Your Personal AI trains on the memory blocks inside your memory stack. When you’re asking questions to your AI, the quality and accuracy of the responses depends directly on the data you upload onto the platform.
If you’re starting from scratch, you can set yourself up for success by gathering as much data as you can, sorting and and getting it ready for upload. Here are tips and considerations to help you save time.
Planning for AI Personas
When gathering files, links and media to train your AI, you can already start sorting it by Personas. While all the information will be saved to your main AI and your primary memory stack, the training process is handled at the Persona level.
For example, if you have marketing data and operations data, you can separate those into two folders and then upload data to your Marketing AI and your Operations AI.
Forming the first memories
When starting from scratch, it’s important to round up all the foundational data you’ll use to train your model. Depending on the type and content of the Personas you’re training, here are the kinds of data you should consider including. This is a non-exhaustive list: the more you upload, the more data you have at your disposal to ask questions and get high-value answers.
Business plans and strategy documents
Uploading business plans and other strategy documents such as SWOT analyses gives more context about your business and its positioning on the market. With these memories present, your AI can guide you through decision-making processes and help you explore new product and service possibilities.
Operational documents
Standard operating procedure (SOP) documents, operational efficiency reports and project management breakdowns can be uploaded into your AI, unlocking multiple advantages for your entire team. You can create an AI to assist new employees get up to speed as well as provide managers with a big-picture perspective of productivity and improvement opportunities.
Communication and collaboration data
As you upload emails, memos, meeting minutes, feedback and support logs you can ask questions of the major topics discussed, giving you a more contextual view of how work is moving forward. You can also use this data to explore events and timelines, building a thread of actions for diagnosing issues and improving workflows.
Customer and market data
Stacking customer and market data, such as profiles, CRM records, market research reports, surveys and polls will be useful to improve or build new products and services. At the same time, it will let you ask questions about your customers and audience, useful for exploring marketing personas or getting a feel of how people would react to changes in your business.
Connecting cloud platforms for automatic sync
While there’s a manual element when training a new Persona from scratch, you can reduce the initial setup time and ongoing maintenance by connecting your cloud accounts to Personal AI. At the moment you can connect:
Google Drive
Microsoft OneDrive
Gmail
Outlook
Once connected, you can choose individual files, folders and email labels that you’d like to sync. The changes will be automatically synced with Personal AI, keeping your AI up to date.
Growing your AI
We recommend cloud sync for keeping your AI up to date with the latest data about you and your business. However, if you have data coming in from other sources, it’s best to gather it in a single place such as a folder on your computer or cloud drive, and schedule a weekly or monthly timeslot to keep training your AI. The compounding effect of more memory blocks will greatly improve response quality, unlocking new insights.
.svg)
