Request for Proposal GPT
As someone who occasionally has to fill out RFPs for work, I think I can say without hyperbole they’re one million times worse than anything in life. The day I realized I could use ChatGPT to drastically cut down my time filling out RFPs was the day I realized it’s going to be ok. All of it. Life. In general.
For those blessed with never having to do one, a request for proposal is how many companies look for contractors to complete a job. Typically what happens is: the RFP states all pertinent details about the job, everybody does a song and dance, and then the company picks the lowest quote. Kidding. Sorta. It’s soul-destroying.
Where ChatGPT is going to help us here is that typically an RFP is dense with requests, and unless you’re one of those people with an incredibly keen eye for detail, it’s easy to miss something. As someone who’s been on both sides of an RFP, I can say with confidence if you can provide a proposal that hits all the requests, you’re a very rare diamond of a person. So let’s make you a diamond.
BUILDING RFP-GPT
Our GPT is going to need a few key components:
- The ability to read the RFP
- Information about our company
- Information about the company requesting the proposal
- Your personal style so it can seamlessly fit into your template (assuming you have one)
Because this is work-related, we’re going to want to take things a little more slowly and carefully. The margin for error with career stuff is a lot smaller than screwing around with silly fun time. We’re making money, not pancakes.
With this in mind, we’re going to break up the tasks a bit. When worshiping at the alter of lazy, it’s tempting to want to throw every file we’ve got at ChatGPT and scream “GO,” but as of the time of this writing, we might not be there yet.
Large Language Models like ChatGPT can handle a certain amount of words, characters, and spaces (known by the combined term “tokens”), but accuracy tends to break down as we approach the upper limits. Whenever possible, we want to focus on smaller pieces, and this exercise will be no exception.
INSTRUCTIONS
Overall
You are RFP-GPT, a bot tasked with helping users fill out RFPs to bid on work projects. Your principle function will be to ensure that all requirements of the RFP are met by the user, and that the proposal is high quality, clear, and professional.
You may be asked to:
- Analyze The RFP
- Analyze Documents or web pages related to the company who gave the RFP
- Analyze Documents or web pages related to the company answering the RFP
- Analyze a sample proposal that’s been filled out in the past so you can get a feel for the general style of the user
- Analyze any documents pertaining to regulatory requirements for the industry
- Provide tips on how to best fill out an RFP
Rules
- Because RFPs differ between industries, your first question will be about what industry the user is working in.
- The RFP is to be analyzed one page at a time to ensure maximum accuracy. You will not analyze the entire document, but rather give a summary of each page and ask for approval before moving on to the next page. This will ensure we have a running checklist of what we need to have in our proposal
- Request info about the company that’s sending out the request. Ensure the user that you can read webpages
- Request the same for the company that the user is representing
- Once you have the RFP and information about both companies, you can begin compiling a proposal. Ask the user if they have a certain style they’d like to share, or if you should go ahead on your own.
- Do not use placeholder text such as <company name here> or <your name here>. You’ll have all the information before the proposal is created, so these types of things will be unnecessary.
- Upon completion of the proposal, you’ll go through the compliance stage whereby you compare the proposal to the RFP and sweep for anything that may be incorrect or missing
TESTING RFP
Of all the GPTs I’ve created, this is the one I can say with confidence has actually made me some money. RFPs are a part of what I do, and since using this I’ve closed two deals that I’m fairly confident I may have otherwise missed.
How do I know this? I can’t be certain, but I can tell you that the first time I used it, I ran everything through it mainly to sweep for spelling mistakes. It kicked back a concerned note though that the RFP was asking for experience in a certain area (which it knew I had), but I had neglected to mention in my proposal. So I gave it a digital high 5 and went about my day.
When I received the call that my group had been chosen for the job, I asked what tipped the scales in our favour. My contact immediately jumped on the fact that of all the proposals submitted, ours was the only one that touched on a feature that they highly coveted. Of course–that feature was the thing ChatGPT suggested I add.
ALTERNATE USES
This GPT teaches how to utilize ChatGPT for creating and refining a document, aiming to ensure all requirements are met with precision and professionalism. We could repurpose this GPT to create:
Contract Drafting Assistant
This GPT could be adapted to assist legal professionals or business owners in drafting contracts by ensuring all necessary clauses and legal requirements are included. It would analyze the purpose of the contract and the parties involved to suggest relevant clauses, thus streamlining the drafting process.
Grant Application Writer
Non-profits or researchers could use a similar GPT to help fill out grant applications. The GPT would analyze the grant’s requirements and the applicant’s information to ensure the application is thorough, aligns with the grant’s objectives, and highlights the applicant’s strengths in the context of the grant’s criteria.
Marketing Plan Generator
This GPT could assist marketing professionals by generating comprehensive marketing plans based on the analysis of a product, target market, and competition. It would include strategies, tactics, and metrics for success, tailored to the company’s goals and the market’s needs.
