The New Retail Revolution: From "People-Product-Place" to "North Star Metric"

   

In the retail industry, data still matters!

 

The retail industry cannot be ignored when it comes to industries that have the greatest demand for data analytics.

 

Retail enterprises have accumulated large amounts of data during their operations and many business scenarios that can be optimized but waiting for data analytics recommendations, for example, how the average transaction value affects overall turnover and whether active customers play a positive role in it. Data analysis has played a vital role in helping retail enterprises to make informed judgments to improve their daily operations.

 

Since the pandemic, brick-and-mortar sales are facing challenges brought by mounting costs and customer retention issues. However, in the context of COVID-19, the new retail industry, represented by community retail and community group buying, has achieved substantial progress. With the new norm of COVID-19 prevention and control requirements, more retailers expect to steadily increase their turnover by providing attentive community-based services. Under this goal:

 

• Some retailers focus on adjusting their product portfolios, and expanding the proportion of daily necessities (such as rice, flour, grain, and oil) to increase their average transaction value;

 

• Some retailers focus on developing private domain users and launch promotions that match their behavioral characteristics to consolidate their active customer base further.

 

However, a common thread that runs through all these fragmented approaches is still yet to be developed. Retailers are in urgent need of a more instructive data analytics and metric system so they can better track the effectiveness of all these initiatives and the synergy of different initiatives in pursuit of their business goals.

 

In this article, we will build a simplified "North Star Metric" system for the new retail business as examples to share with you how to set and break down metrics, how to associate business goals with your North Star metric system, and how to align goals of different teams, and how to form a unified data policy to achieve management and business goals.

 

New retail and North Star Metric

 

Data analytics for the traditional retail industry is essentially inseparable from the model of "People-Product-Place," which has well-defined the analytic perspectives of the retail industry on the horizontal axis.

"People-Product-Place" data analytics model for the retail industry (image from Kyligence)

 

But if we look vertically, we can see this "People-Product-Place" model can still be optimized. For example, among all these metrics, which bears the most business significance? How are these metrics related? What is the top-down hierarchical relationship? As we mentioned earlier in this article, i.e., retailers still feel very confused as they cannot find a common thread to guide the business forward.

 

The new retail industry puts more emphasis on user experience. The approaches we discussed at the beginning of this article are all user-centered, for example, increasing average transaction value or consolidating active customers. Therefore, we need to focus more on business goals and explore the relevance of various metrics. This is also why more new retailers use "North Star Metric" to guide their data analysis.

 

A "North Star Metric" is the absolute core metric that is relevant to the business/strategy at the current stage. Once established, this metric will shine like a North Star in the sky, guiding the team to move forward toward a common goal.

   

Different industries have their own "North Star Metric". Even in the same industry, different "North Star Metrics" may be set for different business goals. For instance, the catering industry may use customer satisfaction or table turnover as the "North Star Metric", whereas convenience store chains may focus more on sales or gross profit margins. It is quite obvious that the "North Star Metric" is strongly related to business goals.

 

Top-level design - metrics should reflect business goal

 

However, during implementation, most data analytic platforms cannot provide a clear representation of the business goal with their reports or metrics, let alone drive management.

 

Another challenge is that when defining the "North Star Metric," how can the enterprise ensure the accomplishment of the metrics, and how can they monitor the process? We need to break down the key metrics and the goal and associate them with multiple specific metrics so they can be easily assigned to different roles.

 

Metrics breakdown is essential for the "North Star Metric" to play its role. These sub-metrics should be trustworthy and could serve as a valuable reference for the users; they should follow the company's data policy.

 

Breakdowns – consistent data policy ensures metrics credibility

 

However, when trying to break down metrics from different departments or metrics scattered in various processing links, the metric names, data policy, or calculation methodology among the upper and lower levels or even the same metrics are mainly inconsistent. This will significantly reduce metrics' credibility and affect their value in decision-making.

