# What is prompt engineering? Definition and business implications

> Prompt engineering is the discipline of writing prompts methodically. It is not a rare profession but a transverse competence, which consists of formalising, testing, and standardising the instructions sent to an LLM to obtain reliable and reproducible outputs at the scale of an organisation.

## Definition
Prompt engineering covers the set of techniques that turn a vague request into a structured instruction producing a reliable output. The main techniques include: assigning a role (“You are an expert in...”), providing examples (one-shot, few-shot), decomposing the task into steps (chain of thought), specifying the output format (JSON, markdown, numbered plan), and introducing negative constraints (“do not mention...”). In 2023, the term designated an emerging and well-paid profession (up to 335,000 dollars per year in an Anthropic job listing published in 2023). Since 2024-2025, this segmentation has largely collapsed: models have become more tolerant of imprecise prompts, and best practices have spread. Prompt engineering is no longer a profession; it is a skill integrated into business roles (sales, legal, marketing), in the same way as Excel mastery or Google search.

## Example
The NBER study by Brynjolfsson, Li, and Raymond (2023, *Generative AI at Work*) observed 5,179 customer-support agents equipped with an AI assistant. The average productivity gain was 14%, but varied strongly by experience level: 34% gain for novice and low-skilled agents, and a minimal effect for the most experienced. The central lesson for executives: AI-driven productivity is not a property of the model, but of usage practice. Without a collective prompting discipline, gains remain uneven and not capitalisable.

A larger-scale case, in a customer-support context: Klarna, February 2024. The Swedish fintech group announced that its AI assistant (based on OpenAI) had handled 2.3 million conversations in one month, two-thirds of its customer-service chats, in more than 35 languages: the equivalent of the work of 700 full-time agents. Errands were resolved in under 2 minutes versus 11 previously, with customer satisfaction on par with human agents. In May 2025, Klarna changed course. Its CEO, Sebastian Siemiatkowski, acknowledged that cost had weighed too heavily in organising the service and that quality had suffered. The company is hiring human agents again, while AI still handles about two-thirds of inquiries. The lesson: an AI assistant cannot be managed on volume alone. Response quality has to be measured, tested and corrected over time.

## For a leader
Hiring a “prompt engineer” as a dedicated job, or conversely letting every employee improvise prompts case by case. The first path has been obsolete since 2024: models have become tolerant of imprecise instructions, and the competence is no longer rare. The second fragments usage, prevents capitalisation, and produces heterogeneous results depending on individuals, with no possible improvement track. | Building an internal library of prompt templates by use case, shared, versioned, tested. Identifying the 5 to 10 recurring tasks per function (sales writing, meeting summary, CV analysis), formalising a template for each, measuring average output quality, iterating. This library becomes a documentary asset of the company, transferable, and each improvement benefits all users simultaneously. The skill diffuses instead of remaining captive to a few individuals.

*Prompt engineering, fine-tuning and RAG: three ways to adapt an LLM to a use case*

| Technique | What it changes | Typical cost | When to choose it |
| --- | --- | --- | --- |
| Prompt engineering | Instructions and examples given to the model at each call: the model itself is unchanged | Low: a few hours to a few weeks per use case | The default: fast to iterate, reversible, no retraining. |
| RAG (Retrieval-Augmented Generation) | The model receives relevant excerpts from a document base with each question | Medium: vector search infrastructure + source curation | When domain knowledge lives outside the model (internal documents, product catalogue, an evolving knowledge base). |
| Fine-tuning | The model is retrained on a proprietary corpus: behaviour and style change durably | High: annotated dataset + compute cost + maintenance on every new base model | Highly specific editorial style, niche vocabulary the model doesn't cover, or latency drop on a single well-scoped task. |

## FAQ

**Do you need an expert to do prompt engineering?**
No. In business, prompt engineering is primarily a writing discipline: clear context, explicit output format, short examples. Documented gains (14 % average productivity uplift, NBER 2023 study of 5,179 support agents) come from good practices that can be shared, not from technical expertise.

**How do you structure a good prompt?**
Four building blocks: role (who the model is speaking to), context (data, constraints, expected tone), task (action verb and success criterion), output format (explicit structure). Adding one or two examples (few-shot) improves format fidelity. Iterate on 3-5 versions before fixing a production prompt.

**Prompt engineering or fine-tuning: what's the difference?**
Prompt engineering adjusts what you ask the model at inference: fast, no model training cost, reversible. Fine-tuning retrains the model on proprietary data: expensive, slower, but useful for a very specific style or expertise.

**What does a bad prompt cost?**
On a high-volume generation use case (support, extraction, summarisation), a poorly calibrated prompt inflates input tokens and produces off-format responses that need rework. Measure the cost per request and the share of outputs to redo before and after each prompt change.

**Is prompt engineering becoming obsolete?**
No, it's evolving. Recent models (for example Claude Opus 5.5, GPT-6) tolerate vague prompts better, but the performance gap between a mediocre and a well-crafted prompt remains significant in production. Techniques evolve (chain-of-thought, tool use, structured output); the discipline stays.

**Should companies document their prompts?**
Yes, in a versioned internal library. Each production prompt should have: an identifier, a date, a use case, the target model version, the measured quality metric, and an example of expected output. Without this discipline, prompts duplicate, drift and become un-auditable: treat them as code.

## See also

- [Prompt](https://wyp.agency/en/glossary/prompt/)
- [Prompting patterns](https://wyp.agency/en/glossary/prompting-patterns/)
- [LLM (Large Language Model)](https://wyp.agency/en/glossary/llm/)
- [Hallucination](https://wyp.agency/en/glossary/hallucination/)
- [Usage discipline](https://wyp.agency/en/glossary/usage-discipline/)
- [Map and territory](https://wyp.agency/en/glossary/map-and-territory/)
- [Context engineering](https://wyp.agency/en/glossary/context-engineering/)

## Sources

- Generative AI at Work, Brynjolfsson, Li & Raymond, NBER Working Paper 31161, 2023 (https://www.nber.org/papers/w31161)
- Prompt engineering overview, Claude documentation (Anthropic) (https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/overview)
- Klarna AI assistant handles two-thirds of customer service chats in its first month, Klarna press release, 27 February 2024 (https://www.prnewswire.com/news-releases/klarna-ai-assistant-handles-two-thirds-of-customer-service-chats-in-its-first-month-302072740.html)
- Klarna reinvests in human talent for customer service, Customer Experience Dive, 9 May 2025 (https://www.customerexperiencedive.com/news/klarna-reinvests-human-talent-customer-service-AI-chatbot/747586/)
- Fortune, Anthropic "prompt engineer & librarian" job listing, 9 March 2023 (https://fortune.com/2023/03/09/new-ai-jobs-chatgpt-like-assistants/)

---
Entry from the WYP.agency AI glossary · https://wyp.agency/en/glossary/prompt-engineering/ · Last updated : 2026-10-14