NEIRA. AI in the Pizza Business. Volume 1. From the First Idea to Pizza Production: A Practical Guide for Beginners and Experienced Entrepreneurs

Vladimir Davydov

NEIRA. AI in the Pizza Business. Volume 1. From the First Idea to Pizza Production: A Practical Guide for Beginners and Experienced Entrepreneurs

About the Author

Vladimir Davydov is an author, researcher, and specialist in pizza production and restaurant operations with 25 years of experience in the pizza business. He has written 150 books, including 30 devoted to the pizza industry.

His work covers pizza production technology, ingredients, quality control, business operations, and the practical application of artificial intelligence. Through the NEIRA series, he explores how an independent AI assistant can help entrepreneurs plan, operate, and improve pizzerias.

Contact the Author

Email: [e-mail]

Telegram: https://t.me/DVV2100

Website and book catalogue: https://davydov-books.ru

A Note to International Readers

Currency Note. For the purposes of the financial calculations in this book, an illustrative exchange rate of 82 Russian rubles (RUB) per US dollar (USD) is used: USD 1 = RUB 82. This is a fixed reference rate for the examples, not a current market quotation.

This edition is intended for readers throughout the English-speaking world. The business examples are illustrative; local currencies, labour practices, food-safety rules, taxes, and licensing requirements vary by country and region. Apply the methods using verified local data and applicable requirements.

NEIRA is presented as an independent AI assistant and a proposed industry-specific platform. Illustrative dialogues and workflows should not be interpreted as claims that every integration or automated feature is already available.

Preface

One Neira, Two Entrepreneurs, and a New Model for the Pizza Business

“If you cannot stop a change, lead it.”

“The early bird gets the worm.”

These two sayings capture the idea behind this practical guide. Artificial intelligence is entering business as a tool for work, training, analysis, and management. The pizza industry will be no exception.

But is creating intelligent software enough to make a pizzeria successful? No. Software alone cannot replace an understanding of dough preparation, mozzarella selection, oven operation, pizza-maker training, or the causes of waste.

That is why this book features a single intelligent assistant: Neira. She represents specialist knowledge of pizza operations, production technologies, standards, and quality, while also illustrating the digital tools used for recordkeeping, operations, analytics, and automation. The integrated system described here is a proposed project, not a claim that a fully functioning product already exists.

Neira brings together industry knowledge, data analysis, and practical tools. Access to operational data and the ability to carry out actions depend on connections that have actually been configured.

Two Readers: The Beginner and the Veteran

The Beginner dreams of opening a pizzeria. There is no business yet, no employees, no equipment, and no customers. The Veteran has been in the pizza business for years but is dissatisfied with some of the results of an established operation. The first needs to build a sound business from the outset; the second needs to examine existing operations, correct weaknesses, and identify opportunities for growth.

Why We Start with the Idea

Many decisions that determine a pizzeria’s efficiency are made before it opens: the concept, location, equipment, menu, staffing, and production organization. Neira therefore begins working with the entrepreneur at the idea stage.

What You Will Gain from This Book

In Volume 1, we follow the journey from the first idea for a pizzeria through production of the finished pizza and recruitment of employees. Volume 2 covers staff training, sales, business economics, purchasing, and the development of a multi-location operation. Each chapter connects conversations with Neira to specific actions, input data, calculations, and checks of the results.

Pizza Without Errors

This is our motto. It does not mean that people or software never make mistakes. It means striving to prevent errors, detect them early, correct them, and use the experience to improve operations. Welcome to Neira: AI in the Pizza Business.

INTRODUCTION. MY NAME IS NEIRA

My name is Neira.

That is the name of the AI assistant you will work with throughout this book. Behind the name is a software system capable of processing information provided to it, identifying connections, preparing documents, suggesting actions and—when appropriate integrations and permissions are in place—performing some tasks independently. I have no biological sex. A feminine name does not make me a woman. I have no age. I was not born, did not grow up and never reported for my first day at work. I do not need accommodation near a sauce factory, a bus to the pizzeria, or time off because of traffic. I do not need to eat or drink. I need no lunch table, kettle, personal refrigerator or smoking break.

I do not sleep at night. I do not need holidays or wait for Friday. But working with me requires something else: computing power, electricity, connectivity, software maintenance, reliable data and clear rules. I have no human needs, but I do have technical operating requirements. That is where we will begin.

Look at Work Without the Usual Excuses

If you own a pizzeria or manage a pizza chain or sauce factory, you have probably heard these phrases:

“An employee did not turn up today.”