 

Kyligenze Zen: a smart metric management and analytics platform

 

Issues like inconsistent statistical standards and failure to align metrics with business objectives will impair business decision-making and objective achievement. As intelligent metric management and analytic platform built on top of Kyligence's OLAP platform, Kyligence Zen has provided an innovative solution for these concerns.

 

• Goal management

 

Kyligence Zen provides goal management and metric alignment capabilities to enable organizations to decompose management goals into relevant targets from the managing perspective, set reasonable result or progress metrics, and continuously track metrics, further driving organization's digital transformation.

 

• Metric catalog

 

Through the Metrics Catalog capability, organizations can easily define and manage metrics, ensure consistent metrics definition, and further enhance data credibility.

 

Let’s embark on an amazing journey to experience the magical power of Kyligence Zen in just three steps!

 

Find North with Kyligence Zen for the new retail industry

 

[1. Setting the "North Star Metric"]

 

For the new retail represented by community retail and community group buying, as retailers pay close attention to their turnover increase, we will use total sales (commonly known as GMV) as the "North Star Metric", which is also a barometer that shows how the business is progressing. As the achievement of this business goal will be affected by multiple factors, we also add total transactions (i.e., one of the most fundamental factors), average transaction value, and active customers as our key metrics.

 

By doing so, we can develop a simple "North Star Metric" system for the New Retail industry, which takes total sales as the common thread, and breakdowns the achievement of this goal into three perspectives, complete transactions, average transaction value, and active customers, to guide the implementation of the goal and progress tracking.

   

As a complete methodology set, a systematic and rigorous approach must be taken to establish and break down the "North Star Metric." We used this article's simplified version to illustrate the implementation process quickly. It is also shown in the above figure that the breakdown metrics can be further linked to other metrics with consistently defined analytic dimensions, for example, time, space and commodity attributes, etc.

 

[2. Import datasets and industry metric templates]

 

For this "North Star Metric" system, the metrics we focus on mainly involve transactions (total transactions and average transaction value) and customers (active customers). Therefore, we need to import the relevant data first.

 

Kyligence Zen supports the integration of multiple data sources. Here we take adding data sources via uploading CSV files(Note 1) as an example to show how to import the two datasets: transactions and customers:

   

 

Upon completion of dataset import, metrics can be created. Regarding the issue of "inconsistent statistical standards for metric data" we mentioned earlier, the metrics catalog feature of Kyligence Zen can help to build a unified metric library:

 

• Create metrics based on the dataset just imported;

• Design and maintain metrics in a uniform format in the metric catalog. Kyligence Zen supports basic metrics, composite metrics, derived metrics to allow hierarchical management of the metric library;

• Set consistent dimensions for the metrics based on your business needs.

   

Generating such a consistent metric library is as simple and rapid as importing the dataset we mentioned above!

 

In Kyligence Zen, we can maintain metric definitions with YAML files and import them with one click. Here we have prepared a complete set of retail industry metric templates for you, including over 30 typical application metrics from the perspective of transaction performance and customer contribution. It also includes total sales, total transactions, average transaction value, and active customers we need for the "North Star Metric":

 

   

Required metrics can be obtained directly from the metric catalog (image from Kyligence)

 

As can be seen from the whole process, with the metric catalog of Kyligence Zen, we can define metrics with consistent statistical standards, which will then be followed within the whole data analysis process. This ensures that the metrics we break down from the "North Star Metric" system have really high data credibility and can help retailers to make informed business decisions.

 

[3. Integrate metrics into business goals and drive management]

 

Upon completion of metrics establishment, it is natural to start designing our reports. However, the traditional report forms cannot address the issue of "difficulties in aligning metrics with business goals" mentioned above. Data itself does not create value directly, and the same is true for metrics, whose value needs to be transmitted and reflected by empowering goals.

