partner with and serve people, companies, and government institutions around the world to help them adapt. The Google Cloud region in Delhi NCR will help our customers adapt to new requirements, new opportunities and new ways of working, like we’ve helped so many companies do in the region: 

• InMobi scaled a personalized AI platform to support 120+ million active users. “With the arrival of the Google Cloud Delhi NCR, InMobi Group sees the opportunity to continue closing the gap between our users and products,” says Mohit Saxena, Co-founder and Group CTO of Inmobi. “Glance, especially, has been serving AI-powered personalised content to over 120 million active users. We can’t wait to continue giving them truly meaningful experiences that are speedy, scale well, and are relevant to them, by expanding the use of our current tools working on Google Cloud with the opening of a new region.”
• Groww now supports a sizable user base. “Google Cloud provides great technology that enables us to build and scale infrastructure to millions of users, and the new Google Cloud region in Delhi NCR will continue to help more businesses and startups in India access powerful cloud-based infrastructure, products and services,” says Neeraj Singh, Co-founder and Chief Technology Officer, Groww.
• HDFC Bank is positioned for the future. “At HDFC Bank, we are harnessing technology platforms to both run and build the bank. As we progress to be future ready, the objective is to invest in future technologies that give us scale, efficiency and resiliency. Towards this the Google Cloud region in Delhi NCR will enable us to enhance our resiliency and help us in building an active-active design framework for our new generation applications on cloud,” says Ramesh Lakshminarayanan, CIO, HDFC Bank.  
• Dr. Reddy’s Lab built a modern data platform with Google Cloud. “At Dr Reddy’s, we pride ourselves in helping patients regain good health, acting quickly to provide innovative solutions to address patients’ unmet needs and in accelerating access to medicines to people worldwide. Our Google Cloud-powered data platform is helping us realize these objectives and we welcome Google’s investment in the new Delhi NCR region as helping us and other businesses in India make further contributions to our social and economic future,” says Mukesh Rathi, Senior Vice President & CIO, Dr. Reddy’s Laboratories.
• “To survive the disruption caused by the pandemic and to succeed in the long term, organizations need to become digital natives, so they can be more agile, explore new business models and build new capabilities that boost resilience. A cloud-first strategy plays a key role in enabling businesses to do this,” said Piyush N. Singh, Lead – India market unit & lead – Growth and Strategic Client Relationships, Asia Pacific and Latin America, Accenture. “Harnessing the potential of cloud requires the right data infrastructure and this expansion by Google Cloud will undoubtedly help Indian enterprises in their digital transformation journeys.”

A global network of regions

Delhi NCR joins 25 existing Google Cloud regions connected via our high-performance network, helping customers better serve their users and customers throughout the globe. As the second region in India, customers benefit from improved business continuity planning with distributed, secure infrastructure needed to meet IT and business requirements for disaster recovery, while maintaining data sovereignty.

With this new region, Google Cloud customers operating in India also benefit from low latency and high performance of their cloud-based workloads and data. Designed for high availability, the region opens with three availability zones to protect against service disruptions, and offers a portfolio of key products, including Compute Engine, App Engine, Google Kubernetes Engine, Cloud Bigtable, Cloud Spanner, and BigQuery. 

Supporting India’s recovery with training and education

Google and Google Cloud will also continue to support our customers with people and education programs. We’re investing in local talent and the local developer community to help enterprises digitally transform and support economic recovery. 

Through the India Digitization Fund, we expanded our efforts to support India’s recovery from COVID-19—in particular, through programs to support education and small businesses. In addition to expanding internet access, and investments to help start-ups accelerate India’s digital transformation, we’ve grown our Grow with Google efforts. Businesses can access digital tools to maintain business continuity, find resources like quick help videos, and learn digital skills—in both English and in Hindi.

Helping customers build their transformation clouds

Google Cloud is here to support businesses, helping them get smarter with data, deploy faster, connect more easily with people and customers throughout the globe, and protect everything that matters to their businesses. The cloud region in Delhi NCR offers new technology and tools that can be a catalyst for this change. To learn more, visit the Google Cloud locations page, and be sure to watch the region launch event here.

