Authors: Paul Harmon and Josh Fleischer
In November of 2022, ChatGPT, a groundbreaking new chatbot that produces remarkably natural and accurate text responses, launched. The sheer effectiveness of the tool made it a topic of significant discussion, and it very quickly became a well-known tool not only in the AI and machine learning world, but everywhere.
No doubt, interest in AI has been on the rise recently, outpacing interest in machine learning or statistics by a long shot. Google trends show a sharp uptick in searches for “artificial intelligence” that spiked substantially in the past year. This is due in part to the popularity of tools like ChatGPT, the popular image-generating AI DALL-E that rose to prominence in late 2021, and others.
Undoubtedly, AI (and related tools/technologies like data science, statistics, and machine learning) has become mainstream — and this may only be the beginning of mass adoption of AI by businesses, schools, and individuals who want to use tools to simplify their everyday lives.
Already, the effects of these tools are being felt. For instance, technology news outlet CNET recently revealed that they were using automated chatbot-type tools to write articles for several months dating back to November 2022.
Schools, both K-12 and universities, are having to rewrite rules around plagiarism and rethink their curriculums as they combat a growing problem with students turning in AI-generated work. In the business world, organizations looking to operate more efficiently and accurately are adopting an increasing number of AI-based tools into their standard operating procedures. For the enterprise, tools like ChatGPT can be an opportunity for radical change, both structurally and in terms of process.
The implications of ChatGPT in the enterprise
AI and related technologies help organizations make better decisions or automate processes at the enterprise level. This brings many benefits for organizations, but tools like ChatGPT can be potentially threatening to workers who see their jobs being automated away.
Striking a balance: ChatGPT and the human component at work
For many organizations, harmonization between investment in new technologies and investment in the fundamental skills of their employees is to create systems where AI can complement (not replace) human capital.
People are good at building trusted relationships with customers, connecting disparate concepts, and relating them together. They’re also good at breaking large problems down into smaller solvable ones. Computers (and AI in particular) are good at processing large amounts of data to identify trends, draw insights, and make predictions — be it in the form of numbers, images, or text. While AI-based tools are improving rapidly, it’s highly unlikely that AI capability will ever replace a person’s ability to process nuance, think creatively, or build trust.
It might be easy to think that ChatGPT was built overnight. For many, it certainly appeared that way after it was rolled out. However, the tool was created from a highly iterative process that involved supervised and unsupervised machine learning, validation from human stakeholders, no small amount of data, and many revisions, updates, and tweaks over time. Organizations that look to build similarly complex AI tools shouldn’t expect to build something as sophisticated as ChatGPT right out of the gate. Instead, they should focus on setting realistic goals first, instead viewing AI as a long-term investment that takes time, synergy, and cross-functional commitment to obtain effective results.
What should companies be doing today to take advantage of tools like ChatGPT?
In the face of an increase in use of automation, AI and machine learning, and data-driven decision-making, there is an ever-increasing need for organizations and their employees to be “data literate.” Organizations must realize that their employees — particularly the ones who help manage teams, optimize processes, and make critical decisions — are at a serious disadvantage if they aren’t armed with the ability to use data to make better decisions.
Uses of ChatGPT and AI for Optimizing Sales Processes
At Atrium, we work closely with customers who want to improve their sales processes, be it in financial services, medical fields, ad sales, higher education, or other industries. ChatGPT provides an automated way to provide conversational, human-esque responses in a chatbot, a quicker response, and the ability to remember context.
However, it does have limitations that should be considered prior to using. For instance, ChatGPT lacks an API for easy use. While this may change in the future, it makes it harder to automate today. Additionally, it is limited to 3K words of context, and its “awareness” is limited to its last training date. (This is likely always going to be a limitation.) Nevertheless, it provides a powerful tool when used correctly.
Here are a few examples of how we see ChatGPT (and other AI tools like it) potentially being useful to enable sales professionals to focus more on what they do best, helping to fuel a data-driven intelligent experience:
Efficiently balance sales and service
Tools like ChatGPT may be useful for helping organizations manage sales and service activities by providing their representatives with a more natural, accurate chatbot to interact with costumers who have routine questions, claims, or concerns. This frees up employees to tackle more complex cases, dig into more complex issues, or operate more efficiently in balancing sales and service tasks.
Enhance self-service in sales
Many companies have large numbers of potential customers who reach out to advisers with sales reps at different stages of the purchase decision. Often, reps may have to manage relationships with various customers with different needs. By enabling ChatGPT to answer routine questions and facilitate self-service, reps can focus on doing what they do best: building relationships with potential customers.
Expand solutions to multilingual environments
ChatGPT may be used to take insights generated by models and automatically translate them into different languages, or distill them into more business-friendly language.
Enable smarter business change enablement
As organizations roll out new technologies or processes, tools like ChatGPT may be helpful in providing organizations with the right information they need to disperse to end users, expediting the process of business change enablement and driving buy-in for strategic initiatives. ChatGPT might also be a useful tool for deciphering feedback from end users, administering surveys or questionnaires, and answering specific questions from end users.
Translate and improve communication
ChatGPT can be used to translate technical (relatively short) documentation, journal articles, and other documents into succinct summaries to facilitate more efficient, effective communication with stakeholders.
Create inputs to other AI initiatives
Organizations that go “all in” on AI and machine learning may be able to leverage tools like ChatGPT or DALL-E to generate net-new data that can be used to drive other machine learning initiatives, such as predictive analytics around opportunity/lead conversion, forecasting, or proactive identification of customer churn.
If used correctly, AI-powered tools should not simply displace employees or automate jobs away. Instead, these tools will enable technology to complement an organization’s greatest resource: its people.(And surprise — this post was not written by ChatGPT!)
Learn more about our data science consulting services and how we can help you identify revenue streams and grow faster.