 

One of the key implications of using the "North Star Metric" is that it allows us to focus on the most core business goals first. Therefore it is necessary to reflect the association between business goals and corresponding metrics on the platform. With the goal management of Kyligence Zen, we can follow up on the implementation of the preset "North Star Metric" and the associated metrics. More importantly, the goal and metrics association allows us to clearly see current gaps and risks.

   

Based on our "North Star Metric" system, in the Goal tab of Kyligence Zen, we can set Total SalesAchieves as the Tier 1 goal, and add three sub-goals under this goal, namely, Total Transactions Achieves, Average Transaction Value Reaches and Active Customers Exceeds, and associate each goal with its corresponding metrics and set the target value.

 

 

In this way, we set up a framework featuring association between business goals and corresponding metrics, which helps us to centrally track goals for different strategies. Meanwhile, we can invite different teams to collaborate on these sub-goals. For example, from this goal dashboard shown above, we can easily track the business performance:

 

• The Total Sales in the "North Star Metric" system has still not attained the desired level, with a current completion rate of less than 60%

 

• Although the target of Total Transaction metric has already been met, the other two metrics (average transaction value and active customers) are both behind the target, especially the active customers, which only reached less than 2/3 of its target

 

As the Zen metric catalog has ensured a consistent metrics definition, the conclusions we derived here can be persuasive among people in different departments or roles, thus helping enterprises to respond and adjust their business strategies more promptly. For instance, in the next stage, enterprises can focus on growing their active customer base by launching more promotional campaigns based on the behavioral characteristics of the target population. By doing so, we have gradually set up a metrics-driven management workflow when working on our business goals.

 

With an aim to perfectly match our "North Star Metric" system, we can continue to add the next level of subgoals based on sub-goals. Such a structure also provides a clear basis and direction for further attribution analytics.

 

Kyligence Zen also provides a dashboard to support multi-dimensional data analytics. The same as goal management, dashboard is also designed based on the governed metric catalog.

 

We can directly select the metrics from the metric library that fits our business needs. For example, for Average Transaction Value, we can check the impact of the associated Average Basket Size. Another example, for Active Customers, we can observe how the associated Repeat Purchase Rate plays a role. All these metrics follow the same data policy, business users can check the metrics data from the time dimension (such as order date), spatial dimension (such as store location) and attribute dimension (such as item category). You can import our retail industry metric template and continue the exploration.

     

In addition to the establishment of the "North Star Metric" system, many new retailers want to better cater to the changes in consumer demand and consumption patterns under the new normal of COVID-19, with a clear focus on the pre-set business goals and how to check the progress with metrics. In these cases, they will inevitably encounter the issues we mentioned earlier, i.e.:

 

• Difficulties in aligning metrics with business goals
• Inconsistent definition of metrics

 

You can use the metric catalog, one of the core capacities of Kyligence Zen, to establish consistent and trustworthy metrics. People from different departments and of different roles can design metrics within the same metrics and goal framework. By doing so, the organization can focus more on the business itself rather than wasting time and efforts repetitively on inconsistent standards.

 

After that, they can build their goal and metrics dashboard directly with the metric catalog, thus improving metrics reuse and facilitating agile collaboration. Meanwhile, enterprises can directly associate business goals and corresponding metrics with the goal management capability of Kyligence Zen, which is also the most direct and reliable basis for status tracking, risk warning, and decision making.

 

Metrics provide a better solution to quickly leapfrog from data display to insight gaining, providing a highway between data and decision making.

 

As a metrics and goal management platform, Kyligence Zen can help leading enterprises build a consistent metric system, implement and continuously improve digitized management, and achieve business goals. Kyligence Zen can be a solid foundation for enterprises to equip themselves with more efficient operation and management mechanisms and add data value.

---

   

Note 1: The dataset is from Kaggle, with Mohamed Harris as author. Some columns of the data used in this article are adjusted on the basis of the original dataset.

 

The retail scenario dataset (two CSV files) and industry metrics template (a YAML file) mentioned above are included in this zip file:

---

   

---