Skip to main content
Guide 12 min read

Prompt Engineering
for Business

Practical techniques to get better results from ChatGPT, Claude, and other AI tools. Written for business professionals, not engineers.

The difference between getting a useless response and a brilliant one from AI usually comes down to how you ask. Prompt engineering isn't a technical skill: it's a communication skill. Here are the techniques that make the biggest difference in everyday business use.

Think of prompting like briefing a new team member. The better your brief, the better the result. You wouldn't say "write me something" to a colleague. Don't say it to AI either.

6 Techniques That Actually Work

These aren't theoretical. Each one is demonstrated with real before/after examples from common business tasks.

1

Be Specific About Context

Weak prompt

Write a sales email.

Strong prompt

Write a follow-up sales email to a UK-based SME owner who attended our AI workshop last Tuesday. They expressed interest in automation but are concerned about cost. Tone: professional but warm. 150-200 words.

Why it works: The more context you give, the more relevant and useful the output.

2

Define the Output Format

Weak prompt

Summarise this report.

Strong prompt

Summarise this quarterly sales report in 5 bullet points, each starting with a key metric. Include a one-sentence takeaway at the end. Use plain English.

Why it works: Specifying format saves you from reformatting the output afterwards.

3

Give It a Role

Weak prompt

What should I do about declining sales?

Strong prompt

You are a business consultant specialising in UK retail. Our Q3 sales dropped 12% year-on-year. Foot traffic is steady but conversion rates fell. Our main competitors launched aggressive discount campaigns. What are three data-backed strategies to recover without racing to the bottom on price?

Why it works: Assigning a role focuses the AI's knowledge and tone appropriately.

4

Show, Don't Just Tell

Weak prompt

Write a proposal for a client.

Strong prompt

Here's a proposal we used last year that won the project: [Paste example] Write a similar proposal for this new client using the same structure and tone, but adapted for their specific needs: [describe client needs].

Why it works: Examples are the most powerful way to communicate what you want.

5

Break Complex Tasks into Steps

Weak prompt

Analyse our entire customer database and create a marketing strategy.

Strong prompt

Step 1: Given this summary of our customer data [paste summary], identify the top 3 customer segments by revenue. Step 2: For each segment, suggest one campaign idea tailored to their characteristics. Step 3: Write the subject line and opening paragraph for each campaign email.

Why it works: Step-by-step prompts produce more accurate and structured results.

6

Ask It to Ask You Questions

Weak prompt

Create a training plan for my team.

Strong prompt

I need to create an AI training plan for my 15-person team. Before you start, ask me 5 questions about my team, their current skill levels, and our business goals so you can give me the best possible recommendation.

Why it works: Letting the AI gather information first leads to far better outputs.

Ready-to-Use Business Prompts

Copy these prompts and adapt them to your needs. Each one is designed for a common UK business task.

Email drafting & replies

You are my executive assistant. Draft a reply to this email that is polite, concise, and moves the conversation forward. Keep it under 100 words.

Meeting summaries

Summarise these meeting notes into: 1) Key decisions made, 2) Action items with owners, 3) Open questions. Format as a short email I can send to attendees.

Report analysis

Analyse this data and highlight: the 3 most important trends, any anomalies or concerns, and one recommendation for each trend. Use simple language a non-technical director would understand.

Proposal writing

Using this structure [paste structure] and this client brief [paste brief], write a project proposal. Match the tone and style of this example: [paste example].

Policy drafting

Draft a [policy type] policy for a UK SME with [number] employees in the [industry] sector. Include sections on: scope, responsibilities, procedures, and review date. Comply with UK employment law.

What Not to Do

Share confidential data, passwords, or personal information
Trust facts, statistics, or citations without verifying them independently
Use AI output for legal, medical, or financial advice without professional review
Copy-paste AI output without reading and editing it first
Assume the AI understands your industry's regulations and compliance requirements

GDPR Reminder for UK Businesses

When using AI tools in a business context, remember your GDPR obligations:

Anonymise data

Remove names, emails, and identifying details before pasting into AI tools.

Check terms

Review the AI tool's data policy. Some train models on your inputs.

Get consent

If processing customer data with AI, ensure your privacy policy covers it.

Keep records

Document what AI tools you use and what data you share with them.

Want Your Team Trained on AI?

Our AI training programmes include hands-on prompt engineering workshops tailored to your industry. From half-day foundations to executive briefings.