In the modern business landscape, there is a dangerous misconception that every negotiation exists in a vacuum. We often treat each deal as a standalone event, closing the door and fighting for the best possible terms before moving on to the next one.

But seasoned dealmakers know a fundamental truth: there are no episodic negotiations.

We live in a highly connected ecosystem where one’s behaviour in one boardroom will inevitably influence his reception in the next. When at the table- the idea shouldn’t be just negotiating the contract. It should be about constructing a long-term negotiating brand.

So, what should that brand be?

Many professionals oscillate between two extremes: the unyielding “jerk” vs the litigation-averse “pushover.” But the most successful negotiators strike a deliberate balance: they are incredibly cooperative by default, but formidably stiff when crossed.

Here is how to build a world-class negotiating brand that protects your boundaries without destroying your relationships.

The Danger of the “Conciliator” Brand

When Joel Peterson, a veteran negotiator with billions of dollars in transactions under his belt, departed a major real estate organisation, he possessed a distinct brand. He was known as a “conciliator”—a reasonable, win-win dealmaker who hated litigation.

He thought this reputation would ensure a smooth exit. Instead, former partners immediately slapped him with a lawsuit. Why? Because his brand was too accommodating. They assumed he was a “giver” who would simply cave to their demands rather than fight back. That assumption led to four gruelling years of litigation.

The lesson here is critical: if your brand is entirely anchored on being a “nice guy” who avoids conflict at all costs, bad-faith actors will use it against you. You must leave room in your brand for retaliation.

Peterson adapted his reputation to become someone who is highly reasonable and eager to build win-win deals, but who becomes “very stiff” the moment an opposing party acts unfairly or violates core principles. You want people to know they can do great business with you, but they do not want to end up in a fight with you.

Enforcing Boundaries Without Burning Bridges

How do you transition from being accommodating to enforcing strict boundaries without ruining the deal? It comes down to a few core tactics:

1. Lower Your Emotional Baseline: When most people try to set a firm boundary, they raise their voice, bluff, or use “high velocity words”. This is a mistake. When you scream or throw a tantrum, your counterpart assumes you are just venting and will calm down by tomorrow.

True power lies in a low emotional baseline.

During a tense negotiation over a billion-dollar property swap, Peterson sat across from six partners. When the senior partner made an outrageously unfair demand, Peterson didn’t yell. He calmly stood up, expressed sincere sadness that their long partnership had to end this way, and left to catch his plane.

Before he even reached the airport, the opposing partners were calling his phone, eager to walk back their demand and make a deal. Because his delivery was perfectly calm, they knew his boundary was real.

2. Establish “Rules of the Road”: If you are doing serial negotiations with the same partner, you can maintain your firm economic boundaries by establishing mutual expectations.

For example, Peterson had a long-term partner who simply couldn’t feel satisfied with a deal unless he was able to negotiate the price down.

Instead of fighting this dynamic, Peterson adapted.

For their next 250 deals, he simply offered prices 5% to 8% higher than his actual target. The partner would knock the price down, securing a psychological “win,” and Peterson would land at his exact required number.

Being reasonable simply means designing a process where the other side feels good about giving you what you want.

3. Leverage the Undeniable Power of Likability: There is a pervasive myth that being an effective negotiator requires being cold or ruthless. In reality, there is “almost no upside in being a jerk”.

Even the most hardened professionals are swayed by human connection. Following his four-year legal battle, an opposing litigator admitted to Peterson: “I wanted to hate you... but I couldn’t help but kind of liking you”. That litigator confessed that despite his professional training, his personal affinity for Peterson subconsciously influenced his approach.

By separating the people from the thorny problems, you can maintain fierce boundaries while remaining a genuinely likable partner.

The Ultimate Asset: Trust. At the end of the day, a world-class negotiating brand is built on a foundation of trust. If you continuously demonstrate Character, Competence, and Power (the empowerment to actually make a decision), people will actively seek out opportunities to do business with you.

Negotiations aren’t zero-sum wars.

They are simply conversations designed to solve mutual problems.

Build a brand as a flexible problem-solver who refuses to be taken advantage of, and you will never struggle to find a willing partner across the table.

PS- The inspiration behind writing this post is a masterclass by Joel Peterson about negotiation at Stanford.

For decades, the standard prescription for institutional inefficiency has been a single, unyielding word: digitization. We took paper files, turned them into PDFs, and moved physical queues into digital waiting rooms. But changing the medium of bureaucracy didn’t change its fundamental nature. The pace remained slow, the tasks remained highly repetitive, and the friction between the citizen and the state persisted.

True administrative transformation doesn’t happen when we simply record state processes electronically. It happens when we make those processes intelligent.

Drawing from a recent training on AI in Governance , this essay explores the transition from a passive digital state to an active, predictive framework—examining how we can leverage these technologies across public systems while strictly maintaining our ethical and mathematical guardrails.

1. Mapping the Friction: From Workflow to Automation

The first rule of applying any advanced technology to public service is deceptively simple: you cannot automate a process you do not fully understand.

Before drafting a single Request for Proposals (RFP) or outlining complex software requirements, public administrators must conduct a rigorous audit of existing workflows. We must isolate the precise entry points where government machinery slows down, categorizing tasks by their structural inputs and outputs—whether they are text, images, or audio data.

Once these friction points are mapped, Generative AI ceases to be a buzzword and becomes a highly targeted tool:

• The Administrative Scribe: In high-volume frontline operations—such as healthcare—speech recognition and generation models can dynamically capture and transcribe human interactions (like a doctor-patient consultation), updating central databases in real time without manual data entry.

