Beyond the Note-Taker: What Advisers Are Getting Wrong About AI

A few weeks ago I was on a call with a firm owner. Directly authorised, 270-odd households, about 80 million under advice, young client bank, seven staff. They'd done everything right by the industry's playbook. They'd invested in best-of-breed technology. They'd been careful and considered. They had Microsoft 365, a market-leading back office system, a client portal, market-leading research and analysis tools, AML and risk profiling tools, and two separate AI note-takers because they couldn't decide which one was better. Plus Outlook, which had somehow become their workflow tool by default. And then this firm owner said something that stopped me in my tracks. "I do not want best of breed. I do not want a back office system that is going to give me best of breed because I've got best of breed now and it doesn't work." That wasn't prompted. It wasn't coached. It was someone who had lived with the fragmented model long enough to see through it. They went on: "We have multiple workflow options and none of them talk to each other. We don't have a single version of truth." And then, plainly: "I want a simple life. I want to be able to just have one system that does most of what we need it to do." That conversation crystalised something I've been arguing for a while now. The advice industry's relationship with technology - and AI in particular - is fundamentally broken. Not because the individual tools are bad. They're often very good. But because the model they sit inside is the wrong model. And no amount of clever AI bolted onto the wrong model will fix it. Let me explain what I mean. Blockbuster Had Great Customer Service Too Imagine for a moment that AI has arrived but Netflix hasn't been invented yet. Streaming doesn't exist. You still go to Blockbuster on a Friday night, wander the aisles, pick something, drive home, watch it, drive back on Monday to return it. That's the model. Now someone puts an AI screen in the Blockbuster entrance. You walk in, it asks you a few questions - what mood are you in, what did you watch last, who's watching with you - and it recommends the perfect film. Brilliant. A genuinely better experience than wandering the aisles for 20 minutes. You might even tell your friends about it. But it's still Blockbuster. You're still driving there. You're still limited to what's physically on the shelf. You're still returning it by Monday. You've improved the experience inside the old model. You haven't changed the model. Now imagine Netflix arrives. Every film, on demand, instantly. And then you put AI inside Netflix - personalised recommendations, smart discovery, contextual suggestions based on everything you've ever watched and everything you've searched for. That's a categorically different proposition. Not marginally better. Fundamentally different. That is what is happening right now in financial advice technology. The industry is putting AI screens in Blockbuster and congratulating itself on innovation. And I say that with some directness because I think the stakes are high enough to warrant it. Fifty Note-Takers and Counting There are now roughly 50 AI note-takers in the financial planning space. Fifty. You didn't know you needed a note-taker two years ago. Now it's the most important thing since sliced bread. Each one does more or less the same thing. It listens to your client meeting, transcribes it, generates a summary, maybe pulls out some action points. Fine. Useful. I'm not going to pretend otherwise. Recording and transcribing meetings is genuinely better than scribbling notes on a pad while trying to maintain eye contact with a client who's telling you about their retirement fears. But here's what's actually happened. You had 15 pieces of technology in your tech stack. Now you have 16. All the AI note-taker has done is added another piece of disconnected software to an already fragmented pile. It sits alongside your CRM, your back office system, your client portal, your compliance tool, your research platform, your document system, and whatever else you've accumulated over the years. It doesn't talk to any of them. Not really. Maybe there's a Zapier integration or a CSV export, but the data doesn't flow. It just pools in another silo. And here's the trajectory that should concern you. The note-taker companies know they need to expand to survive. So they'll start adding compliance checking. Document generation. Email drafting. Gradually trying to become more than a note-taker. But they're building outward from a single feature, not downward from a unified foundation. They don't have your client record, your fact find, your previous suitability reports, your compliance framework, your fee structure, your firm's investment philosophy. They have a transcript. That's it. You could deploy the greatest piece of technology ever in the world, but if you deploy it in the old paradigm - where is it going to go? That's the question nobody seems to be asking. We're so busy evaluating individual tools against each other that we've forgotten to question the model those tools sit inside. The paradigm is the problem. Not the tools within it. AI Is Electricity, Not an Appliance Most people in this industry haven't clocked this yet. The market treats AI as a product. You buy an AI tool. It does a thing. You evaluate it against competitors. You renew or you switch. That's the procurement model the advice industry knows and it's applying it faithfully