For independent creators and small business owners, artificial intelligence is not about replacing the human touch; it is about automating the friction. The primary value lies in handling the administrative drudgery—scheduling, invoicing, and data organisation—so that makers can focus on their craft. The critical skill for any small operator is distinguishing where an AI model is merely “good enough” for routine tasks versus where human oversight remains non-negotiable.
Case study: The private tutor’s second memory
Sam Finnegan-Dehn operates a part-time tutoring business in London, offering mathematics and philosophy lessons to university students. While his core expertise lies in delivering high-level instruction, the logistical burden of running a solo enterprise often stalls growth. He is responsible for curriculum planning, sourcing reading materials, drafting assignments, and managing finances, all alongside his full-time career.
To manage this load, Finnegan-Dehn has integrated AI into his digital workflow to act as a secretarial assistant. He utilises the technology to maintain a running record of client progress and to synthesise new research, effectively creating a second memory that links disparate notes. While he has tested various large language models, including Claude and ChatGPT, he currently relies on Notion AI due to its seamless integration with his existing notebook ecosystem.
The tool does not generate the teaching content itself. Instead, it transcribes client meetings (with consent) and produces automated summaries. Finnegan-Dehn reviews these summaries to identify patterns, such as a specific technique that fails to resonate with a student, allowing him to adjust his pedagogical approach for subsequent sessions. The system also assists with goal setting, drafting lesson notes, invoicing, and synchronising social media updates.
Regarding strategic planning, Finnegan-Dehn uses the AI to bridge the gap between abstract ambitions and concrete actions. He defines a “North Star” objective, such as a target client count for the end of the year, and prompts the AI to generate a step-by-step roadmap based on his stored business profile. He then evaluates the proposed steps to prioritise his daily tasks.
The tool: Notion AI
Notion has long been a dominant force in note-taking software. Its AI add-on, launched in late 2023, has since expanded to interact with numerous productivity platforms, including an email client, calendar integrations, and a newly released autonomous agent. While this connectivity raises valid privacy concerns, it creates a powerful virtual assistant for those with limited bandwidth.
The tasks Notion AI excels at are largely rote: synchronising data between documents or searching through historical notes. This utility is particularly appealing to small business owners who lack the time for menial work. Other industries are seeing similar adaptations; for instance, Grandma’s Quilt Shop in Yuma, Arizona, employs Rain, a suite tailored for craft enterprises, to generate inventory descriptions and pricing for fabric designs. The owners report that this implementation reduces the time required to list items by between 60% and 80%.
However, there are drawbacks. Finnegan-Dehn has noted that the AI can occasionally feel “clunky” in its execution. Furthermore, the Notion AI add-on carries a subscription cost of $20 per month. As with any new technology, small business operators must weigh the potential efficiency gains against the cost of simply performing the task manually.
User tips
When evaluating whether AI can streamline your operations, consider the following principles:
- Look before you leap. Large language models rely on the data you input. To get the best results, you must feed them information in a convenient format. For many notebook services, this means using the platform exclusively for note-taking to avoid manual uploads later. Carefully weigh your options before committing to a specific AI-powered ecosystem.
- Work to your strengths. Identify the skills you currently lack in-house and determine if AI can assist with training or perform the tasks entirely. However, remain vigilant: AI models hallucinate and make errors. Ensure that human oversight remains in charge whenever accuracy is critical.
- AI isn’t always the best tool. It is perfectly acceptable to use established, off-the-shelf solutions when they are safer. For example, utilise existing payment processing platforms like Shopify or Square rather than attempting to build a custom solution using AI-generated code.
- Consider using local models for sensitive information. Online AI models carry the risk of leaking sensitive data, and companies often collect user prompts. Even if your business does not handle personal information, you may prefer not to share certain details publicly. In these cases, running an open-source model locally on your own laptop or desktop is a secure alternative to proprietary services like ChatGPT or Claude. Several large language models are now capable of running on standard hardware.
Key takeaways
- Small businesses should deploy AI primarily to automate administrative friction, such as scheduling and data entry, rather than attempting to replace core creative or strategic judgement.
- Privacy and accuracy require human oversight; operators must choose between proprietary cloud models and local open-source solutions based on the sensitivity of their data.
- Cost-benefit analysis is essential; the subscription fees for AI tools must be justified by the time saved against the risk of implementing a bespoke, potentially unreliable system.
- Integration is key; the most effective workflows utilise AI that sits within existing ecosystems, such as note-taking apps, rather than requiring constant manual data transfer.
Stay ahead of AI. Get the most important stories delivered to your inbox — no spam, no noise.
