Market share and aisle share: Category analytics for CPG Industry

Reading Time: 5 minutes

Consumer Packaged Goods (CPG) companies, rarely have visibility of their market, competitors, or selling patterns because they depend on retailers as their end-customer. The presence of retailers in the process of selling CPG products to end consumers creates the need for multiple varied data access for CPG companies. As such, retail audit data provides visibility across product hierarchies, channels and areas. For example, since CPG companies rely on retailer data, they need to utilize their most accurate sources, like household panel data, to understand customer penetration, frequency, etc., across multiple hierarchies. It is of utmost importance for a category manager at a CPG company to analyze these datasets with multiple other datasets like media, leaflets, perfect store, EPOS, etc, and obtain business insights from it.

Analyzing market share: Why is it necessary?

For CPG category managers seeking to improve their sales performance and market share, the quickest route to success is often decoding business insights from category analysis focusing on increasing sales value and market share. This is successful when backed by exceptional product quality, product presence in stock, customer penetration, and share of aisle.

Analyzing market share is an industry-wide challenge impacting multiple functions. It affects the:

  • Product
  • Rate of sales
  • Weighted distribution
  • Frequency of customers
  • Spend per visit

Considering the differing objectives across functions to derive actionable insights, the market share is often looked at with inter-dependent functional data residing in disparate systems. This departmentalized approach to analysis forces the aggregation of data and information to a very high level.

This limits the ability to:

  • Track, trace, and analyze (on a single pane of glass) disparate data in the market share and take business decisions accordingly.
  • Quickly identify the root cause of complex systems and increase sales across channels and products.
  • Identify the erroneous sub-brands/variations and performances.
  • Detect and avoid hurdles in the sales cycle across retailers before they occur.

Fortunately for the business leaders tasked with managing and increasing market share, customer penetration, frequency, and share of aisle, there is hope. It comes in the form of Decision Intelligence platforms like Fosfor Lumin.

The data

This blog post presents Lumin’s capability to analyze the category sales data for a global CPG company. Capacities of Lumin by Fosfor include:

  • Proactively identifying anomalistic trends in sales value and subsequent diagnosis across channels, product hierarchy, pack type, variant, etc.
  • Identification of the market share drivers and simulation of their impact on sales.
  • Accurate prediction of the overall sales through multivariate forecasting.

Our dummy data consists of retail audits, household panels, and multiple internal data sources, i.e., EPOS system and the perfect store system. This data will be hosted on a cloud data warehouse (AWS Redshift) as virtualized views or reflections. The data’s schema consists of monthly market share data across different attributes. Key attributes include:

  • Key dimensions: Product type, pack type manufacturer, brand, sub-brand, pack size, variant, retailer, trade, etc.
  • Key measures: Market share, sales value, share of aisle, etc.
  • Date dimension: Audit dates

Descriptive analytics to help CPG companies explore the market share data

Let’s explore the data for the market share by brands in February 2023. We will start with a simple descriptive analysis by asking a few questions below.

What is the share of sales by brands?

Image 1: Share of sales

Fosfor Lumin understood our question in natural language and converted it into an analytical query along with the time filter. Lumin clearly calls out in Image 1 that Aspect is the market leader with the majority market share. It would be interesting to understand its spread across the different time frames. As a business, taking business decisions by looking out for sales value trends and understanding growth rates is essential.

By clicking the connected insights button, Lumin provides a set of simple and advanced analytical questions closely related to the metric or KPI being explored. Connected insights also suggest the growth rate in Lumin.

Image 2: Connected decision insights

As the next logical step, as seen in Image 2, it would be prudent to identify the trend where the highest share in sales has been observed. We ask Lumin this next question.

What is the monthly growth rate of Aspect’s sales value?

Image 3: Growth rate of Aspect. This Image also highlights a few features of Lumin viz, data export, chart switcher and download data

It is worth noting that Lumin also allows to:

  • Export the underlying data
  • Flexibility to switch the chart visualization

As you can see in Image 3, Fosfor Lumin dynamically selected the best-fit visualization to present the insight while also generating a contextualized narrative (in English) to explain the same. It is evident that August 2022 has shown a huge increase in sales. This information is interesting because it allows users to understand what drives this increase and from when. Since August has witnessed growth, why not explore the different pack sizes that contributed most to this outcome?

