> For the complete documentation index, see [llms.txt](https://docs.scaleconnect.com.au/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.scaleconnect.com.au/scaleconnect-docs/insights/brokerage-dashboard-1/ai-powered-operations.md). # AI-Powered Operations ScaleConnect uses AI across the platform to reduce manual work and amplify your capacity. From generating portfolio reviews to scheduling meetings, drafting responses, and surfacing insights — the AI handles the repetitive analysis so you can focus on advisory work and client relationships. *** ## The AI Stack ScaleConnect's AI isn't a single feature — it's a layer that runs through the entire platform. Each component serves a specific purpose, and together they cover the full broker workflow. | AI Component | What It Does | Where You Access It | | -------------------------- | -------------------------------------------------------------------------- | ------------------------------------------------------- | | **Broker Brain** | Loan book intelligence — queries across your entire client base, voice I/O | AI Brain button (top navigation) | | **Client Assistant** | Per-client AI — deep dives into individual portfolios, meeting prep | Client detail page → AI Assistant tab | | **AI Reviews** | Automated portfolio reviews generated from simulation data | Client detail page → AI Reviews, or bulk from dashboard | | **Calendar Agent** | Scheduling with availability checking, trigger monitoring, auto-booking | Settings → Calendar Agent | | **AI Message Suggestions** | Contextual reply drafts in broker-to-client chat | Client chat interface | | **Client-Facing AI** | ScaleApp chatbot — client self-service with their own portfolio data | ScaleApp mobile app (client side) | Each component draws from the same underlying data — client profiles, portfolio metrics, simulation outputs, and the [knowledge base](https://github.com/scaleapp-au/scaleconnect-docs/blob/main/insights/client-notes.md). They share context but serve different workflows. *** ## How AI Reduces Manual Work The value of AI in ScaleConnect isn't novelty — it's time. Every AI feature replaces a manual process that used to consume hours of your week. ### Portfolio Reviews **Before:** Writing a client review meant pulling data from multiple sources, drafting a narrative, formatting it, and sending it. For a book of 100 clients, that's 30–60 minutes per client per review cycle. **Now:** [AI Reviews](https://github.com/scaleapp-au/scaleconnect-docs/blob/main/insights/ai-reviews.md) traverse the client's simulation data — every property, loan, cashflow line, goal, and event — and generate a structured review automatically. The AI handles the data analysis and narrative. You add personal recommendations and send. The workflow: 1. Select a client (or a batch of clients). 2. Generate the review — the AI analyses their simulation output and knowledge base. 3. Review the draft — scan the AI's analysis and recommendations. 4. Add your personal touch — write specific advice, reference recent conversations, note upcoming opportunities. 5. Send — the review is delivered to the client via ScaleApp and stored in their knowledge base. What used to take 30–60 minutes per client now takes 5–10 minutes, with a higher-quality output because the AI catches details you might miss in a manual review. ### Meeting Scheduling **Before:** Tracking rate expiries, equity thresholds, and review dates in a spreadsheet. Manually reaching out when a trigger event hit. Booking meetings through back-and-forth emails. **Now:** The [Calendar Agent](https://github.com/scaleapp-au/scaleconnect-docs/blob/main/insights/calendar-agent.md) monitors your client base for trigger events, checks your availability, and handles the scheduling workflow. When a client's fixed rate is expiring in 90 days, the agent doesn't wait for you to notice — it prompts the outreach or books the meeting directly. ### Client Communications **Before:** Opening a client message, reviewing their file, and drafting a response from scratch — context-switching that eats into your day. **Now:** AI Message Suggestions generate a contextually relevant draft reply based on the conversation history and the client's portfolio data. You review, edit if needed, and send. The AI gives you a 70% complete starting point for most routine messages. ### Knowledge Management **Before:** Scribbling meeting notes in a notebook, forgetting to update the CRM, losing context between meetings. **Now:** Notes, transcripts, and AI conversations flow into the [knowledge base](https://github.com/scaleapp-au/scaleconnect-docs/blob/main/insights/client-notes.md) automatically. Upload a meeting transcript and it's indexed. Save an AI response as a note and it persists. Every piece of context you capture makes the AI smarter for next time. ### Client Self-Service **Before:** Clients calling or emailing you with routine questions — "What's my LVR?", "When does my fixed rate expire?", "How much equity do I have?" **Now:** The [client-facing AI](https://github.com/scaleapp-au/scaleconnect-docs/blob/main/insights/product-docs/scaleconnect-docs/scaleai/scaleai-for-clients.md) in ScaleApp answers these questions 24/7, using the client's real data. You handle the strategic conversations; the AI handles the day-to-day