Data science can be applied to business problems to improve practices and help to strengthen customer satisfaction. In this blog, we address how the addition of Looker extends the value of your Google Cloud investments to help you understand your customer journey, unlock value from first-party data, and optimize existing cloud infrastructure.    

Data is a powerful tool 

While everyone doesn’t need to understand the nuts and bolts of data technologies, most people do care about the value data can create for them — how it can help them do their jobs better. Within the data space, we are seeing a trend of ongoing failure to make data accessible to “humans” –   the industry still hasn’t figured out how to put data into the hands of people when and where they need it, how they need it. 

What if everyone in your organization could analyze data at scale, and make more-informed, fact-based decisions?   Data and insights derived from data are valuable but only if your users see it.  We think Looker helps solve that. 

Solutions tell a larger story of how it all fits together 

Across all verticals and industries,  businesses benefit from knowing and understanding their customers better. Many business goals are to increase revenue by improving product recommendations and pricing optimization, improving the user experience through targeted marketing and personalization, and reducing churn while improving retention rates.  To help reach  these goals, key strategies should focus on understanding customer needs, their motivations, likes and dislikes, and using all available data – in other words, put yourself in the shoes of your customer.    

Our goal with Looker solutions is to offer the right level of out-of-box support that allows customers to get to value quickly, while maintaining the necessary flexibility. We aim to offer a library of data-driven solutions that accelerate data projects. Many solutions include Looker Blocks (pre-built pieces of code that accelerate data exploration environments) and Actions (custom integrations) that get customers up and running quickly and lets you build business-friendly access points for Google Cloud functionality like BQML, App Engine and Cloud Functions.  

Below, you’ll find a sampling of the newest Looker solutions. 

Listening to customers by looking at the data

Looker’s solution for Contact Center AI (CCAI), helps businesses gain a deeper understanding and appreciation of their customers’ full journey by unlocking insights from all their company’s first-party data. Call centers can converse naturally with customers and deliver outstanding experiences by leveraging artificial intelligence. CCAI‘s newest product —CCAI Insights — reviews conversations support agents are having, finding and annotating the data with the important information, and identifying the calls that need review. We’ve partnered with the product teams at CCAI to build Looker’s Block for CCAI Insights, which sets you on the path to integrating the advanced insights into your first-party data in Looker, overlaying business data with the customer experience.

Businesses can better understand contact center experiences and take immediate action when necessary to make sure the most valuable customers receive the best service. 

Realizing full business value of First-Party data

Looker for Google Marketing Platform (GMP) provides marketers the power to unlock the value of their company’s first-party data to more effectively target audiences.  The Looker Blocks and Actions for GMP offer interactive data exploration, slices of data with built-in ML predictions and activation paths back to the GMP.  This strategic solution continues to evolve with the Looker Action for Google Ads (Customer Match), the Looker Action for Google Analytics (Data Import) and the Looker Block for Google Analytics 4 (GA4).

• The Looker Action for Customer Match allows marketers to send segments and audiences based on first-party data directly into Google Ads. Reach users cross-device and across the web’s most powerful channels such as Display, Video, YouTube, and Gmail.  The entire process is performed within a single screen in Looker, and is able to be completed in a few minutes by a non-technical user. 
• The Looker Action for Data Import can be used to enhance user segmentation and remarketing audiences in Google Analytics by taking advantage of user information accessible in Looker, such as in CRM systems or transactional data warehouses.
• The Looker Block for Google Analytics 4 (GA4) expands the solution’s support with out-of-the-box dashboards and pre-baked BigQuery ML models for the newest version of Google Analytics.The Looker Block offers up reports with flexible configuration capabilities to unlock custom insights beyond the standard GA reporting. Customize audience segments, define custom goals to track and share these reports with teams who do not have access to the GA console.

