The Joe Reis Show - The Information Product Canvas: A Shared Language for Data
I was lucky enough to be able to chat to Joe Reis about my recently published book, "The Information Product Canvas," and my diving into writing a second book.
The Joe Reis Show - The Information Product Canvas: A Shared Language for Data
I was lucky enough to be able to chat to Joe Reis about my recently published book, "The Information Product Canvas," and my diving into writing a second book.
Watch
Listen
https://open.spotify.com/episode/7fJrhBZG3rXgwFg3gg7iee?si=Z4nJfCAkSLiX8EeyGFkVcQ
Google NotebookLM Mind Map
Google NotebookLM Brief
Detailed Briefing: Shane Gibson - The Power of the Information Product Canvas
Date: October 26, 2023
Source: Excerpts from "Shane Gibson - The Power of the Information Product Canvas" (Podcast Transcript)
Key Speakers:
Shane Gibson (SG): Author of "The Information Product Canvas".
Joe (J): Interviewer, author, and podcast host.
1. The Gruelling Journey of Authorship
Shane Gibson kicks off by reflecting on the arduous process of writing a book, particularly his first, "The Information Product Canvas," and the even more challenging experience of tackling the second.
Writing as a "Nightmare": SG describes writing as an "awful" and "not enjoyable" process, despite being a prolific writer who doesn't consider himself a "natural writer." This contrasts with Joe's enjoyment of writing as a means to "think more clearly about a topic."
Forced Effort vs. Natural Enjoyment: SG admits, "I had to force myself to spend time doing the writing process." His "buzz" comes from simplifying complex ideas into comprehensible words: "how do you take those things I could talk about for ages and just waffle on and get them into... a thousand words that make sense to somebody else."
The "Vanity Project" & Secondary Benefits: His primary motivation was to "scratch" an "itch" he'd always had: "I just want to write a freaking book and I've tried a few times and I've failed and I wasn’t going to fail this time." A secondary, and deeply satisfying, benefit has been seeing strangers successfully use his "pattern" without his direct involvement, solving "a small problem in their data work."
Beyond Just Publishing: The real challenge wasn't just writing the book, but making it a book he "didn't hate," particularly in an era where AI can "crank out a book in an hour." True value, SG suggests, comes from distilling complexity into simplicity, a process that "takes all the time."
The Importance of "Voice": SG learned from an editor that his early drafts lacked his unique "voice," making the content hard to understand for a general audience. This underscored the need to tailor the book for those "outside our domain," making it "easy to read."
Self-Publishing Woes: SG highlights the significant effort involved in self-publishing beyond writing, including navigating Amazon Kindle Direct Publishing, which he calls a "bloody nightmare." He notes the platform’s inconsistent print quality, the need for continuous uploads to check layout changes, and the unpredictable review process upon final submission. He also jokes about his book mistakenly appearing in the "online dating" category on Amazon UK due to the subtitle "stakeholders love it."
2. The Core Problem: A Lack of Shared Language in Data Teams
SG identifies a decades-old, pervasive issue in data teams: a fundamental communication breakdown between data professionals and stakeholders.
The Misunderstanding Loop: "data teams talk to stakeholders ask them what they want stakeholders tell them the data team think they understand they go away and build something they give it back to the stakeholder and the stakeholder says that's not what I wanted." This often leads to frustrating exchanges like "but that's what you ask for."
Learned Behaviours & Disparate Languages:Stakeholders: Have been "trained by us as data people to talk in terms of reports and lists and data requests," e.g., "I want a dashboard and I want to have this table and I want these three fields on it."
Data Teams: Converse in terms of "source systems," data formats, and "conform[ing] the dimension" – technical details irrelevant to business outcomes.
The Missing Link: "there was no shared language in the middle." This void led to the development of the Information Product Canvas.
3. The Information Product Canvas: A Shared Language and Solution
The Information Product Canvas, inspired by the Business Model Canvas, provides a structured, collaborative solution to the communication gap.
Inspiration from the Business Model Canvas: SG and his team "lucked across this idea of the business model canvas," a tool for summarising business strategy "on a page with a small number of boxes." Its success lay in "this combination of a template that was easy to understand and a process to fill it out." Crucially, it was open-sourced, leading to other iterations like the Lean Startup Canvas.
The Canvas's Impact: Using an A3 paper with "12 boxes" and "a specific language of how we filled it out," stakeholders began providing "what we needed." The key shift was focusing "more on what's the action and the outcome that they're trying to achieve not what's the data they wanted."
What is an "Information Product"? SG’s term, which predates the popularity of "data product," defines a bounded entity that encompasses:
Multiple Delivery Types: "a dashboard a report a data service an AI chatbot an MCP service."
Necessary Data: "should contain the data that is needed to to power that product."
Targeted Personas: "focused on a specific set of personas."
A Clear Boundary: Like different breakfast cereals, they are distinct products with different flavours (delivery, data, personas) but serve the same fundamental purpose (breakfast/solving a business problem).
Iterative & Product Thinking: The canvas facilitates an "information value stream," integrating "product thinking" into data work. This process moves from:
Ideation & Discovery: Identifying business problems that "potentially we can solve... with data" and brainstorming "three to five potential ways we could use data to solve that."
