The first in a series of investigations to determine the future of Loyalty Consulting
Your email inbox is buzzing with information about the impact of new AI driven search engines on your life. The impact will be on both your personal and business life, so the invitation to learn is compelling. If you haven’t experimented with Chat GPT from Open AI, you will very soon or risk being labeled as out of touch with the latest trends.
OpenAI is an AI research and deployment company. Their stated company mission is “to ensure that artificial general intelligence benefits all of humanity.” The backstory of ChatGPT Creator Sam Altman adds more context to the company’s mission.
Microsoft, Google and others have embarked on a race to put this amped up search capability to use. Microsoft is an investor in OpenAI and just announced a new version of its Bing search engine, which uses ChatGPT technology to understand questions and generate answers. Microsoft says that this new search engine runs on the next generation of OpenAI's language model, said to be significantly more capable than the version of ChatGPT that has been available since November 2022. Google’s search engine dominance is under threat and is not sitting still. Google has launched its own information site explaining how it is “Advancing AI for Everyone”.
The technology has quickly generated controversy. There is a widely circulated article written by New York Times technology columnist Kevin Roose that documents his two hour conversation with the Bing Chatbot that ended in the author feeling “deeply unsettled, even frightened” by the chatbot’s emergent abilities. A growing number of early users of ChatGPT or search engines based on that technology, have reported issues ranging from incorrect or incomplete answers to incorrect math calculations. Some echo the creepiness that Mr. Roose experienced. Microsoft, which has invested billions in OpenAI, defended its Bing upgrade as a work in progress.
In the recounting of his backstory, Mr. Altman acknowledged that the program’s outputs to date often contain factual errors. “It does know a lot, but the danger is that it is confident and wrong a significant fraction of the time” Altman wrote recently on Twitter.
We have been chatting with the machines, asking questions to see how valuable it can be to a brand marketer who wanted to practice some “DIY” loyalty strategy creation. We won’t cover everything we have been asking ChatGPT, but share this example to illustrate the level of accuracy and proficiency the chatbot has in answering basic questions related to customer loyalty.
We posed this question to Chat GPT as our first test:
Question #1:
“Please tell us the key elements to developing a loyalty program strategy”. Please note that the answer shown below omits detailed aspects of the response and is limited to top-line elements only.
Answer #1:
Developing a loyalty program strategy involves several steps to ensure that the program is effective and meets the needs of your customers. Remember that a successful loyalty program is one that provides value to both the customer and the business. By understanding your customers and tailoring the program to their needs, you can create a loyalty program that drives customer retention, engagement, and ultimately, revenue growth.
Here's a general outline of the steps involved:
- Define your objectives
- Identify your target audience
- Choose the right rewards
- Design the program structure
- Implement the program
- Measure and optimize
Question #2:
We next asked Chat GPT:
"What do you recommend to develop a loyalty strategy using cash back for a coffee chain.”
Answer #2:
Developing a loyalty strategy using cash back for a coffee chain could involve the following steps:
- Define your objectives
- Choose the right rewards
- Design the program structure
- Implement the program
- Measure and optimize
- Offer exclusive benefits
We conclude the following from this brief experiment:
- ChatGPT has access to the volume of content that all of us have been posting online for many years. It is no surprise that the general list of recommendations to create a loyalty strategy would ring true with most loyalty consultants. Would we all share the same list? No. There would be variation, embellishment, and customization. But the chatbot can regurgitate the foundation very well. As a research tool, it is powerful and time-saving.
- Drilling down to seek recommendations for a specific business type and loyalty structure, the chatbot left us wanting. There were five common elements among six total responses. It seems that general recommendations from the chatbot for the entire category of loyalty strategy are not versioned much for a specific business channel.
- When you seek examples of specific program elements, you won’t always find what you need. The cutoff date for the information in the ChatGPT database is September 2021, which means that it has access only to information up until that time. This recency issue will probably be cured at some point, but for now, it represents a weakness in using ChatGPT for specific needs.
The purpose of our investigation was not to bash or attempt to discredit ChatGPT. This is an incredibly powerful tool that will save time and add efficiency to business. We did want to know to what extent the AI engine could report on nuances of loyalty strategy or provide answers in sufficient depth to allow a retail team to create a strategy in lieu of recruiting outside assistance. There is a big gap on that front today, though we expect it to gradually close as the tool becomes more informed and more powerful.
Round One goes to the Loyalty Consultant.