Iqbal Ali is an experimentation consultant and coach who empowers product teams to run successful experiment programs. With over 20 years of expertise in UX design, development, data, and optimisation, he helps his clients improve the quality and velocity of their experiment programs. He is also the author of the "CRO Tales" comic series on LinkedIn, self-published comics creator and graphic novelist.
Chat interfaces are all talk. When we actually want to do stuff, the best we can get out of them are clear instructions. But what if we wanted more generative AI involvement with the doing part? In previous episodes, we tackled this by creating workflows in n8n with connected...
Last time, we created a research workflow. This workflow connected to a spreadsheet containing our user feedback, then randomly sampled it and extracted relevant themes. We used a small model, so the entire workflow was highly energy-efficient. Wouldn’t it be nice if we could quantify the themes by counting...
We have a spreadsheet full of user feedback. We’ve put them all into a vector store, and now we can search the feedback semantically using theme labels, specific mentions, etc. (as we did in previous videos). But there’s a problem. This workflow still requires us to familiarise ourselves with...
How do you create effective workflows that deal with large volumes of user text data? Like, say you want an LLM to analyse large volumes of user feedback for insights as a workflow. In this instance, what you need to learn is a process called Retrieval Augmented Generation (RAG)....
We’ve done quite a bit in n8n, but we haven’t really created any meaningful connections. And by meaningful connections, I mean connecting to the likes of Google, Slack, Airtable, or other tools you might be using. So far, we have a manual process for entering user text into a...
I don’t know about you, but I get bored manually trudging through an interaction with an LLM. We say something, it says something back, and on and on, until we get what we need. Worse still, in previous steps we took a single interaction with ChatGPT to analyse user...