From clinical notes to patient insights at scale

Taking a look at the Healthcare vertical, the Looker Healthcare NLP API Block serves as a critical bridge between existing care systems and applications hosted on Google Cloud providing a managed solution for storing and accessing healthcare data in Google Cloud. The Healthcare NLP API uses natural language models to extract healthcare information from medical text,  rapidly unlocking insights from unstructured medical text and providing medical providers with simplified access to intelligent insights.  Healthcare providers, payers, and pharma companies can quickly understand the context and relationships of medical concepts within the text, such as medications, procedures, conditions, clinical history, and begin to link this to other clinical data sources for downstream AI/ML.   

Specifically, the natural language processing (NLP) Patient View (pictured below) allows you to review a single selected patient of interest, surfacing their clinical notes history over time. It informs clinical diagnosis with family history insights, which is not currently captured in claims, and captures additional procedure coding for revenue cycle purposes.

The dashboard below shows the NLP Term View which allows users to focus on chosen medical terms across the entire patient population in the dataset so they can start to view trends and patterns across groups of patients. 

This data can be used to: 

• Enhance patient matching for clinical trials 
• Identify re-purposed medications
• Drive advancements for research in cancer and rare diseases
• Identify how social determinants of health impact access to care

Managing Costs Across Clouds

Effective cloud cost management is important for reasons beyond cost control — it provides you the ability to reduce waste and predictably forecast both costs and resource needs. Looker’s solution for Cloud Cost Management offers quick access to necessary reporting and clear insights into cloud expenditures and utilization. 

This solution brings together billing data from different cloud providers in a phased approach: get up and running quickly with Blocks optimized for where the data is today (Google Cloud, AWS or Azure) as you work towards more sophisticated analysis for cross-platform planning and even cloud spend optimization with the mapping of tags, labels and cost centers across clouds.

The Looker Cloud Cost Management solution provides operational teams struggling to monitor, understand, and manage the costs and needs associated with their cloud technology with a comprehensive view into what, where and why they are spending money.

Making better decisions with Looker-powered data

Leading companies are discovering ways to get value from all of that data beyond displaying it in a report or a dashboard. They want to enable everyone to make better decisions but that’s only going to happen if everyone can ask questions of the data, and get reliable, correct answers without using outdated or incomplete data and without waiting for it. People and systems need to have data available to them in the way that makes the most sense for them at that moment.  

It’s clear that successful data-driven organizations will lead their respective segments not because they use data to create reports, but because they use it to power data experiences tailored to every part of the business, including employees, customers, operational workflows, products and services.   

As people’s way of experiencing data has evolved, more than ever before, dashboards alone are not enough. You can use data as fuel for data-driven business workflows, and to power digital experiences that improve customer engagement, conversions, and advocacy. 

From Nov. 9 – 11th, Looker is hosting its annual conference JOIN, where we’ll be showing new features, including how we help to:

• Build data experiences at the speed of business
• Accelerate the business value with packaged experiences 
• Unleash more insights for more people in the right way – Deliver data experiences at scale

There is no cost to attend JOIN.  Register here and learn how Looker helps organizations build and deliver custom data-driven experiences that goes beyond just reports and dashboards, scales and grows with your business, allows developers to build innovative data products faster, and ensures data reaches everyone.

As Googlers transitioned to working from home during the pandemic, more and more turned to chat-based support to help them fix technical problems. Google’s IT support team looked at many options to help us meet the increased demand for tech support quickly and efficiently. 

More staff? Not easy during a pandemic. 

Let service levels drop? Definitely not. 

Outsource? Not possible with our IT requirements. 

Automation? Maybe, just maybe…

How could we use AI to scale up our support operations, making our team more efficient?

The answer: Smart Reply, a technology developed by a Google Research team with expertise in machine learning, natural language understanding, and conversation modeling. This product provided us with an opportunity to improve our agents’ ability to respond to queries from Googlers by using our corpus of chat data.  Smart Reply trains a model that provides suggestions to techs in real time. This reduces the cognitive load when multi-chatting and helps a tech drive sessions towards resolution.

In the solution detailed below, our hope is that IT teams in a similar situation can find best practices and a few shortcuts to implementing the same kind of time saving solutions. Let’s get into it!

