Excel Your KPIs with AI Copilot Start for free today
Your AI Copilot for Data
Definitive Guide to Decision Intelligence
Subscribe to our newsletter>
Get the latest products updates, community events and other news.
You may have heard of the concept of a metrics store, but examples of it in the market are few and far between. Some notable examples are from Airbnb, Uber, and LinkedIn. However, these companies refer to it as a metrics platform, which begs the question: what sets a metrics platform apart from a metrics store? If you share this same question, then this article is for you. It aims to clarify the differences between the two.
In the first two generations of architectures, distributed big data architectures such as traditional data warehouses or Hadoop have played a critical role in proving that people can extract real value from massive data. Still, their overall technical complexity ultimately limits its adoption to a small group of enterprises.
A metrics store is, in the simplest words, a virtualized middle layer between upstream data warehouses/data sources and downstream business applications to allow users to create and define business metrics as code / using GUI.
It abstracts away the complexity of underlying data warehouses, and provides end-users easy-to-understand metrics to easily reuse and analyze in downstream analytics, data science, and business applications.
A metrics Store typically serves downstream applications through either natively integration with the applications or an API-first approach to push metrics to be reused in downstream applications.
It is also commonly referred to as Headless Business Intelligence (BI), as the supporting use case of the metrics store is typically in-house business analytics. Unlike traditional BI reporting, a metrics store decouples metrics definition from the BI reporting and visualizations, providing a headless form of the metrics to enable more use cases beyond traditional BI, including data science, and analytics process automation.
The concept advocated by the metrics store is that “metrics can be defined once and then reused anywhere.” That means a metrics store can be used flexibly across BI visualizations, SaaS integrations, and an API, opening up tons of new use cases that were not previously possible with BI reporting.
For example, a SaaS company can consolidate data from various tools, such as Website Analytics, Product Analytics, and CRM, into a metrics store. Using this data, the company can calculate a lead score based on user activities on the website (e.g., content browsing, whitepaper downloads) and product usage (e.g., signups, key feature usage). If the score exceeds a certain threshold, it can be pushed to the CRM, and a salesperson may want to reach out to the user since the score indicates a certain level of interest in the product.
It is a balancing act to find a solution that satisfies the governance/control/consistency of data (IT-driven) and businesses' agility/self-service/autonomy (Business-driven). Implementing a metrics store requires businesses to centralize their business definitions and might take away part of their freedoms, and businesses were once self-services with metrics in their preferred BI tools.
The metrics store is a virtualized layer in the D&A stack, which means it is dependent on the performance of the data sources it is connected to. Despite many optimized caching designs, organizations may experience performance issues for high-concurrent usage of metrics stores. While a metrics store decouples metrics logic from data warehouses or ABI tools, it is still influenced by the D&A ecosystem to provide native support. However, the native support of ABI tools and data warehouses for a metrics store is currently limited, typically coming from cloud D&A.
To distinguish the metrics store from the metrics platform. Let's first look at the emerging architecture for modern data architecture from A16Z. To form a modern business intelligence architecture, not only a metrics layer (aka metric store) is needed, but also a data warehouse and a data modeling layer.
It means to form a metric platform, a company not only needs to establish a standardized layer of metrics but also needs to make sure the supporting data warehouse and data modeling go along well with its metrics store.
What is a metrics platform, then? A metrics platform combines three components: the data warehouse, the data modeling, and the metric store.
To mitigate the risks of adopting a metrics store, a company needs to closely evaluate the how these three components can work together. For example, when providing the metrics store as the consumption layer to the business users, the data team who manages the data warehouse and data model needs to make sure that when metrics are added or updated, the data warehouse needs to compute and back-fills metrics correctly.
After all, the metrics store is typically a standalone logical layer on top of the data warehouse; the metrics' performance is highly dependent on the underlying data warehouse. It is critical to ensure the underlying data warehouse works organically with the metrics store.
Kyligence Zen is a low-code metrics platform that connects the storage tier to the analysis and output tier. It provides three components: data warehouse, metrics store, and data modeling. This consolidation reduces the learning curve, shortens the startup time, and reduces the total cost of ownership for users.The benefit of this combination is that your company doesn't need a dedicated data engineering team to set up your metrics store. As long as you can pull data from your business application, you can get started quickly.Kyligence Zen provides unique values to your business:
2. Built with business users in mind: Our metrics platform was originally inspired by one of our commercial bank customers. We saw this customer successfully enable their business teams to self-serve, define, and analyze their business metrics by rolling out our metrics platform. This makes our metrics platform unique. We built Kyligence Zen with business users in mind from day one. It is a no-code platform, so business users can easily navigate through it.
3. Analytics-as-code: We recognize that successful metrics platforms require collaboration between the data engineering and business teams. As your team grows, you will eventually have a dedicated data team to curate and govern your data. Our product is designed to meet the needs of data engineers so that the metrics platform you build with Kyligence Zen can grow with your business needs.
In this article, I explain the concept of a metrics store, which is a virtualized middle layer between data sources and business applications that allows users to define business metrics as code or using a GUI. This abstraction eliminates the complexity of underlying data warehouses and provides users with easy-to-understand metrics for downstream analytics and business applications. I also discuss considerations for migrating to a headless metrics store from traditional BI and distinguish the metrics store from the metrics platform.
At Kyligence, we understand the challenges of implementing a metrics platform. For a large company, building a metrics platform is not simply adding an extra metrics store on top. The key consideration when adopting a metrics store is how to make the data modeling, data warehousing, and metrics store work organically.
For startups and small-sized companies, Kyligence Zen can be the first step to jump-start your metrics platform building. It can enable more use cases, such as in-app analytics and growth analytics, with ease.
In summary, the metrics platform brings significant value to the enterprise: with the completion of the data analysis, the BI report may end its lifecycle. In contrast, enterprises will stick with the metrics platform that tightly integrates with their business workflow and be reused anywhere, generating more possibilities.
If you are interested in learning more about building your metrics platform, here are some next steps:
2. Sign up and quickly try out the metrics that interest you by applying a metrics template to your metrics store in Kyligence Zen.
3. Read more to learn about the metrics store here: https://kyligence.io/blog/understanding-the-metrics-store/
4. You can also read this article for more information: https://thenewstack.io/demystifying-the-metrics-store-and-semantic-layer/
Learn more about Kyligence Zen here or try out Kyligence Zen for free.
The following references were used in creating this document:
Learn about the fundamentals of a data product and how we help build better data products with real customer success stories.
Unlock potentials of analytics query accelerators for swift data processing and insights from cloud data lakes. Explore advanced features of Kyligence Zen.
Optimize data analytics with AWS S3. Leverage large language models and accelerate decision-making.
Optimize data analytics with Snowflake's Data Copilot. Leverage large language models and accelerate decision-making.
Discover the 7 top AI analytics tools! Learn about their pros, cons, and pricing, and choose the best one to transform your business.
Discover operational and executive SaaS metrics that matter for customers success, importance, and why you should track them with Kyligence Zen.
Unlock the future of augmented analytics with this must-read blog. Discover the top 5 tools that are reshaping the analytics landscape.
What website metrics matter in business? Learn about categories, vital website metrics, how to measure them, and how Kyligence simplifies it.
99 Almaden Boulevard Suite #663
San Jose, CA 95113
+1 (669) 256-3378
Ⓒ 2023 Kyligence, Inc. All rights reserved.
Already have an account? Click here to login
A complete product experience
A guided demo of the whole process, from data import, modeling to analysis, by our data experts.
Q&A session with industry experts
Our data experts will answer your questions about customized solutions.
Please fill in your contact information.We'll get back to you in 1-2 business days.