to AI. It's the wrong framework. AI is a resource. Like electricity. In the early days of electrification, factories bought individual electric machines - an electric loom here, an electric press there. Each one was better than its steam-powered predecessor. But the real transformation didn't happen until factories were redesigned around electricity as a ubiquitous resource - wired into every station, powering everything, invisible in its operation, obvious in its impact. That took decades, by the way. The resistance to rethinking the factory layout was enormous. Sound familiar? AI should be the same. Nowadays we just turn a light switch on. We don't think about electricity as a product we procured. It's just there, doing whatever needs doing. AI should be a thing that's ubiquitous around us, that we deploy wherever we need to. For that to work, the AI and the platform have to be built by the same people, sharing the same data. The AI has to see everything - the client record, the meeting transcript, the previous documents, the compliance history, the firm's tone of voice, the investment philosophy, the fee structure. When you ask it to draft a suitability report, it shouldn't need to be told the context. It should already know. This is where I'll make a claim that might sound arrogant but I believe is factually accurate: we're the only advice tech company in the world, as far as I'm aware, who have a proper back office and our own AI. Most tech companies in this space have one or the other. The back office providers don't have AI - they partner with or integrate third-party tools. The AI companies don't have a back office - they have a feature looking for a home. We have both. Built together. Sharing the same data. And that means AI isn't a product we've added to a platform. It's a resource the entire platform calls upon wherever it's needed. When we built meeting intelligence, the AI doesn't just transcribe. It generates a structured summary across 14 categories. It extracts facts and updates the client record. It flags vulnerability indicators. It identifies action items. It feeds into document generation, so when the adviser sits down to write the suitability report the next day, the AI already has the meeting context, the fact find, the previous documents, the compliance framework. It drafts 80 to 90 percent of the report. The adviser refines and signs off. That's not a note-taker. That's electricity. The meeting intelligence is just one place it shows up. The same AI powers compliance checking, marketing content, client communication, onboarding, workflow automation, and - this is where it gets genuinely interesting - the client-facing experience. The Wrong Question Let's talk about the advice gap. It's the industry's favourite talking point and I think it's the wrong frame entirely. The numbers are well-rehearsed. There are roughly 39 million adults in the UK who don't receive regulated financial advice. The industry's response to this, broadly, has been: if we make advisers more efficient, they can see more clients, and the gap narrows. Technology is positioned as the enabler of that efficiency. Let's do the maths on that. The UK has roughly 4 million people receiving regulated financial advice. The average adviser has about 115 active clients. If technology makes advisers 20% more efficient - and that's a generous assumption - the advised population goes from 4 million to maybe 4.8 million. That's 800,000 more people. There are 60 million out there who still don't have anyone helping them with their financial future. Who cares about the 800,000? I don't say that dismissively. Every additional person who gets quality financial advice is a good outcome. But the framing itself constrains the thinking. Calling it an "advice gap" implies the answer is more advice. More advisers, or more efficient advisers, serving incrementally more clients. Think about it differently. We don't talk about a "personal training gap" because 80% of the UK doesn't have a personal trainer. We talk about a fitness problem. An exercise problem. The solution isn't more personal trainers. It's getting more people moving. Personal training is one part of the continuum - the intensive, personalised end. But gyms, apps, classes, group coaching, wearables, content, communities - that's where the scale lives. Financial advice is personal training. It's the high-touch, one-to-one, regulated end of the continuum. Valuable. Important. Not going anywhere. But it's not the answer to 60 million people being financially sedentary. What those 60 million people need is the financial equivalent of exercise. Education. Tools. Engagement. Guidance. Coaching. Not regulated advice - most of them don't need it yet, and many of them know it. What they need is someone (or something) to help them understand their pension, make sense of their ISA options, think about protection, plan for a house deposit, work out if they're saving enough. To get financially fitter. The question isn't how do we close the advice gap. The question is how do we help the entire population exercise financially. And you cannot answer that question inside the current model. You need orders of magnitude greater scalability. Not 20% more clients per adviser. 