AI assistants empower personalized and effective support that will save time and money, and enhance creativity and quality.
Building your own AI assistants or GPTs (generative pre-trained transformers) opens up new avenues for every role, title, job, and skill in the architecture, engineering, and construction industry. With personalized GPTs, you can “control” the entire AI process from prompting to the knowledge base to the instructions to the output parameters. AI is now in the palm of your hand.
I recently gave a hands-on demonstration specifically for marketers on how to build a GPT within OpenAI’s ChatGPT platform. You can find the link to the online seminar here.
By building your own GPT-based chatbot, you can:
- Access customization options.
- Save and automate your process.
- Develop personalized training of the GPT.
- Focus on repeatable tasks.
- Fine-tune its dedicated knowledge base on your own data, which leads to better performance for your specific use case. It also offers an added layer of confidence in the outputs (accuracy and validity) since you give it the detailed directions and the source of its information.
The value of GPTs in AEC marketing. At the heart of any successful marketing strategy is the ability to connect, engage, and convert. For the AEC industry, where projects are vast and stakeholders diverse, “building” AI assistants empower personalized and effective support to your marketing team that will save time and money (and sanity!) and enhance creativity and quality. Notice I said “will” and not “can” or “should.” I am that certain.
Demystifying GPTs: The architects of conversation. GPTs are designed to understand, respond, and interact in ways that mimic human communication. Imagine having a digital assistant capable of drafting project descriptions, enhancing resumes, and even engaging in real-time chat for conference swag ideas. For AEC marketing, this translates into elevated conversations with your CRM database to know more about your clients’ needs, streamlined proposal drafting, and new insights about your services and market sectors, all generated from vast, personalized data sets into precise answers specifically for you.
Developing your own GPT: A step-by-step approach. The journey to creating a specialized GPT begins with a clear understanding of your objectives. Simply, what do you wish to achieve? What do you need help with the most right now? Once you have your idea, and the goal is set, the following steps outline the process:
- Data gathering. The foundation of an effective GPT lies in the data it learns from. It can obviously be connected to the internet, however uploading your own files to train your GPT will lead to better outputs. You want to load up its virtual library with accurate, clean, meaningful, and rich information.
- Training. With your data ready, the next phase is training your GPT. This involves feeding it the collected data and providing detailed, adapted instructions. The more descriptive you can be, the more you are allowing it to learn from the patterns and terminologies specific to its purpose. Use the GRACES framework by giving it goals, roles, audience, context, examples, and steps. And don’t forget the tried-and-true who, what, how, where, and when.
- Customization and testing. Tailor your GPT to reflect your voice and tone, the technical specifics of your needs, and the types of queries it will handle and how. Continuous testing and refinement are crucial to ensuring your GPT truly delivers on its function. (Sometimes, I even ask the GPT what else it “needs” to be the best, and then I incorporate its recommendations into its configuration.)
- Deployment. With initial testing complete, your GPT is ready for use. You can keep it private; share by invitation-only with a link; or publish on the GPT Store to a public audience. Integrate your new AI assistant into your daily workflow wherever it serves your objectives best.
- Evolution. Regularly update your GPT based on new data, feedback, and technological advancements to keep it functioning at its best and highest use.
Real-world applications: Enhancing client engagement. The examples I showed during the seminar included several GPTs that I have built for myself:
- Marketing Framework Wizard (This GPT is in the public GPT Store and can be found here. Give it a try and let me know what you think! I built it to help marketers generate better content using proven frameworks.)
- PIE Business Development Planner (This GPT is in the public GPT Store and can be found here. Give it a try and let me know what you think! I built it to help anyone in AEC work through a BD process from planning to implementation to evaluation. It is trained on my proprietary PIE process, presentations, handouts, and workbook.)
- Interview Summarizer (Helps me to find themes and insights from transcripts with clients and from my own recordings.)
- Network Navigator (I built our CRM in this. We can have conversations with our client and vendor contact data.)
- AI in AEC Marketing (This specifically helps us find new and creative ways to market our AI intelligence/specialty.)
- Write Like Kristin (I love this. I have trained it on samples of my writing so that it can mimic my style to generate new content on any topic.)
- RFQ Master (Brilliant. This is trained to analyze proposal requirements and generate not only specific technical details, but winning strategies as well.)
- Competitor Analysis (Exactly what it says. I have trained it to generate differences and similarities between firms and outline value-driven differentiators that my clients can use to win work.)
Continuous improvement. As with any powerful tool, GPTs come with some maintenance. Ensuring the AI’s responses remain aligned with your needs and brand and continuous learning from interactions and updated knowledge sources are areas requiring vigilant oversight. Yet, the time you dedicate to creation or maintenance will pale in comparison to the amount of time you reclaim when using AI. Your GPTs will deliver unparalleled opportunities to create, understand, grow, learn, and succeed in ways you didn’t think possible. Again, notice I said “will” and not “can” or “should.” I am still that certain.
Embracing the future of GPTs in AEC. AI is here to stay, and what’s next is now. The integration of GPTs into marketing activities within the AEC industry heralds a new era of better (more tailored and personalized!) client engagement in every aspect of the project life cycle – from relationships to business development to proposals/sales to technical design to delivery. The future of the AEC industry, powered by GPTs, is not just about automation but about creating deeper, more meaningful work and connections to our clients, our fellow co-workers, and each other.
The AI marathon has begun. Are you walking, running, or sprinting? Even at the very beginning of this race, we already have the power to build our own AI assistants. Right now. It’s here, I’ve done it, and I can show you! You can use this technology today, take advantage of its real capabilities, and tailor it to your unique narrative.
As we move forward, remember, the true value lies not in the technology itself but in how we leverage it to automate or augment tasks so we can get to the more important stuff in life, like the people, ideas, or causes we care about. (This is my definition of AI!) And as a testament to the ingenuity and vision that defines the AEC industry, we have to take the first step across the starting line. The course is ours to shape.
Zweig Group’s AI & AEC Tech Summit in Atlanta May 22-23 will connect AEC industry pioneers to explore and collaborate on the transformative impact of technological advancements shaping the field. Learn more here!
Kristin Kautz, CPSM is Zweig Group’s artificial intelligence (AI) consultant and trainer. Learn more about Kristin here.