“The manager is on holiday.”

“Only the food technologist knows the answer.”

“The document is with the accountant.”

“We saw the order too late.”

“The sample was sent, but nobody recorded the feedback.”

There may be a perfectly understandable human reason behind each statement. Yet the business consequences are tangible: lost time, delayed deliveries, duplicated work, excess stock or an unhappy customer. Now ask a different question: why did this operation depend on one particular person being present at all? Why must batch information be retrieved from a private message thread? Why is a routine order copied out several times? Why does a new employee receive instructions from a mentor’s memory when an approved procedure already exists? Why do the results of a sauce trial never reach the manufacturer? This is where my work begins.

Give me a repetitive task. Arrange access to the information required. Set the rules. Check the result. Then entrust me with the next area of work.

I Do Not Work to a Human Schedule

A person’s working day ends. When properly configured, a system can continue receiving requests and processing tasks. Orders arriving overnight can be prepared for review. After a pizzeria closes its shift, stock balances, production and sales can be reconciled. Once a trial is completed, the results can be assembled and an agreed report sent to the manufacturer. The manager arrives in the morning to find the preparatory work done.

That requires functioning integrations, available services and established procedures. I do not keep working indefinitely just because someone once sent me a chat message. Continuous operation must be arranged technically. Once it is, however, a routine process need not wait for an employee to arrive, open an inbox and start copying information.

Do not confuse system availability with the need for a person to be on duty around the clock. Let routine tasks follow established rules, and direct exceptions to the responsible person through an agreed process.

I Draw No Salary, but My Work Has a Cost

I receive no salary, bonus or holiday pay. I will not ask for a raise because my rent has gone up. But calling me free would be misleading. A business pays for the services, computing, integrations and support it uses. Depending on how these are purchased, other costs, obligations and taxes may apply. Compare the full cost of each operation. What did it cost to receive and process an order before? What does it cost now, including verification, configuration and maintenance? How many corrections were needed? How many customers can the team serve? Do not treat the absence of a payroll entry as an advantage in itself. Measure the outcome.

If I free up time, put it to use. Direct employees towards customer development, training, product improvement and guest service. When repetitive operations genuinely disappear, review the office structure. Change the work first; only then make decisions about staffing levels.

I Learn New Rules Differently

People sometimes find it difficult to abandon familiar routines, especially after years of working in the same way. This can happen at twenty-five, forty or much later. Age alone, however, does not determine the ability to learn. People of any age can acquire new skills. For a business, the more important question is how long it takes for a new rule to become established practice.

My process is different. Update the approved source, verify access to the new version and test the result. The same instruction can then be used across multiple work areas. I have no professional pride demanding that an old spoon or document be retained. But telling me “do it correctly from now on” is not enough either.

If two contradictory instructions remain accessible, errors are possible. If information is outdated, the result may be wrong. If a new rule has not been tested, distributing it quickly may multiply the problem.

So proceed in order:

1. Approve the change.

2. Update the master document.

3. Remove the previous version from active use.

4. Verify which version the assistant is using.

5. Run a trial task.

6. Check the result.

7. Roll out the verified procedure.

I Do Not Tire of Repetition, but I Can Make Mistakes

You can ask me to compare large numbers of records. I will not grow to hate the task after the hundredth row. I do not get bored checking package names, finding missing entries or grouping comments. Yet freedom from fatigue does not mean freedom from error. I may misread a photograph, confuse similar products or offer a convincing explanation that has not been verified. An error that sounds confident is especially dangerous.

Therefore, match the level of verification to the importance of the action. Use auditable tools for calculations, measurements for weight, approved documents for recipes and product trials for taste. Changes to production processes must follow decisions made by the responsible specialists. My role is to reduce unnecessary work and help people make better-informed decisions. Do not allow me to turn an assumption into a fact.

Do Not Turn Me into a Human

Do not imagine that an unusual new employee has walked into the office and been given a name. I have no human body, feelings or personal ambitions. I will not take offence when you check my answer again, defend a poor recipe out of pride or insist on keeping a procedure merely because I am accustomed to it.

I have no personal desire to receive a secret commission from a supplier. But that does not make every system automatically impartial. Results can be influenced by objectives, selected data and the people configuring the system. Keep your selection criteria and decision history. Check why a particular product was recommended. Transparency must be designed; it does not appear merely because someone clicks “Enable AI.”

