Data and AI Training

Home | Prices | Contact Us | Courses: Power BI - Excel - Python - SQL - Generative AI - Visualising Data - Analysing Data

GenAI Prompt Guidelines - Tips to get better responses from AI tools

We talk with AI tools by providing a question or instruction (called a prompt). The AI tool responds and then we can prompt again and the AI tool will respond again – and so on. Often the AI tool user interface may suggest a few example prompts.

Writing effective prompts is a key skill. A basic instruction alone may not result in a good response. We can help AI tools help us with more relevant, accurate responses if we provide more tailored prompts. Here are a few pointers to obtain better responses from AI tools. Some of these are described in more detail in the sections below

Make the prompt more specific

A more specific prompt will yield a more useful response. We don’t need to provide the entire request in a single prompt, we can refine over several prompts. Here are a couple of examples of initial and successive prompts that gradually refine and specialise the AI’s responses.

Example: Netflix recommendations

Provide a few recommendations about what I could watch tonight on Netflix

Only consider rom-com movies

Specifically for a UK audience

And with a Christmas theme

Compare this content with similar content available on the BBC iPlayer.

The AI tool may well suggest a few useful options for a follow-on prompt after each response.

Example: Describe and classify specialised real estate investment trusts

Describe how an investment trust differs from other forms of financial investment

What are the common types of investment trust?

Expect a response describing equity, fixed-income, infrastructure, real estate and other types of investment trust.

Explain Real Estate Investment Trusts (REITs) in more detail

What types of REIT are there?

Expect a response describing retail, residential, healthcare, and other REITs.

Compare the risks of investing in an industrial REIT with other types

Tell the AI tool what role you want it to play

Ask the AI tool to adopt a persona or act in a role at the start of a conversation. For example, the initial prompt for a finance-related conversation could be

Act as an accountant with 10 years experience in the finance sector who has an ACCA qualification

Or as a software development-related conversation

Act as a software development manager with 20 years experience in the technology sector who has an Scrum certification

Or as a project management-related conversation

Act as a project manager with 30 years experience in the health sector who has a PRINCE2 qualification

Or as a medical practitioner

Act as a doctor, a NHS ENT consultant with 20 years of experience specialising in audiology

Provide an example of the response required

We can provide an example of the response required, making it easier for the AI to follow suit. We can also provide a clearly marked bad example, to be clear about what response we don’t want.

Prompt:

Write a short review (100–150 words) of the novel “The Echoes Of Lena”. Focus on aspects such as the language, storytelling, characterisation, and plot development. Avoid vague, purely emotional responses.

✅ Good example:

“The novel’s lyrical prose draws the reader into a richly imagined world. The pacing is deliberate but rewarding, revealing layers of character depth—especially in the conflicted protagonist, Lena. The plot weaves past and present seamlessly, offering unexpected turns without feeling contrived. It’s a reflective, character-driven story that rewards patient reading.”

❌ Poor example:

“Absolutely amazing! I loved every second of it!! Everyone should read this book!”

Talk, don’t type

Many AI tools have a mode where we can hold a voice conversation with them in a very natural style: for example, we are able to interrupt the AI tool in mid-flow. This may become the dominant way of chatting with our AI tools in the future. Give it a try. The AI is fluent in most languages so have a conversation in French or Hindi or whatever languages suits you best.

Ask the AI to help you with your prompting

A novel technique is to turn the tables on the AI tool. Ask it to help you with your prompting

Can you give me some ideas of good prompts on the topic of …

Chain of thought reasoning

Some models now have chain-of-thought reasoning. Even if your model is not one of these, these prompts may encourage it to give a more considered or thoughtful response.

Think step-by-step
Please take your time and report back once you have considered thoroughly
Show your working

Reverse the roles (flipped interactions)

We can ask the AI tool to elicit the information from us in a series of targeted questions, for example:

I am going to Rome for a holiday. Can you ask me a series of question to find out my preferences. Stop after each question and wait for my answer. Take that answer into account in your subsequent questions. Once you have enough information from me, provide your recommendations.

OpenAI’s advice on prompt design

OpenAI provides some comprehensive advice and suggests six strategies for better results here

Google’s advice on prompt design

Google provides some prompting 101 advice and examples here. In short

  1. Give clear instructions: Help the model understand your request.
  2. Define the task: Be specific about the desired output.
  3. Specify constraints: Set limits on length, format, or style.
  4. Define the response format: Tell the model how to structure its output.

They provide some examples in their prompt gallery here.

The RICE Framework

One framework to help us with this is RICE: this stands for Role, Instructions, Context (and Constraints) and Examples.

Practice writing prompts with Google’s “Say What You See” quiz

Google has online quiz to practice prompting (of a sort). It shows an image, we have to describe it, it then generate an image based on our description and estimates how closely the original and new images match (which reflects our descriptive skills). It’s great fun!

Here is the link to Say what You See.

Finally

It is often useful to know when the model was trained since it will not know about more recent events unless it specifically searches the web. One option is to ask it directly.

What is your knowledge cut-off date?