The Core Problem: Administrative Overhead at Small Scale
Small businesses face a structural disadvantage. Every hour a founder or solo operator spends on invoicing, note-taking, or scheduling is an hour not spent on the work that generates revenue.
The challenge is not a lack of ambition or capability — it is bandwidth. AI models, particularly large language models integrated into existing productivity workflows, are increasingly capable of absorbing the rote, repetitive layer of business operations without requiring significant technical expertise to deploy.
The key insight is not that AI does everything well. It is that AI does enough of the right things well enough to matter.
Case Study: A London Tutor Builds a Second Memory
Sam Finnegan-Dehn works in fundraising by day and tutors mathematics and philosophy students from his London home by night. His tutoring practice demands more than teaching — it requires lesson planning, reading curation, assignment creation, invoicing, and staying current with academic research.
He turned to AI not to replace his teaching judgment, but to handle the connective tissue of his business. He describes the result as having a second memory — one that links ideas scattered across digital notebooks and surfaces them when relevant.
Why Notion AI Won the Comparison
Finnegan-Dehn experimented with Claude and ChatGPT before settling on Notion AI. The deciding factor was integration. His tutoring notes already lived inside Notion, so an AI layer built directly into that environment eliminated the friction of importing or re-entering information elsewhere.
The practical applications he relies on include:
- Meeting summaries — Notion AI records client sessions (with consent) and generates automated summaries, which Finnegan-Dehn reviews to identify where a teaching technique is falling flat and adjust his approach accordingly.
- Goal decomposition — He writes a “North Star” objective, such as reaching a target number of clients by year-end, and asks Notion AI to generate the concrete steps required to get there based on his existing profile and notes.
- Invoicing and lesson notes — Routine documentation that previously consumed time is now drafted by the AI and reviewed by him.
- Social media synchronisation — Content is generated and synced across platforms without manual duplication.
The pattern here is consistent: Finnegan-Dehn retains strategic and creative control while delegating the mechanical execution to the model.
Industry-Specific AI: The Rain Example
Not every small business needs a general-purpose productivity assistant. Some benefit more from AI tools built specifically for their sector.
Grandma’s Quilt Shop in Yuma, Arizona, uses Rain — a software suite designed for craft businesses — to generate inventory descriptions and pricing for its fabric stock. The owners report that this reduces the time required to list items by 60 to 80%. That is not a marginal efficiency gain. For a small retail operation, it represents the difference between a manageable workload and a bottleneck that limits growth.
The Rain example illustrates an important principle: when an AI tool is trained and structured around the specific vocabulary, logic, and workflows of an industry, it outperforms a general-purpose model on that narrow task. Small business owners should evaluate both categories — horizontal tools like Notion AI and vertical tools like Rain — before committing to a solution.
The “Good Enough” Threshold
AI does not need to be perfect to be valuable in administrative contexts. Drafting an invoice, summarising a meeting, or generating a task list from a stated goal does not require flawless output — it requires output that is close enough to be quickly reviewed and approved. That threshold is already met by current models for a wide range of secretarial and organisational tasks.
Known Limitations
Finnegan-Dehn himself describes certain Notion AI behaviours as “clunky.” This is a fair and honest assessment of where the technology currently sits. Integrated AI assistants can be inconsistent in how they interpret context, and the quality of output depends heavily on the quality and structure of the input data.
LLMs also hallucinate. They generate plausible-sounding but factually incorrect information with enough regularity that any task requiring precision — financial calculations, legal language, medical guidance — should remain under direct human oversight.
Cost is also a real consideration. Notion AI’s add-on runs at $20 per month. For a part-time business or a solo operator with thin margins, that cost needs to be weighed honestly against the time it saves.
1. Commit to the Ecosystem Before You Start
LLMs perform best when they have access to well-organised, relevant data. For notebook-based AI tools, this means using the platform for note-taking from the beginning rather than trying to migrate existing records later. Evaluate your options carefully before locking into an AI-powered workflow, because switching costs are real.
2. Target Your Weakest Operational Areas First
Identify the tasks that consume the most time relative to the value they produce, or the skills that are genuinely absent in-house. AI is most impactful when it fills a real gap rather than augmenting something already working well. Start there.
3. Do Not Use AI Where Established Tools Are Safer
Payment processing is a clear example. Platforms like Shopify or Square are battle-tested, compliant, and reliable. Attempting to build or customise payment infrastructure using AI-generated code introduces unnecessary risk. AI is a tool for augmentation, not a replacement for proven infrastructure.
4. Use Local Models for Sensitive Data
Online AI services process your inputs on external servers, and data handling practices vary significantly between providers. For any business information that is commercially sensitive, personally identifiable, or simply private, running an open-source model locally — on a laptop or small desktop — eliminates the exposure entirely.
Several capable LLMs can now run on consumer hardware without requiring cloud connectivity. This is no longer a technically demanding option reserved for developers. It is a practical privacy measure available to any small business owner willing to invest a modest amount of setup time.
Choosing the Right Tool for Your Context
The comparison between Notion AI and Rain is instructive precisely because they serve different needs. Notion AI is a horizontal productivity layer that works across many business types. Rain is a vertical solution optimised for a specific industry. Neither is universally superior — the right choice depends on the nature of the work.
A useful framework for evaluation:
- Horizontal tools (Notion AI, ChatGPT, Claude) — best for businesses with diverse administrative needs across writing, planning, and communication.
- Vertical tools (Rain and equivalents) — best for businesses with high-volume, repetitive tasks specific to a defined industry or product category.
- Local models (Ollama, LM Studio, and similar) — best for businesses handling sensitive data or operating in regulated environments where data sovereignty matters.
The Broader Shift in Small Business Operations
What Finnegan-Dehn’s approach demonstrates is not a dramatic technological transformation. It is something more modest and more useful: a systematic reassignment of low-value tasks from a skilled human to a capable machine, freeing that human to focus on the work that actually requires their expertise.
The 60–80% time savings reported by Grandma’s Quilt Shop on inventory listing is not an outlier. It reflects what happens when AI is applied precisely to the right problem — repetitive, structured, high-volume — rather than deployed broadly in the hope that something useful emerges.
Small business AI adoption works best when it is deliberate, scoped, and honest about limitations. The tools are ready. The question is whether the workflow is designed to use them well.
The administrative burden of running a small business has always been a tax on the people doing the real work. AI does not eliminate that burden entirely — but applied with precision, it reduces it enough to matter. For a part-time tutor in London or a quilt shop in Arizona, that reduction is not an abstraction. It is hours returned to the work that actually builds the business.
Comments (0) No comments yet
Want to join this discussion? Login or Register.
No comments yet. Be the first to share your thoughts!