10x. 20x. From 115 clients to 2,000 or 3,000. Which sounds insane until you change what "client" means. The Advice Firm of the Future This is what I think an advice firm looks like in five to ten years. Not what I hope. What I think is structurally inevitable if you follow the technology to its logical conclusion. A continuum, not a category. Today, most advice firms have one proposition. You're a client or you're not. You pay fees or you don't. The relationship is binary, high-touch, and expensive to maintain. That limits the firm to maybe 100 to 200 families, depending on the model. The firm of the future has three tiers on a single platform. The first is a digital tier. Two thousand to three thousand people - maybe more - consuming financial education, using tools, working through guided journeys, interacting with an AI coach that knows the firm's philosophy and the individual's situation. These people pay a subscription or access the service through an employer scheme. They never speak to a human adviser. They don't need to. The AI handles it - not giving advice, but providing education, guidance, and engagement. Think of it as a gym membership for your financial life. The second is a pooled remote tier. Two hundred to five hundred families who need more than digital but don't need or want the traditional face-to-face model. They speak to qualified planners by video call. Their experience is technology-led but human-supported. The AI prepares for every meeting, drafts every document, handles the administration. The planner focuses on what a planner should focus on: understanding the client and giving good advice. The third is the dedicated tier. Fifty to a hundred families. The traditional model. Deep relationships, face-to-face, comprehensive financial planning. AI-augmented, not AI-replaced. The adviser still does the advising. But the administration, the document generation, the compliance checking, the meeting preparation, the follow-up - all handled. All three tiers run on the same operating system. Same client records. Same compliance framework. Same AI. Same data layer. A person who starts in the digital tier and builds wealth can graduate seamlessly to the pooled tier and eventually to the dedicated tier. No replatforming, no re-keying data, no starting again. The firm has known them from day one. Multiple propositions from one platform. AI changes the economics. Maintaining a second brand used to cost a marketing department. Now it costs 10% more per additional niche. A firm can have its core brand serving the general population, a specialist proposition for medical professionals, another for business owners, another for divorcees navigating financial separation. Each one speaks authentically to its community. Each one has its own AI - configured with the right tone, the right knowledge, the right compliance guardrails. One CRM, one compliance process, one operating system. Multiple faces. We call this the "niche of one" - the ability to speak to an audience of one person's specific needs at scale. AI makes that economically viable for the first time. The employer play. Consider this. An advice firm's business-owner clients already have something the firm needs: access to employees. Hundreds of them. Each one a potential client, at some point in their financial journey. A workplace financial wellbeing proposition, powered by AI, delivered through the employer, creates a funnel that goes from dozens to thousands. Employees get financial education, tools, guided journeys, access to an AI coach. They become financially fitter. Some of them, over time, need regulated advice. The firm is right there, already in the ecosystem, already holding the relationship. The acquisition cost is effectively zero. Anyone should be monetisable. Any level of wealth, any stage of their financial journey should be monetisable and able to be served profitably. That's not a statement about squeezing money out of people. It's a statement about inclusion. The current model excludes everyone who can't afford traditional fees. That's not a technology problem. It's a business model problem. And AI solves it - if you deploy it in the right model. The Data Question I want to address this directly because it's the question every adviser asks, and rightly so. Where does the data go? What happens when client information meets AI? The short answer: client data is architecturally separated from everything else. Not by policy. By architecture. Two separate systems, two separate purposes. Client records, meeting transcripts, suitability letters, financial plans - all encrypted, all isolated, all held in an environment that never touches the platform's operational data. Personally identifiable information is stripped before any data reaches an AI model and replaced once the output returns. That's an extra layer of safety beyond the encryption itself. ISO 27001 accredited. GDPR-compliant. EU-hosted infrastructure. Encrypted at rest and in transit. I mention this not as a sales pitch but because it matters for the argument. The reason many advisers are hesitant about AI isn't that the technology doesn't work. It's that they don't trust how the data is handled. And the reason they don't trust it is that most AI tools in this space are standalone products that were never designed with financial services data governance in mind. They were designed to transcribe meetings. Data security was an afterthought, addressed with reassuring compliance pages rather than architectural decisions. If AI is going to be electricity - everywhere, handling everything - then the data architecture has to be industrial-grade from day one. You can't bolt that on later. It has to be the foundation. One thing