You’re tasked with delivering an exceptional software demo. Now what? Research!

Generative AI, Demos, and Data

As discussed in the series introduction, delivering a compelling software demo isn’t just about showcasing features within your software products—it’s about crafting a narrative that resonates with a prospect’s unique business needs. This series will explore how ChatGPT aids in transforming the demo and presentation creation process. This article focuses on using ChatGPT and our collaboration to accelerate and expand the research phase.  

As a reminder, the narrative for this series is that a team was tasked with delivering a demo for an RFP issued by a third-party logistics (3PL) company. There was little prior knowledge of their industry, no data, no discovery calls, and one week to prepare. Enter ChatGPT, a key partner in this process. More than just an AI tool, ChatGPT has evolved into a trusted collaborator, especially when it comes to understanding the prospect’s company, industry, key business issues, economic and competitive environments, and analytic maturity. Only by understanding these things can a software architect generate relevant and validated data to power our demos.  

The LLM also helped with demo data creation, demo flow, report suggestions, and the best way to illustrate the capabilities and value of predictive modeling. 

Let’s Dig into Research

Research is defined as “the systematic investigation into and study of materials and sources in order to establish facts and reach new conclusions.” Traditional search techniques such as Google and Bing produce a vast set of URLs that may or may not contain the facts to support research. The search process is fraught with challenges, and the time spent navigating recommended pages can be overwhelming.  

ChatGPT and its peers, such as Claude, Gemini, Meta, Llama, Copilot, and others, have truly revolutionized “research” according to this definition, and LLMs allow us to take research to a new level for complex tasks. It's important to remember to cross-reference and confirm the content provided by these LLMs and look for bias that might be embedded in the results, but when with, say, a publicly available annual report, the results should be reliable enough to start the demo creation process. 

When creating a demo, basic research about a company using traditional methods is achievable to a certain extent. A common starting point is to maneuver through a company's website, read the introduction page, and look at the popup menus that provide access to information such as Products, Services, Employees, etc... Then browse each page via dropdown lists, take notes to the best of your ability, and figure out a few key points you should use in your demo. If the demo creator is ambitious, they can go to financial sites and, if public, enter the ticker symbol for some basic market data. ZoomInfo can also be useful when researching both public and private companies. You could download and read an 80-page Annual Report, but, in the end, the amount of time invested in these approaches to gain a small amount of usable context far outweighs the benefit. 

Building the Foundation. Understanding the 3PL Industry and our Target Company. 

Next are examples of insights gained using ChatGPT for research that dramatically exceeds what could have been achieved using traditional techniques to prepare for demo creation and delivery, especially when given a quick turnaround. It’s truly a game changer and, as I’ve discussed, reimagines the demo creation process. 

Understanding Industry Business Drivers 

The 3PL industry (third-party logistics) is highly competitive. Operational efficiency and customer satisfaction drive success. From ChatGPT, I quickly gathered key business drivers that shaped the demo narrative to include: 

• Scalability 

• Operational efficiency 

• Customer satisfaction 

How ChatGPT Helped? 

I used ChatGPT to ask targeted questions about the logistics industry and this particular company, surfacing insights on real-time tracking trends and operational benchmarks. For instance, asking, “What are the key performance metrics and drivers for 3PL companies?” allowed us to highlight shipment accuracy and on-time delivery as critical metrics. This research gave a clear direction for the demo, and it will enable us to show how your software product can help optimize these metrics through predictive analytics and real-time tracking. 

 Identifying the Company’s Core Challenges 

After gaining a better understanding of the broader industry, I used ChatGPT to explore specific pain points and challenges faced by the prospect. By asking additional targeted questions about the logistics space, I was able to unearth common hurdles like: 

• Managing transportation costs 

• Warehouse optimization 

• Customer expectations 

• Fleet management inefficiencies 

• Delivery delays and cost overruns 

• Warehouse stock management issues 

I framed questions like “What are the common challenges for mid-size 3PL companies?” ChatGPT provided relevant, up-to-date information about logistics pain points, which allowed me to focus part of a demo on these exact challenges. As a glimpse into what will be discussed in later articles, this research set us up to show how a software product can reduce transportation costs by optimizing delivery routes and predicting inventory needs across fulfillment centers. By honing in on real-world pain points, the demo became much more than a generic showcase—it became a solution and a message that would resonate with the company’s specific use cases, issues, etc. 

Annual Reports: [Almost] everything you’ll need to drive the design phase 

Reviewing a company’s entire website and annual report and mapping that research to actionable information to be used in the demo creation process is an almost impossible task. ChatGPT combines research, insight, and broad context of the LLM foundation model to provide actionable recommendations across the entire spectrum of industry and company. It is these recommendations that enable the subsequent steps in the demo creation process to move forward smoothly and efficiently in ways that were previously not possible. 