Challenges in preparing our data

Our tech support service for Google employees—Techstop—provides a complex service, offering support for a range of products and technology stacks through chat, email, and other channels. 

Techstop has a lot of data. We receive hundreds of thousands of requests for help per year. As Google has evolved we’ve used a single database for all internal support data, storing it as text, rather than as protocol buffers. Not so good for model training. To protect user privacy, we want to ensure no PII (personal identifiable information – e.g. usernames, real names, addresses, or phone numbers) makes it into the model.

To address these challenges we built a FlumeJava pipeline that takes our text and splits each message sent by agent and requester into individual lines, stored as repeated fields in a protocol buffer. As our pipe is executing this task, it also sends text to the Google Cloud DLP API, removing personal information from the session text, replacing it with a redaction that we can later use on our frontend. 

With the data prepared in the correct format, we are able to begin our model training. The model provides next message suggestions for techs based on the overall context of the conversation. To train the model we implemented tokenization, encoding, and dialogue attributes.

Splitting it up

The messages between the agent and customer are tokenized: broken up into discrete chunks for easier use. This splitting of text into tokens must be carefully considered for several reasons:

• Tokenization determines the size of the vocabulary needed to cover the text.
• Tokens should attempt to split along logical boundaries, aiming to extract the meaning of the text.
• Tradeoffs can be made between the size of each token, with smaller tokens increasing processing requirements but enabling easier correlation between different spans of text.

There are many ways to tokenize text (SAFT, splitting on white spaces, etc.), here we chose sentence piece tokenization, with each token referring to a word segment. 

Prediction with encoders

Training the neural network with tokenized values has gone through several iterations. The team used an Encoder-Decoder architecture that took a given vector along with a token and used a softmax function to predict the probability that the token was likely to be the next token in the sentence/conversation. Below, a diagram represents this method using LSTM-based recurrent networks. The power of this type of encoding comes from the ability of the encoder to effectively predict not just the next token, but the next series of tokens.

This has proven very useful for Smart Reply. In order to find the optimal sequence, an exponential search over each tree of possible future tokens is required. For this we opted to use beam search over a fixed-size list of best candidates, aiming to avoid increasing the overall memory use and run time for returning a list of suggestions. To do this we arranged tokens in a trie, and used a number of post processing techniques, as well as calculating a heuristic max score for a given candidate, to reduce the time it takes to iterate through the entire token list. While this improves the run time, the model tends to prefer shorter sequences. 

In order to help reduce latency and improve control we decided to move to an Encoder-Encoder architecture. Instead of predicting a single next token and decoding a sequence of following predictions with multiple calls to the model, it instead encodes a candidate sequence with the neural network.

In practice, the two vectors – the context encoding and the encoding of a single candidate output – are combined with dot product to arrive at a score for the given candidate. The goal of this network is to maximize the score for true candidates – e.g. candidates that did appear in the training set – and minimize false candidates.

Choosing how to sample negatives affects the model training greatly. Below are some strategies that can be employed:

• Using positive labels from other training examples in the batch.
• Drawing randomly from a set of common messages. This assumes that the empirical probability of each message is sampled correctly. 
• Using messages from context.
• Generating negatives from another model.

As this encoding generates a fixed list of candidates that can be precomputed and stored, each time a prediction is needed, only the context encoding needs to be computed, then multiplied by the matrix of candidate embeddings. This reduces both the time from the beam search method and the inherent bias towards shorter responses.

Dialogue Attributes

Conversations are more than simple text modeling. The overall flow of the conversation between participants provides important information, changing the attributes of each message. The context, such as who said what to whom and when, offers useful bits of input for the model when making a prediction. To that end the model uses the following attributes during its prediction:

• Local User ID’s – we set a finite number of participants for a given conversation to represent the turn taking between messages, assigning values to those participants. In most cases for support sessions there are 2 participants, requiring ID 0, and 1.
• Replies vs continuations – initially modeling focused only on replies. However, in practice conversations also include instances where participants are following up on the previously sent message. Given this, the model is trained for both same-user suggestions and “other” user suggestions.
• Timestamps  – gaps in conversation can indicate a number of different things. From a support perspective, gaps may indicate that the user has disconnected. The model takes this information and focuses on the time elapsed between messages, providing different predictions based on the values. 