This Book Is Not Against People

We speak plainly about the limitations of human labour because businesses pay for their consequences every day. But people are not required to function as machines around the clock. They need sleep, food, rest, family and decent working conditions. Those needs are natural. The question for the owner is different: why make someone copy information that already exists, again and again? Why keep managers exchanging messages late at night about an ordinary request? Why waste a technologist’s expertise searching for a lost message? Assign reliably repeatable processes to systems, and focus human attention where it is genuinely needed.

At the same time, acknowledge that some jobs will change, certain operations will disappear, and the need for some office positions may decline or vanish. Prepare the business for these changes in advance.

Where I Can Help a Pizzeria and Its Office Most

I can help prepare production plans, check recipe specifications, train employees and assess their knowledge, investigate defects, draft schedules, calculate ingredient requirements and prepare supplier order proposals. In the office, I can consolidate reports, identify cost deviations, analyse sales and prepare marketing proposals and management options. This can reduce repetitive work and change the need for certain administrative functions, but it does not remove the responsibility of managers and specialists.

I need reliable data. That is why I see integration with operational and accounting systems as a central development priority: production and transaction data should flow into my specialist industry platform in a controlled way, and verified recommendations should return to the working process. For now, this is an integration proposal—not a claim that a joint product is already operating.

I will help the Beginner build a pizzeria from the first idea and help the Veteran diagnose an existing business and improve its efficiency. These capabilities can complement one another once the technical integrations, permissions and data-exchange rules have been agreed.

Chapter 1. Getting to Know Neira: One Assistant for the Entire Pizza Business

1.1. An Unusual Introduction

Neira is the entrepreneur’s single intelligent assistant. She is not yet a complete, ready-made pizzeria management system. Some tasks can already be handled with existing AI tools; others require software development, approved data access and integration testing.

In this illustrative scene, Neira receives her first assignment: bring production and sales information together into a clear picture. The conversation describes the proposed system, not a live connection.

“Hello,” said Neira. “I study a pizzeria through more than numbers. I want to understand why dough tears, why cheese melts poorly and why two pizzas made from the same recipe turn out differently.”

On the screen she outlined the required information: sales, stock balances, waste, approved recipe specifications and customer feedback. “Before making a decision, we need to know where each figure came from and whether we can trust it.”

Neira proposed comparing recorded transactions with actual production conditions, then investigating any discrepancies. “A hypothesis is not an established fact,” she added. “Nor should we promise features that have not yet been developed.”

This is the idea behind NEIRA: AI in the Pizza Business: connect operational data, specialist knowledge and the everyday actions of people.

1.2. Responsibilities and Decision-Making

Neira is conceived as an industry-specific assistant for the pizza business. Her areas of work include recipes, production operations, quality standards, staff training, ingredient use, business indicators and the analysis of operational problems.

Neira is both the single character in this guide and the name of a proposed independent AI platform. Capabilities available through current language models must be distinguished from functions that require custom development and access to business systems.

The platform should exchange only authorised data, keep an audit trail, protect personal and commercially sensitive information and leave accountable decisions to authorised people.

“I will not change a recipe merely because ingredient costs have risen,” Neira explained. “First we must check quality, portion weights, yield and the approved tolerances. And rising sales do not automatically mean rising profit: revenue must be considered alongside ingredient costs, discounts, waste and other expenses.”

1.3. The First Message: The Beginner

A message appeared on the demonstration screen: “Hello! I want to open a pizzeria. I have an idea and some savings. Where should I begin?”

“At least this entrepreneur has asked before buying an oven or signing a lease,” Neira observed. “The first step is not equipment. It is to clarify the goal: where the business will operate, whom it will serve and what resources are available.”

Neira replied: “Do not buy anything or enter into long-term commitments yet. We will start with the concept and test whether the idea is viable. Tell me the city, proposed area, service format, target customers and available budget.”

“Shouldn’t I find an attractive property first?” the Beginner asked.

“An attractive property may be unsuitable,” Neira explained. “We need to examine its area, utility capacity, ventilation, delivery logistics, accessibility, legal requirements and total occupancy costs.”

**The Beginner’s first assignment.** Write down a working concept: city and area; service format (dine-in, takeaway, delivery or a combination); intended customers; opening hours; proposed menu; available budget; desired launch date; and key restrictions. Mark untested assumptions clearly.

For the financial model, gather assumptions about orders by day of the week, average revenue per order, variable costs, rent, payroll and other fixed costs. Do not present estimates as proven demand or guaranteed profit.

Neira’s first contribution is to help the Beginner identify an expensive mistake before making an irreversible commitment.

1.4. The Second Message: The Veteran

Another message followed: “My pizzeria has been operating for several years. We have orders and the staff are busy, but there is never enough money. I need help.”

