In our customer-centric pillar of essential thinking, we introduced grassroots values of modern companies. That grassroots bias is a necessary condition for any product-led organization. A “parts culture” and observability are sufficient to make companies product-centric.
Parts are more than epic
The agile methodology (c. 2001) argues for breaking down work into epics and user stories. Project management was at the core of agile thinking. Projects are bespoke work constructs that are popular in the services industry. In fact, agile was borne out of lessons learned in professional services. Unfortunately, customer-visible projects in a product company are the biggest nemesis of product managers and developers – with specialized feature flags per customer, it is almost impossible to upgrade and maintain backward compatibility in a multi-tenant cloud without bespoke code and custom deployment scripts.
The Essential Methodology introduces a central concept of product parts, breaking down the anatomy of a product into smaller components called ‘parts’. Just like a human anatomy is broken down into organs, which are composed of tissues, and tissues of cells, a product can be broken down into its capabilities, which are composed of features (and sub-features), which then ultimately get implemented by APIs and webhooks. Parts can also be internal, entities that are continuously patched, upgraded, and deployed.
Parts bring a product-led thinking to companies otherwise struggling to manage projects. Unlike in the services business, projects are not metered, billed, analyzed, monetized, priced, or invoiced in a product company. Instead, it is product parts that are treated thus. Essential companies use ‘parts’ to bring leverage to their work – consistent, repeatable, packageable facets of a product that are potentially used by all customers, not just one. Parts, not epics, bring a discipline of doing fewer things for most customers, and understanding how to prioritize what matters most to a product (company).
The lifecycle of a product can also be easily managed by a special part called enhancement. Enhancements are unpublished parts that capture customer tickets, feature requests, customer chats, product documents, and developer transactions at one place. They also serve as a unit of integration in modern SaaS companies developing code in a multi-repo setup. Once ready, an enhancement gets published as a new or an improved product part.
Product-led companies will essentially leverage GenAI to auto-generate PRDs, release notes, blogs, SEO pages, digital campaigns, and customer newsletters.
Analytics and time travel
One of the hardest things about being product-led is gathering and analyzing data about product parts at the smallest grain of an API or a webhook, in real time. Product analytics, user analytics, and profile (employee) analytics are core to an essential mindset. Traditionally, product analytics has been expensive, in the hands of a chosen few, and complex due to external dependencies such as cumbersome data warehouses and fragile data pipelines.
The essential philosophy argues for complete data access for maximum people, so as to reduce the number of escalation meetings, product wastage, and incident response times. But more importantly, it is not just about point-in-time data but also historical data, so essentialist companies reduce recency bias in their decisions. Customer satisfaction “as of,” product quality “as of,” employee workload “as of” are examples of going back in time to grasp the history of customers, product parts, users, and people.
Parts supply chain
Once there is appreciation for parts (over epics), essential companies become good at understanding their supply chain – 3rd party cloud services and software, deployed internal micro-services, and people (engineers, designers, and developers). For example, a meaningful awareness of an external service provider’s (a) notifications and outages, (b) advisories and vulnerabilities, (c) new capabilities, and (d) economics of usage, all via automations, is crucial for reducing security risks and financial surprises down the line. Similarly, a continuous understanding of an engineer’s (a) skills and contributions, (b) deep work metrics, and (c) engagement (will and values); and the incidents, upgrades, and root cause analyses (RCAs) in cloud operations are critical data points for a product-led company.
Parts catalog
A product-led company has inventory, orders, and needs to promote its product innovation actively. Part inventory can get quite unwieldy, especially in e-commerce companies trading 3rd party products and services on their website. An inventory could also become unmanageable if unused features aren’t actively pruned and eliminated.
Essential product management mandates that PMs continuously prune, bundle, tier, package, and reprice product parts in their catalog. Only with semantic search can large inventories be managed holistically. The best product-led companies will fundamentally leverage LLMs for such a search culture that is equally exposed to end users for self-serve. PMs must also prioritize, customize, and ship part orders – new feature requests, defective part remediation – by actively joining support data with customer information coming from sales and customer success activities.
Large catalogs can also be only maintained on a continuous basis with the help of GenAI, as mentioned before, to generate blogs, SEO pages, roadmap, release notes, and website landing pages.