Post processing

Suggestions can then be manipulated to get a more desirable final ranking. Such post-processing includes:

1. Preferring longer suggestions by adding a token factor, generated by multiplying the number of tokens in the current candidate.
2. Demoting suggestions with a high level of overlap with previously sent messages.
3. Promoting more diverse suggestions based on embedding distance similarities.

To help us tune and focus on the best responses the team created a priority list. This gives us the opportunity to influence the model’s output, ensuring that responses that are incorrect can be de-prioritized. Abstractly it can be thought of as a filter that can be calibrated to best suit the client’s needs. 

Getting suggestions to agents

With our model ready we now needed to get it in the hands of our techs. We wanted our solution to be as agnostic to our chat platform as possible, allowing us to be agile when facing tooling changes and speeding up our ability to deploy other efficiency features. To this end we wanted an API that we could query either via gRPC or via HTTPs. We designed a Google Cloud API, responsible for logging usage as well as acting as a bridge between our model and a Chrome Extension we would be using as a frontend.

The hidden step, measurement

Once we had our model, infrastructure, and extension in place we were left with the big question for any IT project. What was our impact? One of the great things about working in IT at Google is that it’s never dull. We have constant changes, be it planned or unplanned. However, this does complicate measuring the success of a deployment like this. Did we improve our service or was it just a quiet month?

In order to be satisfied with our results we conducted an A/B experiment, with some of our techs using our extension, and the others not. The groups were chosen at random with a distribution of techs across our global team, including a mix of techs with varying levels of experience ranging from 3 to 26 months. 

Our primary goal was to measure tech support efficiency when using the tool. We looked at two key metrics as proxies for tech efficiency: 

1. The overall length of the chat. 
2. The number of messages sent by the tech.

Evaluating our experiment

To evaluate our data we used a two-sample permutation test. We had a null hypothesis that techs using the extension would not have a lower time-to-resolution, or be able to send more messages, than those without the extension. The alternative hypothesis was that techs using the extension would be able to resolve sessions quicker or send more messages in approximately the same time.

We took the mid mean of our data, using pandas to trim outliers greater than 3 standard deviations away. As the distribution of our chat lengths is not normal, with significant right skew caused by a long tail of longer issues, we opted to measure the difference in means, relying on central limit theorem (CLT) to provide us with our significance values. Any result with a p-value between 1.0 and 9.0 would be rejected. 

Across the entire pool we saw a decrease in chat lengths of 36 seconds.

In reference to the number of chat messages we saw techs on average being able to send 5-6 messages more in less time. 

In short, we saw techs were able to send more messages in a shorter period of time. Our results also showed that these improvements increased with support agent tenure, and our more senior techs were able to save an average of ~4 minutes per support interaction.

Overall we were pleased with the results. While things weren’t perfect, it looked like we were onto a good thing.

So what’s next for us?

Like any ML project, the better the data the better the result. We’ll be spending time looking into how to provide canonical suggestions to our support agents by clustering results coming from our allow list. We also want to investigate ways of making improvements to the support articles provided by the model, as anything that helps our techs, particularly the junior ones, with discoverability will be a huge win for us.

How can you do this?

A successful applied AI project always starts with data. Begin by gathering the information you have, segmenting it up, and then starting to process it. The interaction data you feed in will determine the quality of the suggestions you get, so make sure you select for the patterns you want to reinforce.

Our Contact Center AI allows tokenization, encoding and reporting, without you needing to design or train your own model, or create your own measurements. It handles all the training for you, once your data is formatted properly. 

You’ll still need to determine how best to integrate its suggestions to your support system’s front-end. We also recommend doing statistical modeling to find out if the suggestions are making your support experience better. 

As we gave our technicians ready-made replies to chat interactions, we saved time for our support team. We hope you’ll try using these methods to help your support team scale.