Creating Effective TTO Marketing Summaries Using ChatGPT

At tech transfer offices, you understand the critical role of technology transfer in bridging academia and industry. A key element of this transformation is the ability to craft compelling marketing summaries that effectively showcase the value of your new inventions to potential investors and industry partners.

This tutorial provides a step-by-step guide on how to use ChatGPT, a powerful language model, to create effective marketing pitches for technology transfer, particularly in the context of inventions stemming from academic or research institutions. I used ChatGPT 4 for this tutorial. Subscribe to get all of our free tutorials.

?

Summary of Steps to Create Your Marketing Pitch with ChatGPT:

1. Gather Material: Start by compiling all available information about the invention. This typically includes the invention disclosure submitted to the tech transfer office, any relevant scientific papers, and the patent application.
2. Understand Industry Challenges: Conduct research to understand the problems faced by the industry in relation to this type of technology. You can utilize ChatGPT for initial insights, but also read the materials you've gathered and consult with the inventors for a deeper understanding.
3. Identify Solutions Provided by the Invention: Determine how the invention addresses these industry-wide challenges. This step is crucial in highlighting the unique selling points of your technology.
4. Draft the Marketing Pitch with ChatGPT: Engage in a conversation with ChatGPT, inputting the information you've gathered. The goal is to refine the information into a compelling marketing pitch.
5. Editing and Finalization: Fine-tune the draft provided by ChatGPT. This might involve adding specific details or tweaking the language to better suit your target audience.

?

Case Study: Perovskite Solar Cells from Harvard University

I'll apply this process to a real-world example: a very cool innovation involving thin-film photovoltaics using perovskites from my alma mater, Harvard (see ref 1 for the link). This innovation is supported by an invention disclosure, a scientific paper (ref 2), and a patent application (ref 3).

I’m not reproducing the material here, as I think Harvard does a great job of presenting its innovations – it’s worthwhile to check out their website yourself. Briefly, the invention is for thin film photovoltaics that use perovskites (instead of silicon). Perovskites are transparent, light and efficient – but they have a relatively short lifetime, so they aren’t yet commercially useful.

I want to create a marketing pitch for this innovation from each of the above types of invention information. I could just start by entering the innovation information into ChatGPT – but that won’t provide good results. I’ll show you what happens when I just toss in the patent application and ask ChatGPT for a summary:

Yikes! This isn’t a marketing pitch. It sounds like it was written by a lawyer.

This isn’t the fault of ChatGPT – it’s my fault. I didn’t give it enough information so that it could produce the type of summary that I wanted.

Then I asked for a marketing pitch to a potential licensing partner – without adding any more details:

?

?

This is better but still not great. It doesn’t really sell the innovation to the potential licensing partners – just throws around some generic sales-type words and jargon.

Again, this isn’t the fault of ChatGPT. I haven’t given it enough information to be able to zero in on what I want. I want to show potential licensing partners that this innovation will solve their business problems and help them increase their revenues, while beating the competition. So I need to do a bit of research first. I can ask ChatGPT about the commercial problems for thin film photovoltaics that can be solved by perovskites.

?

Here is the answer from ChatGPT (reproduced as text to also show the links):

Perovskite solar cells present a promising solution to several commercial challenges faced by traditional thin-film photovoltaics. Here are some key aspects where perovskites could offer significant advantages:?

1. Efficiency: Perovskites have shown exceptional potential in terms of efficiency. They have achieved a record efficiency of 29.15%, which is higher than the highest efficiency recorded for crystalline silicon technologies. This high efficiency is especially notable considering that crystalline silicon technology is well-established and mature.?

2. Light Absorption: Perovskites can be tuned to respond to a wider range of wavelengths in the solar spectrum compared to crystalline silicon, which is limited to wavelengths equal to or superior to 1,100 nm. This ability to absorb a broader spectrum of light could lead to more efficient solar panels capable of converting more sunlight into electricity.

3. Cost-Effectiveness: The manufacturing cost for perovskite solar cells is currently on par with the lowest cost for crystalline silicon. As perovskite technology matures and mass-production processes are established, it's anticipated that the cost could drop to around $0.10 per watt, potentially making it one of the cheapest photovoltaic technologies.?

4. Versatile Applications: Perovskites hold potential for diverse applications, including residential, commercial, industrial, Building Integrated Photovoltaics (BIPV), tactical, and even space applications. This versatility stems from their unique properties and the ability to fabricate them into both rigid and flexible forms.

However, there are still challenges to be addressed for perovskite technology to become commercially viable. The major concern is the lifespan of perovskite solar cells, which currently lasts only about 30 months, significantly shorter than the 25-30 years lifespan of crystalline silicon cells. Addressing this durability issue is crucial for perovskite cells to be practical for real-world applications.