Prioritisation: For larger organisations, discovering "20 things we could do" and then prioritising "the next most valuable product to build."
Data Factory/Information Factory: Moving into design, build, deploy, maintain, and enhance phases.
Benefits of the Canvas:Faster Feedback: Increased speed from problem statement to product delivery leads to quicker feedback loops and adjustments.
Deliver More Value: Enables chunking down large problems into "smaller things," delivering value incrementally (e.g., a week for a revenue model part, rather than six months for a lifetime value model).
Versatile Application: Primarily used in the "discovery phase," but also valuable for conceptual data modeling and understanding the "language of the business."
Greenfield vs. Brownfield Projects: The canvas is adaptable for both new projects (greenfield) and existing systems (brownfield).
Greenfield: Can quickly "lightly discover all the things we could potentially rebuild" (e.g., replacing 1,000 legacy reports). Teams can identify "150 potential products," then group and prioritise "the most valuable ones to build first" before filling out the canvas in more detail.
Brownfield: Useful if current requirements gathering is slow, misunderstood, or poorly attended. It aims to fix the problem of delivering "not what they wanted."
4. Challenging Traditional Data Modelling & Embracing Agility
SG critiques traditional data modelling practices, advocating for more agile and collaborative approaches.
Workshop Exhaustion: SG recalls "multi-day" data modeling workshops that were "waste" for stakeholders, highlighting the need to "not steal those three days of time from those stakeholders" and instead "chunk it down into small bits that are valuable."
Isolated "Heroes": He criticises the "data modeling heroes" who "model a really good data model... in isolation," resulting in "no buy in" and "no feedback loop." This reinforces the problem of delivering "what they asked for But it's not what they wanted."
Incremental & Iterative: The product thinking approach promotes "incremental and iterative versus big design up front," a traditional pitfall of "enterprise data model[s]." This is crucial due to "constraints on time constraints on budgets."
Data Product vs. Information Product: While acknowledging market preference for "data product," SG stands by "information product" due to its historical use within his work. He clearly differentiates between a "data asset" (purely a table) and an "information product" (data, logic, delivery, targeted users).
Critique of DIKW Hierarchy: SG questions the Data-Information-Knowledge-Wisdom hierarchy, seeing it as a "convenient abstraction" but potentially "wrong" because "it assumes that that knowledge and wisdom and information are sort of a linear progression from data," ignoring feedback in the other direction. He struggles with its practical application: "where I struggle with data information knowledge and wisdom is like okay I get it but how do I use it?"
The Blurring Lines of Data & Knowledge: Joe notes the increasing vagueness of boundaries in data, especially with AI, integrating "library sciences" (ontologies, taxonomies) into the data world, which were historically "separate gangs on the street." SG is actively seeking to learn from library sciences to better understand concepts like "context" for AI.
Accessibility of Knowledge: Both SG and Joe lament the inaccessibility of knowledge in some domains (e.g., data modelling, library sciences). SG points out that the success of books like Joe's "Fundamentals of Data Engineering" and his own "Information Product Canvas" lies in their ability to "take that complexity and make it simple" for newcomers.
Kimball's Enduring Legacy: SG discusses Ralph Kimball's continued success in data warehousing due to the "readily available" patterns, "easy to read" books, and freely shared ideas. Kimball "made his money out of being the expert... to learn the idea in a better way."
"Mixed Modelling Arts": SG suggests that just as data modelling benefits from combining techniques, publishing could adopt a "mixed analogy," leveraging various approaches.
5. Publishing Strategies & The Future of Content Creation
The conversation delves into the evolving landscape of publishing, marketing, and continuous learning.
Beyond Publishers: The Author's Burden: SG notes the "bullshit" expectation that publishers will handle marketing and sales. Instead, they prioritise an author's "following" and market access. Authors "do all the work they take all your money."
Marketing & Word of Mouth: Marketing typically drives initial sales, but "marketing only lasts for about probably the first month of a book then it's word of mouth." Positive feedback from users, like the person who used his canvas and it "just worked," is the true measure of success.
Continuous Improvement through Feedback:Courses as Feedback Loops: SG plans to run "bad version[s]" of his course for the next book, making them "almost free" to "learn from people what messaging is not getting through." This provides "feedback to make a better book."
Public Accountability: SG uses public deadlines (e.g., LinkedIn updates on writing progress) as a "forcing function" to maintain discipline, especially since he doesn't naturally enjoy writing.
The Evolving Definition of "Data Modelling": SG and Joe discuss whether merely adding metadata or context to data constitutes data modelling. Joe argues "hell yeah," expanding the definition beyond just changing data structure. This aligns with the idea that an "information product" doesn't have to be complex; a "paperclip" is still a product, just like a simple data output can be an information product.
The Iterative Nature of Writing: SG likens writing a book to an "iterative agile process," where the table of contents isn't fixed, and unexpected "new chapter[s] turn up that just makes sense."
AI's Impact on Content: SG muses that in the AI world, "we're going to start paywalling the ideas more because people won't pay for the content." He notes the current inability of LLMs to scrape Amazon's website, highlighting content providers' efforts to control their material.
This briefing summarises the key takeaways from Shane Gibson's reflections on his publishing journey and the innovative approach of the Information Product Canvas, all delivered with that classic Kiwi directness. Chur!