New Technology: The Projected Total Economic Impact™ Of Azure OpenAI Service In Reinventing Customer And Constituent Engagement

Profit Gain From Better Engagement With Existing Customers

Evidence and data. Interviewees shared that with Azure OpenAI Service applied at their organization’s contact centers and support organizations, agents can more easily pull up information while on the call, which allows them to quickly and better address the issue at hand. This can impact customer satisfaction levels and, by extension, the level of customer churn.

• The EVP at a retail and consumer goods organization described the improvement in customer service with the AI-enabled knowledge base. They said, “Our customer support agents and store associates can access our knowledge base to better support our customers.”
• The head of data at a retail and consumer goods company shared that their organization looks for ways to use its Azure OpenAI-based solution to positively improve customer engagement and, by extension, the overall experience with their organization. The interviewee noted: “Our NPSSM (Net Promoter Score) is currently negative 30. [With Azure OpenAI,] we want to bring that back to being positive within three months. One of our objectives with the personalized campaign is to decrease churn.”
• The head of engineering at a manufacturing and mobility company reported that their organization used Azure OpenAI Service to determine the sentiments of customers, which helped improve customer understanding: “Sentiment analysis can uncover pain points or challenges that cause churn, so we can proactively take action on it.”
• The head of data at a telecommunications company described the importance of improving their organization’s customer experience with Azure OpenAI Service: “Telco is an industry where people can easily switch providers. We have to make sure our customer experience is so delightful that customers stay with us, and we can reduce our churn. We are expecting a 10% uplift in our CSAT (customer satisfaction score) from empowering our support agents better.”

Interviewees reported that in addition to creating personalized content for prospective customers, their organizations were able to more effectively personalize materials for their existing customer bases. By being able to create tailored materials for specific customers, the organizations uncovered cross-sell and upsell opportunities, increasing the average revenue per customer.

• The SVP at a media company described the improvement in the engagement of existing customers with Azure OpenAI Service: “Once we get people into our [digital] space, we can personalize that engagement. That can turn into them buying merchandise, consuming our other content, and subscribing to our business.”
• The head of data at a retail and consumer goods organization explained that Azure OpenAI helped their company create content targeted to individual customers, which drove cross-selling opportunities: “The other part of personalized campaign is to increase sales. Cross-selling is rooted in us now understanding the personalized customer demand more.”
• The SVP at a media company reported that their organization used solutions built on Azure OpenAI Service to better understand customer sentiment, which helped it realize upselling opportunities. They said, “By being able to do sentiment analysis more effectively, we can uncover cross-sell and upsell opportunities.”
• The chief architect at the financial services and insurance organization told Forrester: “[From our improved engagement,] we can evaluate whether or not we are actually helping the client more. We can see if their portfolio with us grows. If I see 10% to 20% revenue growth that can be directly attributed to [Azure OpenAI Service,] that would be a good starting point for success to me.”
• Some interviewees even imagined evolving the role of contact center agents to become more of a guide when addressing a caller’s concern, which can include identifying additional buying opportunities. The head of data in telecommunications said, “With the time savings and efficiency gain, the hope is for agents to be more like advisors [to our customers].”

Modeling and assumptions. Based on the interviews, Forrester assumes the following about the composite organization:

• The composite’s annual revenue is $10 billion, and it grows at 2.6% each year.
• There is a gradual adoption of Azure OpenAI Service at the organization. In the low-case scenario, 5% of the organization is affected by Azure OpenAI Service use cases in Year 1, which then grows to 20% in Year 2 and to 30% in Year 3. In the mid-case scenario, adoption starts at 7% in Year 1, then grows to 25% in Year 2 and to 50% in Year 3. In the best-case scenario, adoption starts at 10% in Year 2, then grows to 40% in Year 2 and to 80% in Year 3.
• The composite’s annual customer churn rate prior to using Azure OpenAI Service was 10%.
• By adopting Azure OpenAI Service, in the low-case scenario, the organization improves its retention rate by 10% in Year 1, 15% in Year 2, and 20% in Year 3. In the mid-case scenario, this improvement is 15% in Year 1, 20% in Year 2, and 25% in Year 3. In the best-case scenario, the improvement is 20% in Year 1, 25% in Year 2, and 30% in Year 3.
• The adoption of Azure OpenAI Service also impacts the composite’s average revenue per customer. In the low-case scenario, this improvement to overall revenue is 1% in Year 1, 2% in Year 2, and 3% in Year 3. In the mid-case scenario, these percentages improve to 2%, 3%, and 5% respectively. And in the best-case scenario, they improve to 3%, 5%, and 7%, respectively.