In comparison to other thin-film technologies like amorphous silicon (a-Si), cadmium telluride (CdTe), and copper indium gallium selenide (CIGS), perovskites offer a more efficient option, with only III-V GaAs thin-film technology currently surpassing them in efficiency.

In conclusion, perovskite solar cells have the potential to overcome several limitations of current thin-film photovoltaic technologies, particularly in efficiency and cost-effectiveness. However, their commercial success hinges on overcoming challenges related to lifespan and mass production: https://solarmagazine.com/solar-panels/perovskite-solar-cells/.

Check out these search results(https://chat.openai.com/backend-api/bing/redirect?query=commercial%20problems%20thin%20film%20photovoltaics%20solved%20by%20perovskites) to dig in deeper.

?

ChatGPT gave us lots of great information – BUT it is always important to check the links that ChatGPT provides and actually read the material. AI hallucinations are real, and ChatGPT can make mistakes while reading online material. Plus, the online material itself may not be reliable – you need to judge that for yourself.

?

I then asked ChatGPT to rewrite the pitch, based on the above material, summarizing the commercial promise of perovskites for thin film photovoltaics. Here is ChatGPT's answer (shown as text to fit it all in).?

Harness the future of solar technology with our cutting-edge perovskite solar cells, addressing key challenges in the thin-film photovoltaics market. Our patented method, US20230386759A1, is a breakthrough in solar cell efficiency, achieving an unprecedented 29.15% efficiency rate, surpassing traditional crystalline silicon cells. With the ability to absorb a broader spectrum of light, our perovskite cells promise superior performance in diverse applications, from residential to space technology.

Our technology is not just about superior performance; it's also about affordability. We're on track to reduce costs to as low as $0.10 per watt, making our perovskite solar cells one of the most cost-effective solutions in the market. This positions us uniquely to revolutionize the solar industry, offering both high efficiency and affordability.

However, we're not stopping there. We are tirelessly working to extend the lifespan of our perovskite solar cells, currently at 30 months, to meet and exceed industry standards. Our goal is to combine unparalleled efficiency, cost-effectiveness, and durability, making our technology the preferred choice for a wide range of solar applications.

Join us in this journey towards a sustainable future, where cutting-edge technology meets practicality and affordability. Let's power the world with perovskite solar cells, reshaping the energy landscape one panel at a time.

?

This answer from ChatGPT still has a number of problems. First, it is true of all perovskite solar cells generally, not the invention that I’m working on specifically. Plus, it doesn’t list all of the problems of current perovskite solar cells which the invention overcomes. And it also fails to adequately describe the actual invention and what makes it innovative.

I could simply ask ChatGPT to rewrite the marketing pitch to include all of the problems that the invention solves. However, in my experience, it is better to break this step up into two parts – first asking about the commercial problems generally, and then asking ChatGPT to rewrite the marketing pitch according to the problems solved.

?

I first asked ChatGPT to describe the current commercial problems for perovskite solar cells, which are blocking widespread commercial adoption of this technology. Here is ChatGPT's answer (shown as text to include the link):?

The widespread commercial adoption of perovskite solar cells faces several challenges. A primary issue is the stability of these cells. Despite significant advancements in efficiency, perovskite solar cells have yet to demonstrate long-term operational stability that matches the industry standards set by existing technologies like silicon-based cells. Enhancing their durability and resistance to environmental factors is crucial for their practical application in real-world scenarios. Addressing these stability challenges is essential for perovskite solar cells to become a viable, competitive commercial technology. https://www.nature.com/articles/s41578-022-00521-1

?

Again, it provided a link, which I read - and which demonstrates that the summary relates to a particular type of perovskite solar cell. However, I’ll assume that the problems are similar for all types of perovskite solar cells. Now I’ll ask it to rewrite the marketing pitch according to this information.

?

?It's still pretty generic, plus the actual invention is no longer described. Furthermore, there is a clear hallucination problem - US20230386759A1 is a patent application, not a patent.

I clearly need more information. Fortunately, there are clues in the patent application and also in the published paper about the specific problems solved by this specific invention.

The patent application states:

Previous CVD methods employed hydrogen halides (HX) as the halide source. However, HX can be damaging to vacuum equipment, and can also inflict damage upon certain substrates, such as substrates described herein. CuI stands out among the cuprous halides for its particularly promising electrical properties, but CVD of CuI using this method requires use of anhydrous hydrogen iodide. Anhydrous HI is difficult to source commercially and is difficult to store due to its facile, exergonic decomposition to H2 and I2.