queries. *** ## The Knowledge Layer Behind every AI response is a structured knowledge layer that ensures answers are accurate, contextual, and traceable. ### How Knowledge Is Structured ScaleAI's knowledge isn't just a prompt — it's a multi-tiered system with defined data dependencies and interpretation logic. | Tier | What It Contains | Example | | ------------------------- | ----------------------------------------------------------- | ------------------------------------------------------------- | | **Educational content** | Property investment concepts, tax rules, financial literacy | How negative gearing works, CGT calculations, LVR definitions | | **Calculation logic** | How the simulation engine computes each metric | How equity is derived from property value minus loan balance | | **Product documentation** | How ScaleConnect features work | What each event type does, how cashflows are projected | | **FAQs** | Common questions and answers | "Why is my cashflow negative?" patterns | | **Interpretation rules** | How to read and contextualise specific data points | When an LVR of 80% is healthy vs concerning | Every simulation field has a defined upstream dependency chain (what feeds it) and downstream impact chain (what it affects). This structure is what allows the AI to answer "why" questions and "what if" questions — not just "what is" questions. ### Why vs What Most analytics tools tell you **what** your data says. ScaleAI goes further: * **"What is"** — "Your client's LVR is 72%." (Reading a number.) * **"Why"** — "Your client's LVR dropped from 78% to 72% because the Brisbane property appreciated by 6.2% this year while the loan balance reduced through P\&I repayments." (Tracing upstream causes.) * **"What if"** — "If your client refinances to interest-only, the LVR will hold at 72% instead of continuing to decrease, but monthly cashflow improves by $1,200." (Tracing downstream impacts.) This depth comes from the structured knowledge layer — every metric has a definition, a calculation method, a set of dependencies, and a set of impacts. The AI uses this structure to reason about your data, not just report it. *** ## Time Savings The cumulative impact of AI across the platform is significant. Here's a realistic picture of where time is saved for a broker managing 80–120 clients. | Task | Manual Time | With AI | Savings | | --------------------------------------------- | -------------------------------------- | ---------------------------- | ----------- | | Writing a client review | 30–60 min per client | 5–10 min per client | \~80% | | Tracking trigger events (rate expiry, equity) | 2–3 hours/week | Automated via Calendar Agent | \~90% | | Answering book-level questions | 15–30 min per query (manual data pull) | Seconds via Broker Brain | \~95% | | Responding to client messages | 5–10 min per message | 2–3 min with AI suggestions | \~60% | | Meeting preparation | 15–20 min per meeting | 5 min with Client Assistant | \~70% | | Fielding routine client questions | Ongoing throughout the day | Handled by ScaleApp AI | Significant | ### The Compounding Effect Time savings increase over time. As you build out the knowledge base — adding notes, sending reviews, uploading transcripts — the AI has richer context to work with. This means: * Reviews require less editing because the AI knows more about the client * Meeting prep takes less time because the Client Assistant has more history * Broker Brain gives more nuanced answers because it has more data to draw on The AI doesn't just save you time today — it saves you more time tomorrow than it did yesterday. Each piece of context you feed into the system makes every future interaction faster and more accurate. {% hint style="info" %} These time estimates are based on typical broker workflows. Your actual savings will depend on your client base size, how actively you use the knowledge base, and your review cadence. {% endhint %} *** ## Next Steps * [Getting the Most from AI](https://github.com/scaleapp-au/scaleconnect-docs/blob/main/insights/getting-the-most-from-ai.md) — practical guide to maximising your AI investment * [Ask AI](https://github.com/scaleapp-au/scaleconnect-docs/blob/main/insights/product-docs/scaleconnect-docs/scaleai/ask-ai.md) — the chat interface for Broker Brain and Client Assistant * [AI Reviews](https://github.com/scaleapp-au/scaleconnect-docs/blob/main/insights/ai-reviews.md) — automated portfolio reviews from simulation data * [Broker Brain](https://github.com/scaleapp-au/scaleconnect-docs/blob/main/insights/broker-brain.md) — loan book intelligence across all clients * [Calendar Agent](https://github.com/scaleapp-au/scaleconnect-docs/blob/main/insights/calendar-agent.md) — trigger-based scheduling and outreach * [Knowledge Base](https://github.com/scaleapp-au/scaleconnect-docs/blob/main/insights/client-notes.md) — building the AI's context for each client * [ScaleAI for Clients](https://github.com/scaleapp-au/scaleconnect-docs/blob/main/insights/product-docs/scaleconnect-docs/scaleai/scaleai-for-clients.md) — the client-facing AI in ScaleApp --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://docs.scaleconnect.com.au/scaleconnect-docs/insights/brokerage-dashboard-1/ai-powered-operations.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.