Results. This yields a three-year projected PV ranging from $18.7 million (low) to $93.3 million (high).

Profit Gain From Better Engagement With Prospective Customers

Evidence and data. Interviewees shared that by using Azure OpenAI Service, their organizations were able to more easily generate personalized content, ensuring that marketing materials resonated with individual prospects. Additionally, the organizations were able to rely on AI-powered knowledge bases to better respond to prospects’ inquiries, which helped them avoid lost sales. Collectively, the increased personalization and the improved response cycles led to more prospects converting into sales.

• When asked how using Azure OpenAI Service improves conversion rates, the head of data at a retail organization noted, “With personalized newsletters, we can increase our web conversion rate by two times.”
• The senior director at a manufacturing and mobility company shared how the AI-powered knowledge base helped their organization better respond to inquiries while improving engagement and conversion: “[When evaluating a purchase,] customers want to understand how to use a product, what regulation is behind it, [and] how to learn about it. That can all get pretty complex. The idea is to help this process with AI.”
• When engaging with prospects, answer accuracy is expected to be a key determinant of customers’ purchase decisions. The head of data at a telecommunications company told Forrester, “We’ve seen the accuracy of response to customers increase by 30%.”
• The CTO at a financial services and insurance firm added: “If you think about the sales and marketing funnel of an organization, the ability to put insights into every stage of that journey, knowing who the person that’s viewing the insight [is], what their impact is on the client, what the client is thinking, [and] what they want, you can dramatically improve the interaction [with the client].”

Additionally, interviewees shared that AI-generated content is more optimized for search engines, which drives more prospects to their webpages. With the improvement in personalization, prospects who were directed to websites were also more likely to engage with the organizations’ sales teams.

• The data lead at a retail and consumer goods organization shared: “Personalized newsletters help drive interest into our websites and stores. Anecdotally, we saw a 200% increase in our top-of-funnel growth in certain markets.”
• The head of data at the same organization added: “Without understanding personalized customer demand, you end up doing major marketing campaigns on TV, radio, and other major media. Now, you can do this in a more targeted way, and it’s a cheaper effort.”
• The senior director at a manufacturing and mobility company noted: “Last year, we generated $300 million in revenue opportunities, which represents $1.2 billion in total contract value pipeline. Our marketing spend is measured in the amount of business outcomes we produce with that. We’re using [Azure OpenAI Service] to more intelligently understand customer signals and behaviors, including where the customer is in the buying cycle and what the buyer group is looking at. That should all lead to better performance at a lower cost.”
• The SVP at a media company told Forrester: “We can now localize any content we have to any market and country that we want to enter. We can adapt the content we write depending on different personas.”

Modeling and assumptions. Based on the interviews, Forrester assumes the following about the composite organization:

• Every year, the composite receives 4.2 million top-of-funnel opportunities. These include a mix of both digital and physical and inbound and outbound prospects.
• There is a gradual adoption of Azure OpenAI Service at the organization. In the low-case scenario, 5% of the organization is affected by Azure OpenAI Service use cases in Year 1, and this percentage then grows to 20% in Year 2 and to 30% in Year 3. In the mid-case scenario, adoption starts at 7% in Year 1, then grows to 25% in Year 2 and to 50% in Year 3. In the best-case scenario, adoption starts at 10% in Year 2, then grows to 40% in Year 2 and to 80% in Year 3.
• In the legacy environment, the organization converted 2% of all prospects into actual sales.
• In the low-case scenario, Azure OpenAI Service improves the sales conversion rate by 10% in Year 1, by 15% in Year 2, and by 20% in Year 3. In the mid-case scenario, the improvement is 15% in Year 1, 25% in Year 2, and 30% in Year 3. In the best-case scenario, the improvement is 20% in Year 1, 30% in Year 2, and 40% in Year 3.
• The organization’s revenue per customer is $15,000.
• The organization’s operating margin is 10%.
• In addition to the improvement in sales conversion rate, the use of Azure OpenAI Service also impacts the number of top-of-funnel prospects. In the low-case scenario, this improvement is 5% in Year 1, 8% in Year 2, and 10% in Year 3. In the mid-case scenario, the improvement is 8% in Year 1, 13% in Year 2, and 15% in Year 2. In the best-case scenario, the improvement is 10% in Year 1, 15% in Year 2, and 20% in Year 3.

