AI for SMEs: Luke Chaffey’s playbook to automate and drive ROI | Ep 255 | DevReady Podcast

Luke Chaffey, Managing Director of AIWise, joins host Anthony Sapountzis (CTO and Co-Founder of Aerion Technologies and Co-Founder of DevReady.ai | AI-Powered App Planning for Non-Tech Founders ) in this episode of the DevReady Podcast to unpack how small and medium-sized enterprises can turn AI from hype into business value. From early chatbot and augmented reality experiments to production-ready automation, Luke shares practical lessons on strategy, tooling and evaluation frameworks that keep outputs accurate, consistent and on brand. Expect real examples: cutting document creation time, prioritising high-value leads, and natural-language product search, plus a simple roadmap to get started with AI today.

Luke charts his journey from web development to co-founding Capillary Digital with David Koch, then into startups building AR, AI and chatbots for international clients before launching AIWise. Early prototypes paired AR “place-in-room” visualisation with AI trained on product data to answer questions and support sales, an approach that saw stronger uptake in the U.S. than Australia. Alongside hands-on tech, Luke built authority with 400+ articles and frequent media appearances, emphasising how writing and communication skills accelerate technical leadership and client education.

Inside AIWise, the playbook starts with clarifying strategy and a roadmap, then moves to implementation (or hand-off to internal dev teams) and leadership training. For automation, Luke mixes code and no-code: Python for control, reliability and richer state handling; Make (and, for developers, n8n) for fast proofs-of-concept that clients can self-manage. The north star is embedding AI directly inside core systems and workflows, shipping quick wins via no-code where sensible, then migrating in-house for scale, orchestration (containers, agents) and long-term maintainability.

On common missteps, Luke sees SMEs either assuming AI is “only for big companies” or dabbling without context. The remedy is to start hands-on with models like ChatGPT or Gemini, provide rich business context, and then rigorously validate outputs. He warns about hallucinations and “sycophantic” responses; best practice includes cross-model checks, human fact-checking in unfamiliar domains, and a robust evaluation framework that bulk-tests answers for factuality, tone and correct source use—crucial for customer-facing chatbots.

Results follow when AI targets repeatable work: prioritising referral conversations so teams focus on high-value customers; turning bullet points into polished job descriptions in seconds; and compressing a tax report workflow from eight hours to two by auto-drafting the repeatable 80%. For newcomers, Luke suggests a simple path: start with ChatGPT for everyday tasks (emails, briefings, document drafts), then add no-code automation with Make to streamline processes; explore off-the-shelf tools (e.g., voice with ElevenLabs) before going bespoke. When needs grow, engage experts to productionise and integrate so AI delivers reliable, measurable outcomes rather than one-off experiments.

#AI #Automation #SMEs #SmallBusiness #ChatGPT #Gemini #Makecom #n8n #Python #Ecommerce #AugmentedReality #DevReadyPodcast #AerionTechnologies #LukeChaffey #AIWise