> For the complete documentation index, see [llms.txt](https://eirin-ai.gitbook.io/eirin-ai-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://eirin-ai.gitbook.io/eirin-ai-docs/sentiment-report/what-it-contains.md). # What it contains Every section of the Player Sentiment Report, what it covers, and how to read it. Every report follows the same structure, so you always know where to look. Some sections depend on your sources. For example, the YouTube section only appears when YouTube is connected. ## Sections at a glance | # | Section | What it covers | | -- | --------------------------------------- | ------------------------------------------------------------------------------------------------------------------ | | | **Header** | The report period, the period it's compared with, and five headline cards | | 01 | **Overview** | What happened this period, a daily sentiment chart, and the split of positive, neutral and negative feedback | | 02 | **Release & event sentiment** | Sentiment on each source, where the complaints are, and a timeline of what changed the mood | | 03 | **Top risks & strengths** | What's hurting and what's working, ranked by how serious it is, with mention counts for each source | | 04 | **Topic analysis & community feedback** | Each topic scored and sized, with real player quotes | | 05 | **Trend monitoring** | New complaints, and what has got better | | 06 | **Product backlog input** | How the issues split by type, and what that means for planning | | 07 | **Top actionable recommendations** | Five next steps in order of priority, split into what you can do now and what to pass to your team | | 08 | **Feature requests** | What players are asking for, in order of priority, with quotes | | 09 | **Love / hate matrix** | What players love next to what frustrates them, on the same scale | | 10 | **Community language & keywords** | Player slang explained, plus keywords for your store listing and marketing | | 11 | **YouTube analysis** | What creators think compared with players, what creators miss, keywords and videos. Only when YouTube is connected | ## Section details ### Header * The report period (for example, 14 days), its dates, and the period it's compared with. * Five headline cards: sentiment score, positive share, negative share, reviews analysed, and primary risk. See [About the report](/eirin-ai-docs/sentiment-report/about-the-report.md) for what each card shows. * When needed, a short note explains how the numbers were worked out, for example which sources the main figures use, or why a source was left out. ### 01 · Overview * A one-line summary of how players feel, then a short explanation of what's pushing the mood up or down. * A **daily sentiment score** chart for the period, with the average and labels on key moments, such as a wave of "I quit" posts or a season preview. * The **sentiment split**: how much of the feedback is positive, neutral and negative. * A warning when platforms feel very differently, so you don't average them together by mistake. ### 02 · Release & event sentiment * **Sentiment on each source:** how many items Eirin read, the positive and negative share, and the trend (for example Stable or Declining), with a short note on each. * **Where the complaints are:** each complaint topic with how often it came up, its share, and its status, for example Escalating, Ongoing, or linked to the game version players blame. * **Timeline:** releases, community events and changes in mood, in order, including what's coming up, such as a season launch. If no update shipped in the period, the report says so and shows the community events that changed the mood instead. ### 03 · Top risks & strengths * **What's hurting:** each issue with a label for how serious it is (Critical, High or Moderate), a plain explanation, the total number of mentions, and mentions on each source. * **What's working:** the things players praise, with how often they came up. * A short closing note that reads the two sides together. ### 04 · Topic analysis & community feedback * Each topic with a sentiment score from −1 (very negative) to +1 (very positive), how often it comes up, how many mentions it has, and where. * A short note on why the topic matters. * **Real player quotes**, word for word and in the language they were written in, each with its source and date. This lets you check every number against what players actually said. ### 05 · Trend monitoring * **New complaints:** issues that are appearing for the first time or growing fast. * **Recently improved:** what's getting better compared with earlier periods. * This section is about what's new, not what's loudest. ### 06 · Product backlog input * **How feedback splits:** how the issues divide by type, such as fairness, performance or onboarding. * **What this means for planning:** whether you face one big problem or many smaller ones, and where the easiest wins are. ### 07 · Top actionable recommendations Five next steps in order of priority, each with how serious it is, why it matters, and who it affects. * **For you, do this now:** steps that don't depend on anyone else, such as posting a community update or changing your store listing. * **For the team, pass it on:** steps that need your engineering or live-ops team. ### 08 · Feature requests * What players are directly asking for, kept separate from bug reports because they are planned differently. * Each one with a priority (High, Medium or Low), a real quote, where it came from, and a link to the matching recommendation or topic. ### 09 · Love / hate matrix * What players love and what frustrates them, side by side, shown as the share of reviews that mention each one. * It shows at a glance whether frustration outweighs love, or the other way round. ### 10 · Community language & keywords * **Player slang:** the words your players use, what they mean, whether they're positive or negative, and where they appear. This includes local terms, so the quotes make sense. * **Suggested keywords** for your store listing and paid marketing, each with the reason to use it. * **Warnings:** which keywords to hold back, such as promises about something players currently say is broken. ### 11 · YouTube analysis This section appears when YouTube is connected. * A summary of what videos are saying about your game. * **What creators think compared with players**, with a warning when the two don't match. * **What creators miss:** topics players care about that creators don't cover. * **Creator keywords** and how often they come up, plus **player keywords** from the comments. * **Videos**, with views and likes, each marked as Your product, Adjacent title or Out of scope. ## How the report keeps its numbers honest {% hint style="info" %} * **Out of scope:** when people on a source are talking about a different product, for example a viral video about a game with a similar name, those items are marked Out of scope, shown separately, and left out of the main figures. * **Small samples:** when there isn't much feedback, the report tells you to treat the numbers as a direction, not an exact size. * **No data:** if a source had no reviews in the period, the report says so instead of guessing. {% endhint %} --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation by asking a question. Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters: ``` GET https://eirin-ai.gitbook.io/eirin-ai-docs/sentiment-report/what-it-contains.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.