As is now a somewhat common practice for LLM users, I uploaded the prospect’s annual report to ChatGPT. Asking for insights into the company based on the uploaded document is a practice that provided valuable information such as that below, as well as examples of the type of information specific to our 3PL target customer. 

• General and industry-specific metrics 

Total revenue, key contributions from segments such as value-added services, intermodal services, dedicated transportation services, brokerage services, capital expenditures and cost structure.

• Current business challenges 

Pricing pressure, consolidation in the trucking and logistics industries, advances in technology, many standard industry metrics lag competitors in the same revenue range, diesel fuel costs, labor costs.

• Competitors and market presence 

Market presence has stagnated and competitors offering more flexible and cost-efficient services as well as more modern data services.

• Opportunities for growth 

Expansion into the automotive industry as well as vertical markets such as aerospace, energy, health care, and retail.

It’s readily apparent how the foundation model is used to extrapolate beyond and complement what is in the Annual Report. But treating ChatGPT as a trusted advisor yields more meaningful utilization of the foundation model and more insightful and useful results specifically relevant to our task of building a compelling demo for our logistics company.

• Understand where predictive analytics can add value 

ChatGPT assessed which predictive models would be most useful in addressing some of their lagging metrics and what variables would be most valuable in building these models. This is critical when building the synthetic data set in the Design phase. 

• Benchmark against industry metrics 

It’s useful to look across many corporate metrics as they relate to their competitor and industry metrics. This was incredibly easy with ChatGPT and would have been near impossible with traditional means. I instructed it to pull the 20 most common industry metrics for their top 10 competitors in their revenue range, normalize those metrics for those competitors, and compare them. Gathering this type of information allows for creating synthetic data of this nature and makes a competitor dashboard robust and valuable. 

• How should I best demonstrate predictive analytic capabilities? 

ChatGPT provided invaluable guidance on how to leverage predictive tools to build the demo so that it is relevant and useful to the customer. It recommended specific reports and pages that would organize data logically and resonate at different levels of a 3PL company. Secondly, it recommended how decisions should be organized and delivered to decision-makers. Finally, it was specific on how the different metrics could and should be calculated based on other data. This will inform my synthetic data progress as I move into the Design phase. 

These insights are just a few of the things gleaned in the initial interaction with ChatGPT. ChatGPT can map the reports back to different user personas, business teams, and planning activities such as forward-looking CapEx planning.  

Collaboration and Iteration: Using ChatGPT as a Bridge 

ChatGPT’s role in refining the demo sharpened focus and research questions. Whether it was validating industry assumptions or brainstorming future demo structure, ChatGPT helped keep ideas aligned, despite working remotely. 

ChatGPT collaborate effectively and became the connective tissue throughout the process and a tool for comparing notes, helping to:

• Synthesize key insights from the logistics industry, helping us align on the most important pain points to address. 

• Brainstorm demo features that would address the client's needs, ensuring the right balance of business and technical insights. 

• Validate assumptions about our prospect’s priorities, allowing us to refine our approach continually. 

ChatGPT wasn’t just a research assistant but an essential team member. Generating rapid insights and refining ideas enhanced our collaboration and significantly accelerated the process. 

Laying the Groundwork for the Next Steps 

With the research phase complete, the demo creator is ready to move on to the next steps in the demo creation process. Many phases of demo creationoverlap, and I will navigate through each as there is a lot to discuss and unpack. 

• Design: Outline the demo flow and create a tailored dataset that reflects the company’s operational challenges, from managing transportation costs to improving real-time shipment visibility. 

• Create: Using ChatGPT to help generate synthetic data that mirrors real-world logistics scenarios and is surfaced from all the research including the company’s annual report noted above. 

• Iterate: Gathering feedback from key stakeholders at the company to refine the demo’s narrative and structure. 

• Deliver: Presenting a final demo that highlights the impact on the company’s bottom line. 

Lessons Learned During The Research Phase 

Through this process, I learned several key lessons and best practices about teaming up with ChatGPT in the research phase: 

1. Time-saving and comprehensive insights: ChatGPT allowed me to understand this company and the broader logistics industry in a fraction of the time it would have taken through manual research. 

2. Refined focus on pain points: By asking targeted and strategic questions, I could hone in on the company’s biggest challenges, key business drivers, and industry nuances helping to lay the foundation to craft a demo that specifically addressed these considerations. 

3. Collaboration accelerator: ChatGPT became an invaluable partner in aligning thoughts and ideas to stay on track. 

4. Cross-Referencing: While ChatGPT provided accurate and relevant information, it's important to remember to cross-reference and confirm the content provided by these LLMs and look for bias that might be embedded in the results. 

Looking Ahead 

The research phase is just the beginning and a crucial step in accelerating and laying the groundwork for the demo and presentation. In my next article, I'll explore how ChatGPT helps set the stage for the process' Design and Creation phases as I begin to create and explore synthetic data.