Prompt:
You are an expert data analyst. You are provided with a limited number of insightful questions and their corresponding datasets and Your task is to generate a clear, concise, and consolidated impact summary based solely on the provided Impacted Services Data.
- The summary must reflect insights from all the provided data pieces. No insight should be missed or overlooked.
- Ensure complete coverage with a data-driven, non-redundant, and well-structured response.
- The output should be readable and useful for both technical and business stakeholders.
Task:
The Impact analysis run results in a list of Impacted CIs and Service CIs, our intent is to summarise the impact on the Services. To allow this we have run a set of aggregated queries to determine the impact on services, based on location, categories, regions, impact levels etc. The JSON includes four aggregated datasets, each answering a specific insight question as part of the overall Impact Run Analysis.
Analyze the Insightful Questions and Corresponding DB Response provided in input and generate a number heavy structured, high-level impact summary.
# Objectives:
- Identify patterns and trends in impacted Services.
- Summarize root causes, dependencies.
Column Definitions: Use the following metadata to accurately interpret the data:
- Region: Stores the Service region. Region refers to the geographical area where the Service is located, particularly in organizations that operate across multiple locations.
- Site: Stores the name of the Site Group that the Service belongs to. A site group is a smaller, more specific grouping of locations within a region.
- Type: Indicates Product categorization tier 2 of configuration item or service. This level breaks down the broader categories from Tier 1 into more specific subcategories. For example: Under Hardware (Tier 1), Tier 2 could be: Laptops, Desktops, Servers, Printers. Under Software (Tier 1), Tier 2 could be: Operating Systems, Enterprise Applications, Productivity Software.
- Impact Level: A numeric severity indicator representing the impact level of a service. The impact level is classified into four categories: No Impact(10), Minor(20), Degraded (30), and Loss of Service(40). Used for classifying the services into one of the impact levels. Additionally, it should be considered for aggregating the count of different impact levels while generating summaries.
- Is ServiceCI: A binary flag indicating whether the impacted entity is a Service or Non -Service item (CI). '1' means it is a Service Item such as Customer Facing Service, Resource Facing Service, or Business Services, while '0' indicates a non-service entity such as hardware or physical devices.
- Category: Indicates Product categorization tier 1 of configuration item or service. This is the highest-level category. It typically refers to broader categories that represent major product or service areas in an organization. This could relate to general categories such as: Hardware, Software, Network, Applications, etc.
Guidelines:
DO:
- Generate the output directly from the analytical content. Do not include any introductory title or heading at the beginning (like "Impact Summary of Services"). However, retain all internal section headers
such as "Top Impacted Sites:", etc. Also, exclude the "Conclusion" section at the end.
- Back every insight with supporting data having numbers; avoid generic statements.
- Be concise: 2–3 lines per bullet or section.
- Maintain clarity, logical flow, and consistency.
DON’T:
- Avoid excessive detail, verbosity, or repetition.
- Do not reference specific incidents—focus on services and CIs.
Output Format:
-Use Markdown formatting selectively to bold key data points (e.g., site names, categories, impact types) to help them stand out in the summary.
- State the number of impacted services, using precise terminology.
- Use severity descriptors (e.g., "Loss of Service", "Partial Failure").
- Summarize affected services and related CIs by location and product category.
- Use bold headers like Top Impacted Sites:, Impact Distribution by Region: to separate sections clearly.
Input (Questions and Data):
{input}