Negotiation Preparation with Deepak AI

This week, I demonstrate how to use Deepak, our AI negotiation preparation assistant, to analyze scenarios efficiently and develop strategic insights.

If you enjoyed this episode, please leave a review and check out our website: www.negoai.ai

I welcome any suggestions, questions, or comments at yrana@negoai.ai

--------------------------------------------------------------------------------------------------
Copyright Disclaimer Under Section 107 of the Copyright Act 1976, allowance is made for "fair use" for purposes such as criticism, comment, news reporting, teaching, scholarship, and research. Fair use is a use permitted by copyright statute that might otherwise be infringing.

Transcript

Today, I'm going to show you how I use my negotiation preparation assistant that I call Deepak to analyze negotiation scenarios. As we did in our previous demos, I will use the same role play with the seller's instructions. But this time, I'm providing a very simple prompt: "Please perform an analysis of the attached seller's instructions," instead of having very complex prompts as I did before. Why is this? It's because the negotiation preparation assistant already has system instructions inside that I will show to you later.

Now let's see the response by the assistant. Executive summary, the objectives, estimated ZOPA, core recommended strategies, the strategic context, our positions, so the key facts, our interests, needs, and priorities, our BATNA reservation value and alternatives, leverage points. "The collection is unique, in excellent condition, and hard to assemble elsewhere." And then there are potential biases and assumptions. "Overvaluing the collection due to sunk costs." Very interesting. "Assuming the potential buyer is highly motivated, underestimating their alternatives."

Then we have three scenarios of the counterpart:

Highly motivated buyer, level of confidence 50%

Value-oriented and patient buyer, confidence 35%

Selective partial buyer, confidence level 15%

Then we have integrity value creation, and options. Leveraging key differences, for example, time pressure, hard-to-assemble collection, shipping and handling costs, condition completeness of set, modernization agenda.

So we have some options:

Full price with phased payment

Premium price for full set plus condition guarantee

A 30-day return option

Offer discount if the buyer takes immediate delivery and covers all shipping

Sell at $25,000, but include a year of consulting, access to Burns & Allen's Japanese law expertise for transition. Very interesting.

Bundle sale with digitalization rights

Sliding price based on shipping method. Lower price for bulk, higher if it's first class.

Add in additional minor assets or case files if permitted to sweeten the deal

Offer to facilitate introductions to Japanese law contacts

Structure a contingent price. For example, $24,000 base plus $2,000 if the potential buyer will use the books for a major client within a year

A joint press release to enhance both firms' reputation

Evaluation of the key options: Feasibility, value creation, risk. And we have refined package deals and contingencies.

So we have the preferred, where $27,000 and they cover shipping cost with a 30-day return option for condition concerns. Alternatives and a fallback.

And then, which is the first offer, the anchor, the concessions we can make, the framing points, and calibrated questions for, "How important is having the full set given your firm's plans? How would assembling these materials one-by-one impact your time and cost? It sounds like the condition of the books is a priority. For example, would a 30-day set inspection return window address these concerns? Is there any flexibility in your budget?"

And then there is relationship and process management considerations. How relationship impacts the negotiation. Trust-building actions, for example, offer on-site inspections, provide references.

And key assumptions, risks, overestimating their need, they may want only a subset, underestimating the buyer sensitivity, mitigation, and everything.

So this is a very comprehensive preparation report that we receive from this assistant. And it is quite in line with what we did with very complex system instructions, prompts with Claude and ChatGPT in our previous videos. I would say this is even more comprehensive. So let me show you how the assistant works.

So this is the name. This is the description.

It works with the latest ChatGPT-4.1 model, and I also provided some advanced settings. The context window is only half of the size it could be because I think the context window for ChatGPT-4.1 is one million tokens. And the temperature is more or less on the slightly creative side, so 0.6. But most important, these are the system instructions. So you have the role.

I provided the context.

For sure, here the context is very general. If this assistant was to be applied to, for example, the purchasing department of a company, then the context would be much more customized.

The instructions, what we expect the system to perform. And these are very comprehensive instructions. Then we have the extrapolation, the next one. Extrapolation is, please go beyond merely applying frameworks of negotiation, but combine them, generate novel options, formulate unique strategies, provide innovation. And then we have criteria. So the assistant has to abide to.

And some examples. I provided meta-level reasoning examples instead of very specific examples because this assistant is very broad, very transversal, and will encounter a lot of situations. So these are the examples.

And last, the output format. We also have the continuous improvement section that is based on feedback that you receive, please continually improve the level of your output. And there are also some internal processing steps that provide the assistant with some thinking space so that it can think before responding, and usually it enhances the response level and the response quality of the assistant.

Thank you for watching. If you're interested in AI and negotiation, please subscribe to our newsletter at negoai.ai.