Results. This yields a three-year projected PV ranging from $14.9 million (low) to $75.8 million (high).

Improvement In Contact Center Chatbot Resolution

Evidence and data. Interviewees shared how using Azure OpenAI Service enabled their organizations to deflect more contact center requests with their chatbots. This allowed human agents to focus on more complex engagements in which human intervention is crucial without overwhelming them with simple queries now handled by the chatbot. Having the AI-enabled chatbot also empowers agents to have all up-to-date information at their fingertips, which they can use to seamlessly address the issue at hand. Forrester research found that while contact center hosts massive amounts of unstructured customer and interaction data, they rarely have access to customer and interaction insights.⁴ This gap is easily filled by having AI-enabled chatbots that can access those data points and information as needed.

• The head of data at a telecommunications company shared how using Azure OpenAI Service reduced average handle time at their organization’s contact centers: “Agents spend a good amount of time in any customer contact searching through multiple documents. This impacts the time taken to have that contact and the caller experience overall. By having these Azure OpenAI Service [features,] the agent can find that answer quicker with the help of the chatbot. We hope to reduce agent handover from 36% to 30% in the future.”
• The AI lead supporting government agencies told Forrester, “A very popular use case [of Azure OpenAI Service in government] is empowering our existing frontline workers with some sort of pocket Wikipedia for different domains they work in.”
• The data and analytics lead at a healthcare company said their organization is working to develop a private chatbot to help patients independently resolve basic queries: “Ideally, we can present an OpenAI chatbot that has access to the patients’ information.”
• The head of AI in the education space said, “Our school admins use the [Azure OpenAI Service] internal chatbot to support them with required information in conversations with parents who have questions about their children.”
• The chief architect at a financial services and insurance organization noted: “We want to see if the efficiency gains [from using Azure OpenAI Service] mean our staff [will be] able to meet more clients. We also want to see whether there are impacts on the time spent per client as well as the time for prospective clients to make a decision. If we can improve those metrics by 30% to 50%, I would consider that good enough to start.”

Modeling and assumptions. Based on the interviews, Forrester assumes the following about the composite organization:

• The composite’s contact center handles 4 million calls each year. This number grows at the same rate as the overall organization.
• In the legacy environment, the contact center chatbot was already able to handle 50% of all calls without needing human intervention.
• For the remaining 50% of calls, the use of Azure OpenAI Service reduces 10% of calls in Year 1, 15% in Year 2, and 20% in Year 3 for the low-case scenario. This percentage reduction increases to 15%, 25%, and 35% respectively in the mid-case scenario, and to 20%, 35%, and 50% in the best-case scenario.
• The cost per call requiring human intervention is $10.

Results. This yields a three-year projected PV ranging from $7.5 million (low) to $17.5 million (high).

Go-To-Market Content Generation Efficiency

Evidence and data. Interviewees noted that the amount of content their organizations can create with Azure OpenAI Service is now significantly higher than it was before. In most cases, this meant their teams of content creators could essentially become content reviewers. Beyond creating more content, the organizations could now ask their generative AI tools to personalize short- or long-form content, ensuring brand consistency and achieving personalization at scale.⁵

• The head of AI in the education space explained the efficiencies their organization saw from using Azure OpenAI Service to help teachers create lesson plans: “We used to have educators write their own sample lesson plans. Creating one could take 2 hours for every hour of lesson. By switching the educators to more QA work, the time investment can take less than an hour per hour of lesson plan.”
• The EVP at a retail and consumer goods organization described the time savings their company saw in creating product descriptions and other website content: “We can get a lot more product descriptions produced with the same number of people. That content helps us with SEO, improving the natural ranking of our content’s searchability.”
• The SVP at a media company noted that having Azure OpenAI Service made their team more productive in generating content in different languages. They said, “It’s almost like giving us 50 virtual global translators to translate our content in real time to any language we want.”
• The AI lead supporting government agencies added, “Depending on the case, we are saving as much as 80% of our time by having Azure OpenAI Service review our documents.”

