Extended history and more granular data available within AWS Cost Explorer

AWS Cost Explorer offers an easy-to-use interface to visualize and understand your AWS cost and usage over time. Previously, Cost Explorer provided up to 13 months of cost and usage data at daily and monthly granularity as a free feature, with an option for hourly granularity over the past 14 days as a paid feature. However, customers requiring multi-year analysis or understanding cost drivers with resource level details couldn’t complete these tasks in Cost Explorer. Now, with extended multi-year history and more granular resource level data within Cost Explorer, customers no longer need to leave Cost Explorer to perform the above analysis. Cost Explorer now offers the following features for free:

1. Multi-year data at monthly granularity: you can now access up to 38 months of historical data at monthly granularity, allowing for more comprehensive long-term trend analysis.
2. Resource-level data at daily granularity: Cost Explorer offers resource-level data at daily granularity, spanning over the past 14 days, enabling you to dive into your cost drivers.

This blog will guide you through the various use cases that these features will enable you to address, and show you how to enable and leverage these features within Cost Explorer.

What use cases you can address with these features

Default 14-month data

Monthly Billing Year-over-Year Analysis: Upon receiving your AWS bill for the month, you might want to analyze and compare costs with the corresponding month from the previous year. This comparison will provide insights into how your AWS spending has evolved year-over-year. Prior to the introduction of 14 months of history, the oldest data you had access to was for your current month from the previous year due to the 12 month historical lookback period. However, by the time you got your AWS bill for prior month, you had lost access to the corresponding month data from previous year, prohibiting you from performing year-over-year analysis. But now, with one extra month of historical data, you can perform this analysis.

As seen in the screenshot below, you can now compare October 2023 cost with October 2022 cost in November 2023.

Figure 1. Cost Explorer sample graph: 1-year lookback view

Multi-year data at monthly granularity

Year-over-Year Cost and Usage Analysis: Your business, applications, and architecture have matured in the past few years and you are wondering how your AWS spend has evolved along with that. You can now perform this analysis for the past three years in Cost Explorer and get a better understanding of your year-over-year or quarter-over-quarter spend patterns. In Cost Explorer, you can now select a start date within the past three years and set an end date to any date up to the present day to create multi-year data view. You can filter and group this data by various dimensions, such as service, account, region, usage type to perform comprehensive analysis.

As displayed in the screenshot, you can study how the RDS spend has changed in different regions in the past three years.

Figure 2. Cost Explorer sample graph: 3-year lookback view of Amazon RDS spend by region

Resource-level data at daily granularity

Advanced variance analysis to identify cost drivers: You are responsible to optimize cloud cost in your organization. You have noticed variance in your Lambda spend in the past two weeks and you are wondering what is causing that at the resource level. You can now perform this analysis in Cost Explorer and pinpoint the exact Lambda functions responsible for the variance. You can then discuss these functions with respective teams to differentiate intended from unintended spend.

You can filter Cost Explorer for Lambda service to focus on Lambda cost and usage.

Figure 3. Cost Explorer sample graph: previous visibility into AWS Lambda cost and usage

You can notice your Lambda cost and usage has more than doubled on October 6th as compared to your average. To understand what specific Lambda functions caused this spike, you can group your data by resource.

Figure 4. Cost Explorer sample graph: current experience with Lambda functional detailed view

Now, you understand that three functions are primarily responsible for your Lambda cost spike on October 6th.

Though the graph displays the top 9 functions, you can download costs per Lambda function in a CSV file format for offline analysis and sharing with your team.

Data Transfer cost and usage analysis: You are responsible to optimize cloud cost in your organization. You have seen spike in your Data Transfer usage and you want to understand the reason behind it. You can now perform this analysis in Cost Explorer.

You can start by filtering your Cost Explorer view by Data Transfer usage types, such as “DataTransfer-Out-Bytes” for data transfer out to the internet, to focus on only Data Transfer related usage. You can then group this usage by service to understand what services this increased usage is associated with.

Figure 5. Cost Explorer sample graph: Data Transfer usage by service

From the above graph, you can determine that the increased Data Transfer usage on November 4th is associated with EC2-Instances.

You can now filter your Cost Explorer view to show only the data transfer usage related to EC2-Instances. Moreover, you can group the data by resource, providing insights into the specific EC2 instances involved in the data transfer.

Figure 6. Cost Explorer sample graph: Data Transfer usage by resource

In this example, it is clear that data transfer associated with instance i-096caae10eb96ce07 is responsible for the Data Transfer usage spike on November 4th.

How you can enable this data

You can enable multi-year data at monthly granularity and resource-level data at daily granularity from Cost management preference page available to management account of your organization. Once these features are enabled, they can be used by all accounts in your organization.

Figure 7. Usage estimate in Cost Explorer preference

Enabling multi-year data at monthly granularity: You can click on the checkbox to enable this feature. Once enabled, your data should be available within 48 hours in Cost Explorer.

Enabling resource-level data at daily granularity: You can select specific services you want to enable resource data for. The services are listed in the order of their contribution to your AWS bill, with the most expensive service on top. Once enabled, your data will be available in Cost Explorer within 48 hours.

Please refer to Cost Explorer user guide to understand what IAM actions you need to enable the above settings and step-by-step process.

Conclusion:

With the increased data and granularity available in Cost Explorer, we are enabling customers who want to perform in-depth cost and usage analysis. To learn more about Cost Explorer, visit our user guide.