I've noticed, and I'll give the industry credit here: firms are asking much better questions than they were 12 months ago. The awareness around data processing locations, LLM data handling, and certification specifics is growing rapidly. Firms that couldn't articulate what they needed a year ago are now probing on specific processes. That's a good sign. It means the market is maturing. A Binary Decision I'm going to be direct, because the moment calls for it. Advisers are facing a binary decision. Not eventually. Now. Stay in the old model - best-in-class tools, fragmented stack, incremental AI bolted on the side, gradually increasing complexity with every new tool you add - or move to a new model where everything is connected, AI is interwoven, and digital scalability sits on top. There isn't a third option. There isn't a "wait and see how it plays out" option. Not because the technology won't let you wait - of course it will. You can run your current back office and a standalone note-taker and Excel spreadsheets for the next decade if you want to. The problem is that while you wait, the firms that didn't wait are compounding. More clients generate more data. More data makes the AI smarter. Smarter AI attracts more clients. The gap between firms on the old model and firms on the new model widens every quarter. It doesn't narrow. Netflix was first. Then Amazon Prime and Apple TV followed. The streaming model was inevitable once Netflix proved it worked. Others will follow into the unified model in advice tech. That's fine - competition is good and the industry needs it. But being first matters because the data compounds. For firms, my challenge is simple: stop asking "what AI tool should I buy?" and start asking "what does my business look like in five years?" If the answer is "basically the same but a bit more efficient," you're not thinking big enough. And if you're not willing to disrupt yourself, there are plenty of people that will be willing to disrupt you on your behalf. I know how this sounds coming from someone who co-founded a technology company. Of course I'm going to say the old model is broken - I built a business on the alternative. I'm aware of that. But I also hear it from the other side of the table. That firm owner I mentioned at the top of this essay didn't come to us looking for AI. They came looking for simplicity. For one system. For a single version of truth. They'd arrived at the conclusion independently, through years of lived experience with the fragmented model. Advisers should focus on advising clients and not building tech. But they've almost been driven to it because no one's bothered to properly solve the problem. Firms have become system integrators by default - stitching together disconnected tools, managing data flows between platforms, training staff on multiple systems, reconciling records that should never have been separated in the first place. That's not what they signed up for. That's not where their value lies. The answer is not more tools. It is not a better note-taker. It is not a shinier CRM with an AI tab. The answer is a fundamentally different architecture - one system, one data layer, AI everywhere it's useful, digital scalability on top. Making the Thing That Could Kill Us I'll close with something about how we think internally at Ningi, because I think it applies to advice firms too. We have to be making the thing that could kill us. And further than that, we have to be making the thing that could kill the thing that could kill us. That's three concurrent horizons. Today's product - the back office, the AI, the integrations, the features firms need right now. Tomorrow's threat - the pure-play AI companies, the digital-first propositions, the fintech disruptors who don't carry the weight of legacy architecture. And the threat to that threat - the consumer-direct model, the platform play, the world where regulated advice is one layer in a much larger ecosystem of financial wellbeing. We hold all three in mind simultaneously. We're building the operating system that powers today's advice firm. We're also building the digital-first, AI-native proposition that could replace it. And we're building the infrastructure layer that other brands, communities, and specialists can run on top of. If any one of those models wins, we're the platform underneath it. That's not arrogance. That's survival strategy. And it's the same strategic thinking I'd challenge any advice firm to adopt. Don't just optimise what you have. Build what could replace it. And then build what could replace that. The firms that will thrive over the next decade aren't the ones with the best note-taker. They're the ones that understood, early enough, that the note-taker was never the point. The point was always the model. The paradigm. The architecture of how advice gets delivered. The old model got us here. It deserves respect for that. But it won't get us where we need to go. The question is whether you're building for the next five years, or defending the last fifteen.

This comprehensive analysis draws on extensive industry research, real-world case studies, and insights from leading financial advisers across the UK. Our research team has compiled the most current data and trends to provide actionable insights for advisory practices.

Key Topics Covered: Technology adoption strategies, regulatory considerations, client experience optimization, operational efficiency improvements, and future market predictions.