Modeling and assumptions. Based on the interviews, Forrester assumes the following about the composite organization:

• The composite has 500 FTEs involved in GTM content generation activities, and they spend 50% of their time relative to their other responsibilities.
• With Azure OpenAI Service, in the low-case scenario, these FTEs save 10% of their time in doing content generation work in Year 1. This improves to 20% in Year 2 and to 30% in Year 3. In the mid-case scenario, this improves by 15% in Year 1, 50% Year 2, and 55% in Year 3. And in the best-case scenario, this improves by 20% in Year 1, 55% in Year 2, and 60% in Year 3.
• The average fully burdened annual salary for an FTE involved in GTM content generation activities is $80,000.
• The composite’s productivity recapture rate is 50%. This measures the value of time saved by focusing only on the time spent on creating valuable work or actual cost reduction, rather than the overall time saved.

Results. This yields a three-year projected PV ranging from $4.8 million (low) to $10.9 million (high).

Unquantified Benefits

Interviewees mentioned the following additional benefits that their organizations experienced but were not able to quantify:

• Enablement for global operations related to content translation. For organizations with operations in different regions, the use of Azure OpenAI Service created speed-to-value benefits in translating content to a different language. Previously, a team of translators had to be deployed, and multiple revisions needed to be conducted depending on how much business context the translators had. With Azure OpenAI Service, as long as the organizations ensured the data was set up correctly in the back end, they expected their chatbots to capture context correctly and effectively in creating translated content. The head of engineering at a manufacturing and mobility company shared: “One of the main benefits of genAI is to break the barriers of language, which is important for a global company. Some datasets are written in local language. By having genAI translate this data, it enables sharing of information between business units that used to be extremely complex before. If you previously hired professional translators, there are direct cost savings.”

• Efficiency and productivity gains in product development and management. Interviewees also shared that using Azure OpenAI Service enabled faster connection between customer input and product development. This means customer feedback received by call centers or store associates could be incorporated into future product updates sooner. The chatbots could support engineers and other employees involved in building the product to access the latest customer feedback. The EVP at a retail and consumer goods company said, “It opens up a huge backlog of work that my engineers needed to do. I’m freeing up a significant portion of my engineers' time to focus on the next innovative idea.”

  The CTO at an investment management company added a similar sentiment of how having genAI capabilities through Azure AI Service allowed different roles in different departments pull insights themselves if needed. They noted: “People can now go further and do something themselves. If I have a great idea that requires some data engineering or manipulation, genAI can analyze that for me.”

• Boost in employee satisfaction and competing for talent. Interviewees shared that a key pillar of effective customer engagement is hiring and retaining key talent. The interviewees shared that having tools and capabilities enabled by Azure OpenAI Service is a key differentiator in their ability to attract talent. The EVP in retail and consumer goods added: “Engineering talent is very hard to find. Having the best talent is key to ensuring we can provide market-leading experiences to our customers. We can provide complete differentiations in our offerings.”

Flexibility

The value of flexibility is unique to each customer. There are multiple scenarios in which a customer might implement Azure OpenAI Service and later realize additional uses and business opportunities, including:

• Key enabler for future business growth. Interviewees shared that leveraging Azure OpenAI Service is imperative to remaining competitive in the market. The said time savings for marketing and customer success teams, faster development cycles for new use cases, and deeper customer insights empower their organizations to adjust engagement strategies and ultimately drive business growth. The EVP at retail and consumer goods company told Forrester, “The things that we're working on, the innovation we're driving, the experimentation we’re doing: All this allows us to capture more market share over time by driving the right experience for our customers as well as our associates.”

Flexibility would also be quantified when evaluated as part of a specific project (described in more detail in Appendix A).