Get AI summaries of any video or article — Sign up free
Ryan Young - Is it (layer) cake: Thinking in content levels thumbnail

Ryan Young - Is it (layer) cake: Thinking in content levels

Write the Docs·
6 min read

Based on Write the Docs's video on YouTube. If you like this content, support the original creators by watching, liking and subscribing to their content.

TL;DR

Users often arrive at docs with a specific job to be done, even if their initial intent sounds broad (like “accept payments”).

Briefing

Stripe’s documentation frustration isn’t coming from a lack of information—it’s coming from mismatched “content levels” that force users to make the wrong decisions at the wrong time. After reviewing customer feedback, Ryan Young found recurring complaints: too many options presented too early, information that users couldn’t apply in context, incorrect scope (either too broad or too narrow), unclear next steps, and missing links to the resources needed to finish the job. Even when guides contained the right facts, users still felt stuck because the docs didn’t help them reason through what to do next.

Young frames the problem through a user-job lens. Most people arrive at Stripe docs thinking they want to “accept payments,” but their real goals are more specific—creating subscriptions, adding buy buttons, building checkout sessions, or sending invoices. When the information architecture assumes a single path (or a single type of user), users experience choice overload and disorientation. Feedback also points to a growing mismatch between the original developer-first audience and newer users who either don’t want to write code immediately or don’t want to invest time in coding when payments are a side project. In those cases, “too technical” isn’t just about language—it’s about where the docs ask users to jump into implementation before they’ve understood the shape and scope of the integration.

To diagnose what was missing, Young compares docs to mall navigation. A mall directory helps orient visitors (“you are here” and where stores are), while signs help once they’re close to their destination. But granular store maps and high-level overviews alone don’t cover the middle distance where people need to make decisions—figuring out which path fits their situation before they start executing. That analogy leads to a three-layer “docs layer cake” framework.

At the top is the orientation layer: high-level overview pages that guide users to the right entry point for their use case, such as a payments landing page with direct links to detailed guides. The middle is the decision layer—new, intentionally designed content that helps users reason about options and understand the scope of their integration before writing code. Young argues this layer is absent in many doc systems, which tend to jump from overview to implementation.

Finally, the implementation layer contains the familiar low-level assets: integration guides, quick starts, and API reference material. These are most effective when users already know where they are and what they’re trying to build, so the docs can focus on execution, accuracy, and copy-pasteable steps. Young also notes that some implementation guidance can be embedded in the product experience (tooltips, in-context flows, immersive quick starts) to reduce context switching.

Stripe’s response includes dedicated “no code” documentation that follows the same layered pattern: overview landing pages, low-level UI-driven how-tos, and a decision-level doc that defines integration pieces, explains options in plain language, and ties them to common use cases. Early feedback suggests this middle layer reduces frustration for non-technical and time-constrained users and makes complex concepts like recurring payments easier to explain.

In the Q&A, Young describes how Stripe collects feedback through thumbs up/down prompts (with follow-up multiple-choice questions and free text), supplemented by user interviews and triangulated with page views and CSAT. When negative feedback appears, the team retraces user steps to determine whether the issue is documentation clarity, product gating, or a mismatch between what the UI exposes and what the docs describe. The overarching takeaway: docs work best when they match the user’s journey—orient, decide, then implement—rather than treating all readers as if they’re ready to execute immediately.

Cornell Notes

Ryan Young’s core finding is that documentation failures often come from missing “content levels,” not from missing facts. Stripe’s users arrive with specific jobs (e.g., subscriptions, checkout sessions, invoices) even when they start with a broad intent like “accept payments.” Young proposes a three-layer framework: an orientation layer (high-level entry points and navigation), a decision layer (middle-depth guidance that helps users reason about options and integration scope), and an implementation layer (execution-focused guides, quick starts, and API reference). This middle layer matters most for users who are overwhelmed by options or not ready to jump straight into code. Stripe’s no-code docs adopt the same structure, and early feedback suggests it reduces frustration by clarifying scope and next steps before implementation begins.

Why did users still feel frustrated even when Stripe’s docs were packed with details and code samples?

Feedback clustered around several failure modes: (1) choice overload—too many options scattered across pages made simple tasks feel overwhelming; (2) inconsistent “how to get started” paths—users encountered options that felt contradictory or mutually exclusive; (3) information that couldn’t be applied—pages might contain the right facts but lacked the context needed to act; (4) wrong scope—guides were sometimes too broad or too narrow for the user’s immediate need; and (5) unclear next steps—users finished an initial “getting started” flow but didn’t know where to go next. The result was disorientation: users couldn’t translate doc content into a clear sequence of actions.