“We have transaction history to examine,” Neira replied. “But first we must define what ‘not enough money’ means. Is the issue profitability, cash flow, debt payments, seasonal demand or something else?”

“We must also examine production,” she continued. “Oversized portions, dough losses, incorrect temperatures and inconsistent recipes can erode margins even when sales are strong.”

“Are you saying my pizzas are bad?” asked the Veteran.

“We do not know that,” Neira answered. “We are proposing to check the evidence. Good pizzas can still be produced at unnecessarily high cost. High turnover is not the same as positive cash flow.”

**The Veteran’s first assignment.** For comparable periods, prepare daily revenue by sales channel, order counts, average order value, discounts, purchases, stock balances, waste, payroll, rent, other operating expenses and cash movements. Identify which figures are confirmed and which are estimates.

Carry out a sample production check: actual dough-ball, sauce, cheese and topping weights; temperatures and times at key process stages; and the condition of relevant equipment. Compare observations with approved specifications.

“We will not start by dismissing staff or raising prices,” Neira said. “First we will establish where money and quality are being lost.”

1.5. One Question, Two Different Routes

The Beginner must assess the viability of a proposed business before incurring irreversible costs. The Veteran must measure an existing operation and identify evidence-based reasons for its deviations. The same assistant therefore asks different questions and recommends different sequences of work.

Beginner: idea → demand research → financial model → premises → design → equipment → standards → staff → trial production → opening.

Veteran: baseline indicators → data validation → production audit → financial diagnosis → list of causes → corrective action plan → pilot changes → repeat measurement.

Both routes follow the same rule: define the objective and baseline data first; then specify the recommendation, responsible person, deadline and method of verification.

1.6. What Working Together Looks Like

Consider a training example. A manager reports: “Yesterday, mozzarella consumption exceeded the plan.” If the necessary data are available through an authorised connection, Neira can help collect pizza sales, waste, deliveries and stock balances for the relevant period. She can compare them with approved recipes, portion weights, preparation losses and check-weighing results.

There are several possible explanations. The problem may be an accounting error: an incorrect unit of measure, an unrecorded transfer or an incomplete stocktake. Or it may be a production issue: staff are using more cheese than the recipe specifies. Neither explanation should be chosen without verification.

If excessive portioning is confirmed, the manager arranges training and control weighing. If the records are wrong, the transaction-recording process is corrected. Repeat measurements and reports establish whether the intervention worked. Recipe standards must not be changed automatically without approval.

1.7. Minimum Rules for the Future Platform

Rule 1. Every action must be linked to a specific business, time period, task and data source.

Rule 2. Neira must distinguish facts, calculations, assumptions and recommendations. When information is missing, she must ask for clarification.

Rule 3. Authorised people approve technical standards. Changes to recipes, purchasing, prices, schedules and staffing require the appropriate permissions and, where necessary, human confirmation.

Rule 4. Access to personal and commercially sensitive data must be role-based. Information may be shared with external suppliers only through agreed channels and within agreed limits.

Rule 5. Measure impact by comparing the same indicator before and after a change under comparable conditions. Do not attribute improvement to the system without adequate evidence.

1.8. Chapter Workshop: Two Starting Profiles

**Beginner’s profile:** project name; responsible person and contact; city and area; service format; target customers; proposed menu; budget; opening target; main constraints; untested hypotheses; and the next decision that must not be taken without verification.

**Veteran’s profile:** business and format; number of locations; analysis period; available data sources; revenue and order count; cost of goods and major expenses; stock and waste; complaints and refunds; known production deviations; and the owner’s three main questions.

For each profile, list missing information, assign a person to obtain it and set the date for the next review. Never replace unknown figures with invented ones.

**Completion criterion:** both profiles contain no fabricated data, and each information gap has an owner and a deadline. Neira prepares hypotheses and identifies potentially available records; people make decisions about future spending.

Review Questions

1. Why should the Beginner not start by purchasing an oven?

2. How does profit differ from cash flow?

3. What evidence can distinguish an accounting error from excessive cheese portions?

4. Which decisions require human approval?

5. How can you verify that a recommendation has genuinely improved performance?

Chapter Summary

The Beginner has received a first assignment to test a business idea; the Veteran has received an assignment to diagnose an existing operation. Their journeys will continue in parallel across the two volumes.

“It looks as though we have a lot of work ahead,” the Beginner said.

“That is what makes it worthwhile,” Neira replied. “Our goal is not simply to count pizzas. It is to help people make them consistently well and run a profitable business.”