• Cognitive Leverage for Officials: By functioning as an advanced assistant to public officials, AI can rapidly parse vast libraries of historical case files, legal precedents, and policy notes to deliver faster, highly context-aware responses.

2. The Operational Blueprint: The Smart Outpatient Model

To see this in practice, consider the structural re-engineering of a public Outpatient Department (O.P.D.):

[Citizen Arrival] ──► [Aadhaar + Facial Recognition] ──► [QR Token Issued]
                                                               │
[Database Updated] ◄── [Automated AI Audio Transcription] ◄────┘

By blending identity verification with real-time tracking, the frontend patient experience is completely reordered:

1. Seamless Identity: Integrating Aadhaar with Facial Recognition Systems (FRS) immediately authenticates the citizen upon entry.

2. Dynamic Flow Tracking: Citizens are issued tokens or wristbands embedded with a QR code. This allows computer vision and predictive models to monitor crowd flows, updating database actions dynamically and shifting administrative resources where they are needed most.

3. Ubiquitous Accessibility: Post-consultation instructions and welfare schemes are directly linked to that same QR code, ensuring clear, immediate access for the citizen.

3. The Two Pillars: Predictive vs. Generative Frameworks

For effective public sector procurement, we must avoid treating “AI” as a homogenous entity. It operates across two distinct technical dimensions, each solving a different economic problem within governance:

Predictive AI (Classical Machine Learning)

• The Core: Rooted in pattern recognition and statistical probability.

• The Utility: Used for risk forecasting, fraud detection, and optimized resource allocation. It answers the question: What is likely to happen next, and where should we deploy our budget?

Generative AI (Deep Learning & LLMs)

• The Core: Focused on contextual synthesis and multi-format output generation.

• The Utility: Powering advanced conversational interfaces (via frameworks like Dialogflow or Microsoft Bot Framework) to bridge the vernacular divide, making complex public policies understandable to citizens in their native dialects.

4. The Indian Paradigm: DPI Meets Intelligence

India’s Digital Public Infrastructure (DPI)—built on foundations like Aadhaar and UPI—provides the perfect substrate for these intelligent layers to scale. Across the country, we are seeing the emergence of powerful, localized use cases:

• Vernacular Inclusion via BHASHINI: Moving beyond English-centric governance, the Digital India BHASHINI initiative utilizes generative models to translate judicial transcripts and public welfare data into regional Indian languages instantly.

• Mass Scale Logistics: Systems deployed during massive cultural gatherings like the Uttar Pradesh Mahakumbh use computer vision and predictive AI to monitor vast railway passenger flows, mitigating crowd congestion and proactively ensuring public safety.

• Financial Integrity: Tools like MuleHunter.ai, operating within the banking ecosystem, leverage classical machine learning to scan network anomalies, detecting and freezing fraudulent accounts to protect public funds and subsidy transfers.

• Targeted Social Welfare: In places like Nagpur, the introduction of AI-powered Anganwadis uses anonymized data analysis to deliver precision nutrition and early-childhood learning kits, directly targeting systemic rural poverty.

5. The Mathematical and Ethical Guardrails

When a private software application fails, a user experiences temporary inconvenience. When a public sector AI system fails, a citizen may be stripped of their livelihood, healthcare, or fundamental rights. Therefore, our deployment metrics must be uncompromising.

We must evaluate every model through a strict matrix of mathematical correctness, optimizing heavily for True Positives and True Negatives:

                  ACTUAL REALITY (Recall) [cite: 44]
               ┌───────────────────┬───────────────────┐
               │   Eligible (1)    │  Ineligible (0)   │
 ┌─────────────┼───────────────────┼───────────────────┤
 │  Approved   │   True Positive   │  False Positive   │
P│    (1)      │ (System Success)  │  (Unjust Leakage) │
E│             │         │         │
D├─────────────┼───────────────────┼───────────────────┤
I│   Denied    │  False Negative   │   True Negative   │
 │    (0)      │ (Citizen Denied)  │ (System Success)  │
 └─────────────┴───────────────────┴───────────────────┘

In public policy, a False Negative—wrongfully predicting a citizen is ineligible for a service—is a catastrophic policy failure. Because we must balance Precision (predictive accuracy) with Recall (the reality of who needs the service), we must judge models by their F1-Score. This is the harmonic mean of the two metrics:

F1 = 2.(Precision.Recall/Precision + Recall)

Before any wide-scale deployment, rigorous A/B testing must be mandated to prove that the AI-assisted model delivers services faster and more equitably than the traditional human workflow it seeks to replace.

The Three Questions for Every Administrator

As we venture toward the long-term horizon of Artificial General Intelligence (AGI), our immediate professional focus must remain on safely managed, Applied AI. Every public sector blueprint must continuously answer three non-negotiable questions:

1. Equity: Did the AI model unfairly deny service to a citizen due to biased historical data?

2. Competency: Is the AI model performing worse than a trained human professional for the task at hand?

3. Inclusion: Are any citizens being entirely left out of the state’s digitization efforts?

Technology is an exceptional lever for amplifying efficiency, but it cannot replace institutional empathy. By building resilient Platform Engineering, mastering Prompt Engineering, and keeping a dedicated human-in-the-loop, we can build an agile, predictive state that serves every citizen with absolute precision.

What are your thoughts on introducing AI-driven automation in public administrative spaces? Let’s discuss in the comments below.

If you found this essay insightful, consider subscribing to the Minimalist Manifesto for regular deep-dives into philosophy, economics, and public policy.