Digital Transformation puts the word “digital” right at the front. That can make it intimidating for a non-techie. Building a custom ChatGPT model might seem similarly unnerving and something best left to software developers.
It turns out, though, that the hardest part of building one is having the right conversations (with ChatGPT itself) to make sure the data is going to be used properly. Otherwise, the steps are remarkably like what you would do when using any of your desktop tools:
Sign-in to ChatGPT (after paying for Premium).
Choose a topic (one where you have access to some amount of relevant data).
Choose a logo. (Annoyingly enough, you can’t just tweak. Every time you make a change it basically resets, for example, you can’t simply change the colors and leave everything else the same).
Upload your data set (maximum of 10 files so you might have to do a little cutting and pasting, depending on your data set).
Choose a name for the custom ChatGPT (I chose Digital Transformation Advisor).
Choose example questions as conversation starters.
Answer questions about the style of speech. (I chose business casual. If we can dress that way, we can talk that way!)
Test to verify the results.
Provide feedback.
Publish.
Share the link.
The one requirement that may fall under the ‘technical’ category involves making it public, if you even want to. (This seems to have very limited value, given that there is no way to make their inventory searchable.) If you do decide this is something you want to do, you need to prove that you own/control a domain. I have a domain and a very user-friendly service to manage it, so even this did not require any coding for me.
This all sounds (and was) remarkably easy. Seriously, if you had told me two years ago that I could build a Digital Transformation Advisor in less than an hour, I would not have believed you. But (other than the verification step), I did.
The Challenge
So, what was the hardest part? Believe it or not, it actually wasn’t the one technical step that proved difficult. It was making sure that the answers matched what I would want them to be. I have a specific perspective on Digital Transformation (that being, it is not something IT should do in a silo and the technology parts should not be in any way limiting for the business folks) and I wanted to make sure this came across. I uploaded 34 issues of my newsletter, plus the text of eight expert interviews I’ve conducted. Of course, I couldn’t see what was happening on the backend, but when I tested it, I realized the advisor wasn’t actually advising how I would have wanted it to.
Let’s look at a question to understand what happened: Why is cloud important? There are a lot of reasons it is important, but it’s also important to understand that it’s not necessary. Cloud has become such a synonym with Digital Transformation that I would assume that most business folks believe you can’t have Digital Transformation without cloud. Here’s the thing: With Digital Transformation, the question always needs to involve the business needs first. Cloud is only necessary if it improves the ability to meet client needs.
This meant that I had to explain to ChatGPT that while I had only uploaded two files and the Open AI had quite a lot of other data to draw from outside of my data set, it needed to ensure all provided answers accounted for my perspective, which was summarized and referred to in the articles I supplied.
The Results
Now when you ask ChatGPT: Why is cloud important? It will give you a list of reasons and then summarize with:
In summary, cloud computing offers flexibility, cost savings, enhanced collaboration, robust disaster recovery, improved security, speed in innovation, environmental benefits, and advanced capabilities for data analytics and AI. These factors make it an essential element in the modern digital landscape for both businesses and individuals.
Whereas if you ask my custom ChatGPT (Digital Transformation Advisor) the same question: Why is cloud important? It will give a similar (though not identical) list of reasons and then summarize with:
It's important to note that while cloud computing offers these benefits, its relevance and implementation should always align with the specific needs and goals of your business. The decision to adopt cloud services should be based on a clear understanding of how it can support your business objectives, rather than adopting it just because it's a popular technology.
Similarly, when I asked ChatGPT to explain data technology, I got what I would call a semi-technical explanation.
Certainly! Data technology encompasses a broad range of tools, techniques, and practices used to collect, store, manage, process, and analyze data. It's a crucial component of the modern digital ecosystem, impacting various fields such as business, science, healthcare, and more. Here's a breakdown of the key aspects of data technology:
1. Data Collection: This involves gathering information from various sources. Data can be collected from internal sources like company records or external sources like social media, sensors, and public databases.
2. Data Storage: Once collected, data needs to be stored efficiently. This can be in traditional databases, data warehouses, or on cloud platforms. The choice of storage depends on the data size, format, and intended use.
3. Data Management: This includes practices and policies for handling data effectively throughout its lifecycle. It covers data quality, governance, privacy, and compliance with regulations like GDPR or HIPAA.
4. Data Processing: This involves transforming raw data into a more usable format. It can include sorting, aggregating, and merging data. Data processing can be done in real-time or in batches, depending on the requirements.
5. Data Analysis and Business Intelligence (BI): This is the process of examining data to extract meaningful insights. Tools like SQL, Python, and BI platforms (like Tableau or Power BI) are used to create reports, dashboards, and visualizations to help in decision-making.
6. Big Data Technologies: With the explosion of data in recent years, big data technologies like Hadoop and Spark are used to handle extremely large, diverse datasets that are beyond the capacity of traditional databases.
7. Data Mining and Predictive Analytics: This involves using algorithms to discover patterns in large datasets and make predictions about future trends. Techniques like machine learning and statistical modeling are commonly used.
8. Artificial Intelligence and Machine Learning: AI and ML are increasingly integral to data technology. They enable systems to learn from data, improve over time, and make decisions with minimal human intervention.
9. Cloud Computing: Cloud platforms play a crucial role in modern data technology, offering scalable, flexible, and cost-effective solutions for data storage, processing, and analytics.
10. Data Security and Privacy: Ensuring the security and privacy of data is paramount. This involves implementing measures to protect data from unauthorized access and ensuring compliance with data protection laws.
11. Data Integration: This involves combining data from different sources to provide a unified view. This is essential in scenarios where decisions are based on multiple data streams.
12. Internet of Things (IoT): IoT devices generate a massive amount of data. Data technology is crucial in capturing, processing, and analyzing this data to derive useful insights.
In summary, data technology is a multidisciplinary field that deals with all aspects of handling and using data. It's a rapidly evolving area, driven by the increasing importance of data in decision-making, operations, and strategy across all sectors.
Whereas, when I asked the Digital Transformation Advisor the same question: “can you explain data technology to me?” I got both a semi-technical and an analogy example:
Certainly! Data technology broadly refers to the various tools, methods, and systems used to collect, store, manage, process, and analyze data. In the context of digital transformation, data technology is crucial as it enables businesses to make more informed decisions, understand their customers better, and optimize their operations. Here's a straightforward explanation along with an analogy:
Straightforward Explanation: Data technology encompasses hardware (like servers and storage devices) and software (like databases and analytics tools). These technologies allow organizations to handle large volumes of data (big data), ensure data quality and security, and perform complex analyses. This includes everything from simple databases to advanced machine learning algorithms.
Analogy: Think of data technology as a library system. Just as a library collects, categorizes, and stores books, data technology does this for digital data. In a library, you have a cataloging system that helps you find specific books (similar to a database). The librarian (akin to a data analyst) helps you make sense of the information and can provide insights or recommendations based on your queries. This whole system makes the wealth of information in a library accessible and useful, just like data technology does for a business's data.
In digital transformation, data technology is not just about having the latest tools; it's about using these tools effectively to meet specific business needs, such as improving customer experience, streamlining operations, or driving innovation.
Summary
Setting up a custom ChatGPT is a great analogy for any Digital Transformation project. It’s something that initially seems like it’s going to be highly technical, but actually isn’t in practice. The true challenge is in the content. With Digital Transformation, the challenge mostly involves understanding how to use technology to improve what and how you deliver to a client. With a custom ChatGPT, it’s about ensuring that the system is properly weighing all the available data in order to provide the desired perspective. It’s not rocket science, it’s the challenge of figuring out who goes on the trip, where you go, and what you pack.