Best Practices
Practical guidance for getting the most from your Neuro+ subscription.
Core principles
Start a new chat for each new topic
Each conversation has a memory, but only within that conversation. If you start a new subject in an old chat, the AI carries all the earlier context with it, which can muddy the responses. As a general rule, one chat per topic keeps things clean and focused.
When a conversation has been going for a long time and you notice the quality of responses starting to drift, that is usually a sign that the context is getting overloaded. Start fresh.
Give the AI the relevant background upfront
The AI only knows what you tell it. A vague prompt gets a vague answer. Before you ask your question, give it the context it needs: who you are, your role, the situation you are dealing with, and what you are hoping to get out of the conversation.
For example, instead of asking "how should I handle a difficult team member?", try something like: "I am a team leader managing a senior team member who has been disengaged since a restructure three months ago. I want to have a direct conversation with them next week. Can you help me think through how to approach it?"
The more relevant context you provide, the more useful the response will be.
Be specific about what you want back
It helps to tell the AI what kind of output you are looking for. Do you want a draft you can edit? A set of options to consider? Questions that help you think it through yourself? A summary of the key points? Saying this explicitly saves a lot of back and forth.
If the first answer misses the mark, keep going
You do not need to start over if a response is not quite right. Stay in the same conversation and redirect: "that is not quite what I meant", "can you focus more on the communication side", or "give me a shorter version". The AI responds well to course corrections within a conversation.
Break complex tasks into steps
If you have a large or multi-part task, do not try to put it all in one prompt. Work through it in stages. Ask for the analysis first, then the recommendations, then the draft. This gives you more control over the output and makes it easier to steer.
Choosing the right model
Neuro+ gives you two categories of model. The choice is usually straightforward.
Use Neuro+ Power or Neuro+ Coach when your challenge involves people, leadership, team dynamics, or personal development. These models apply the NeuroPower™ Framework and are specifically built for this kind of work.
Use a General Purpose Model when you need capable AI for tasks where the behavioural science lens is not the point: writing, research, summarising documents, or general questions.
You can switch between models at any point in a conversation without losing context, which makes it easy to shift from analysis to drafting or from reflection to planning within a single session.
Managing context
Every conversation has a limit to how much it can hold. Think of it like working memory: the longer a conversation runs, the more the AI has to keep track of, and eventually earlier parts of the conversation start to fall out of focus.
In practice this means:
- Long conversations can produce less accurate or less relevant responses toward the end
- If you paste in very large amounts of text, important parts may be overlooked
- Starting a new chat resets the context completely, which is often the right move for a new topic
If you are working with a long document, consider breaking it into sections rather than pasting the whole thing at once. Ask about one section at a time.
More detail on context length →
Advanced tips
Save prompts you use regularly
If you find yourself typing the same kind of request repeatedly, save it as a prompt. The Prompts section in the sidebar lets you store custom prompts so you can access them instantly rather than rewriting them each time.
To use a saved prompt, type / in the chat area. This opens a shortcut menu where you can select from your saved prompts. It is a small thing that saves real time if you have regular workflows.
Use Neurons to save your context
A Neuron is a saved context: a name, a set of instructions, and optionally some files that you want the AI to know about every time you start a particular type of conversation. If you do the same kind of work regularly, such as preparing for team meetings, working on a specific project, or advising a particular client, setting up a Neuron means you do not have to re-explain the background every time.
AI is powerful, but it is not infallible
AI models can be confidently wrong. They can misunderstand your situation, draw on outdated information, or produce advice that sounds reasonable but does not fit your specific context. Always apply your own judgment before acting on anything the AI tells you.
This is especially important in leadership and coaching contexts, where advice that seems sensible in the abstract might not account for the real people, culture, and history involved. Use Neuro+ as a thinking partner, not as the final word.
A note on privacy
Be thoughtful about the personal information you share in your conversations. This includes information about other people: colleagues, clients, direct reports, or anyone else involved in a situation you are working through.
You will often get better results by describing a situation in general terms rather than sharing specific names and identifying details. The AI does not need to know who someone is to help you think through how to work with them. Where possible, describe the dynamic or the challenge rather than the person.
Check your organisation's policies on data handling and AI use if you are unsure what is appropriate to share.