In addition, due to purchasing and handling difficulties, HI may also be incompatible with perovskite absorber materials, which are particularly exciting substrates for device applications of CuI (e.g., in perovskite solar cells, smart windows, etc.). Hydrogen halide gases have been shown to undergo halide exchange with methylammonium lead trihalides at 120° C. Because the highest-performing halide perovskite PVs employ careful engineering of the balance of halides Cl, Br, and I in the light-absorbing perovskite layer to optimize their optoelectronic properties, these exchange reactions would lead to undesirable perovskite compositions and lower device quality.

Films are desirably continuous, pinhole-free, and have precisely controlled thickness and composition.

?

The paper states:

However, no prior CVD method could produce continuous thin films of any cuprous halide.

However, these prior methods typically produce films that have inadequate smoothness, purity, continuity, or large-area uniformity for commercial application in optoelectronic devices. The only known CVD or ALD routes to cuprous halides provide either islands or impure films. (the word “prior” was added for clarity)

?

I could also get this information by speaking with the inventors of course – which I strongly recommend doing.

?

In response, ChatGPT wrote (shown as text to fit it all in):

Introducing a groundbreaking innovation in perovskite solar cell technology: our patented method, detailed in US20230386759A1, revolutionizes the production of cuprous halide films, particularly CuI, through an advanced chemical vapor deposition (CVD) process. This technology circumvents the challenges of traditional CVD methods that used hydrogen halides, which were harmful to vacuum equipment and certain substrates, and required difficult-to-source anhydrous hydrogen iodide (HI).

Our method overcomes these hurdles by eliminating the need for HI, thereby enhancing compatibility with perovskite absorber materials and maintaining the carefully engineered balance of halides in high-performing perovskite photovoltaics. As a result, we achieve continuous, pinhole-free films with precise control over thickness and composition, essential for high-quality optoelectronic devices.

Most importantly, this innovative approach significantly increases the lifetime and durability of perovskite solar cells, addressing the primary commercial challenges of stability and operational longevity in the solar industry. Partner with us to harness this transformative technology, paving the way for more sustainable and efficient solar energy solutions.

?

You can see the hallucination problem continues. Plus I want to start with a catchier first sentence. So I asked ChatGPT to rewrite the first sentence, to focus on the problem that perovskite solar cells have a too short lifetime and are not sufficiently durable for commercial use; and to add a sentence that cuprous halides could be a solution, except for their problems (which should then be described).

?

?

You could also ask it to drop the mention of the US patent application, and make other changes if you wish through ChatGPT - or just manually edit. At this point, you should have a strong basis to put the marketing pitch together, so that you can show it to your inventors.

?

To summarize, I performed the following process for this tutorial:

1. Initial ChatGPT Output: Inputting just the patent application into ChatGPT yields a technically dense summary, not suitable for marketing purposes.
2. Revised Approach: Reframing the request to ChatGPT with a focus on potential licensing partners and emphasizing the commercial benefits of the invention produces a more targeted pitch.
3. Incorporating Industry Challenges: By understanding the commercial challenges for thin-film photovoltaics that perovskites can solve, we can refine the pitch to address specific industry needs.
4. Further Refinement: Analyzing the specific problems that this particular invention solves leads to a more customized and effective marketing summary.
5. Addressing Specifics and Avoiding Generalizations: It's important to focus on the unique aspects of the invention rather than making broad claims about the technology class it belongs to.
6. Final Edits: Use ChatGPT to fine-tune the marketing pitch, ensuring it is not only informative but also engaging and directly relevant to the target audience.

?

?

We plan to create a series of tutorials on using less expensive (or free) AI tools for tech transfer marketing. To subscribe, please sign up here: www.rocketsmart.io/subscribe-to-our-tutorials-on-ai-tools-for-tech-transfer

You can also contact me through LinkedIn for more information.

?

Please let me know which tutorials you’d like to see in the future.

?

References:

1.?????? Invention disclosure as published: https://otd.harvard.edu/explore-innovation/technologies/scalable-method-for-manufacture-of-thin-film-photovoltaic-devices/

2.?????? Scientific paper: https://pubs.acs.org/doi/abs/10.1021/acs.chemmater.0c04586

3.?????? Patent application: I selected a published patent application that covers at least similar material from Harvard, with Roy Gordon as an inventor. I do not know if this patent application is the one to which reference is made in the Harvard tech transfer office site, as it does not state a specific patent application number. US20230386759A1

4.?????? Link to patent application: https://patents.google.com/patent/US20230386759A1 (through Google to ensure free access to everyone)

?

?

Image credit: DALL-E

?

?