What does the “mall” analogy add to the docs problem?

Young compares docs navigation to mall wayfinding. A mall directory helps visitors orient themselves (“you are here” and where stores are). Signs help once people are close to their destination. But a system that only provides extremely granular maps (low-level implementation) and very high-level overviews (landing pages) leaves a gap: the middle distance where people need to decide which path fits their situation. That gap mirrors the missing “middle level” in many documentation architectures.

How do the three layers—orientation, decision, implementation—map to user needs?

Orientation is for finding the right entry point and understanding where the user is in the journey (e.g., a payments landing page that links directly to use-case-specific guides). Decision is for reasoning about options and integration scope before execution—users may know what they want at a high level but need help understanding the shape of the integration and what pieces matter. Implementation is for executing the build: integration guides, quick starts, and API reference details (including exhaustive field-level information). Young emphasizes that implementation content works best when users already know their destination.

What changes did Stripe make for “no code” users, and where does the decision layer show up?

Stripe created no-code documentation with the same layered pattern: high-level landing pages for navigation, low-level UI-driven guides (including embedded UI functionality and text steps), and a dedicated decision-level doc. That middle doc lists use cases first, defines typical integration pieces in plain language, and then expands a selected job to explain the overall process while linking to deeper docs. The goal is to help users understand options and scope without requiring immediate code writing.

How does Stripe distinguish whether negative feedback is a docs issue or a product issue?

Young says the team tries to confirm it isn’t a documentation clarity problem first, then retraces user steps to see whether the user could do the task but wasn’t guided correctly, or whether the task is impossible because the product doesn’t expose the required UI. A concrete example: a page received many negative ratings despite normal page views because it described a beta feature, but users weren’t actually gated into the beta experience—so the UI buttons referenced by the docs weren’t visible.

What feedback and metrics does Stripe use to measure documentation effectiveness?

Stripe relies on continuous collection of feedback: every page includes thumbs up/thumbs down with follow-up multiple-choice prompts (e.g., whether something wasn’t clear or was inaccurate) plus free-text comments. Young notes this feedback can be biased toward frustrated users, but filtering out noise still yields useful signal. Interviews are described as the best method but are time- and resource-intensive. For quantitative triangulation, the team looks at page views alongside CSAT to spot pages with high traffic but high dissatisfaction.

Review Questions

  1. Which specific user complaints Young cites point to missing decision support rather than missing technical content?
  2. How would you redesign a docs site that jumps from overview pages directly to code-heavy quick starts using the orientation/decision/implementation framework?
  3. In the mall analogy, what would “signs” and “directory maps” correspond to in a documentation system, and why does that matter for user drop-off?

Key Points

  1. 1

    Users often arrive at docs with a specific job to be done, even if their initial intent sounds broad (like “accept payments”).

  2. 2

    Choice overload and inconsistent entry guidance can make docs feel contradictory, overwhelming, and disorienting.

  3. 3

    Correct information still fails if users can’t apply it in context or if the guide’s scope doesn’t match their immediate need.

  4. 4

    A three-layer framework—orientation, decision, implementation—helps align documentation with the user journey.

  5. 5

    The decision layer is the missing middle in many doc systems: it helps users reason about options and integration scope before implementation.

  6. 6

    No-code documentation works best when it follows the same layered pattern, including a decision-level page that defines integration pieces and ties them to use cases.

  7. 7

    Feedback loops combine thumbs up/down prompts, user interviews, CSAT, and page-view triangulation; negative feedback is investigated by retracing user steps to separate docs issues from product gating problems.

Highlights

Users weren’t frustrated because Stripe lacked information; they were frustrated because the docs forced them to make the wrong decisions too early or couldn’t help them determine the next step.
Young’s “mall” analogy pinpoints a gap between high-level overviews and low-level execution: the middle distance where people need decision support.
Stripe’s no-code decision-level doc lists use cases first, defines integration pieces in plain language, and then expands jobs with process explanations and links to deeper docs.
Negative doc feedback can be caused by product gating—users may not see beta UI elements that the docs describe, turning accurate text into an unusable guide.

Topics

  • Docs Information Architecture
  • User Journeys
  • Content Levels
  • No-Code Documentation
  • Technical Writing

Mentioned