What is the growth contribution to overall sales for pack size in September 2022?

Lumin provides users with some powerful collaboration features to do the following:

  • Share the business decision insights with our colleagues within the organization.
  • Add comments (our understanding, observations, POV, etc.) on the insights while collaborating with a larger team.
  • Create personalized alerts and signals to better track the metrics or KPIs. We can get notified when our metrics breach a threshold that we can define.
  • Capture the insights in an insights storyboard called a workspace as we go about our exploratory journey. The workspace can further be shared with our colleagues within the organization.

As a collaborating user, I would expect Lumin to show me a waterfall chart. It does exactly that. Lumin’s semantic model understands common business terminologies like ratio, share, contribution, top or bottom N, etc. Furthermore, it does not need to be separately trained on that. As is evident from Image 4, the large pack size has contributed the highest to growth, whereas the multi-pack size negatively impacted it.

Image 4: Waterfall chart to depict the pack size impact on growth distribution in Aug 2022

Since we know the greatest change was observed in August 2022, the next obvious question would be: Why has the Sales value for Aspect changed in Aug 2022 vs. a year ago? To answer this, we can ask Lumin the following question.

Why has the sales value for Aspect changed in Aug 2022 vs. a year ago?

  Image 5: Diagnostic of change in sales in 2022

As seen in Image 5, Lumin – the Decision Intelligence platform, performs a cross-dimensional sales value analysis. This includes the product hierarchy (manufacturer, brand, variant) and product size. It also includes a type apart from the geographical hierarchy, evaluating across channels and trades we have already analyzed. In a matter of a few seconds, Lumin presents a detailed narrative and an interactive bubble chart that presents the deep-dive analysis across multiple levels, like product, manufacturer, and size.

For example, if we look through the product lens, we understand that the Sachet product type model contributed the most to the sales increase. The weighted distribution and rate of sales drove it. Also, for Bamboo 143, the sales value was the highest in the sachet for a large pack size.

Now as users, we can very well question these findings. As such, Lumin’s explainability layer becomes very important. It provides users with explanations on how it arrived at these insights, i.e., what was the input data, its sanity, the model that was applied, how we can interpret that model, its reliability, etc.

Now that we know the drivers for sales, we can simulate the same for Aspect by changing the drivers, e.g., reducing the weighted distribution and rate of sales by -10%.

Image 6: Simulation

As is evident from Image 6, a -10% reduction in the sales rate and weighted distribution further reduces sales by 9.2%. We can click on the “How did I arrive at the insight?” link to understand the computation Fosfor Lumin performed to simulate the results.

As we have seen the causes of change in sales of Aspect, why not explore other data sources to learn more about them? With Lumin, you only need to ask the question to land on another data source.

What is the trend of penetration and spend of Aspect?

Image 7: Trend on household panel data

The insight in Image 6 suggests that the penetration has been relatively constant, whereas the spend per visit has reduced. This suggests customer frequency has increased. That is why sales are increasing even though spending per visit has reduced. This environment essentially creates a change either in promotion or placement. Let’s explore the placement with its presence in perfect store data.

What is the share of main aisle by accounts for Aspect in Aug 2022 vs. July 2022?

Image 8: Retailer with maximum Change in Share of Aisle for Aspect

As evident from the Sales chart in Image 7, retailer B drove the increase in share of aisle in Aug 2022.

Never miss a critical change in your sales value with autonomous nudges

We can continue even further and ask even more questions. We can probe the data for more deep-dive analyses across various dimensions. We started our analysis by looking at a geographical cut. Still, we can very well look at manufacturers and brands, product types, package types, etc., and arrive at granular insights enabling increased confidence in enterprise decision-making.

Still, what if Fosfor Lumin could have autonomously detected and nudged us on this anomalistic trend? Is this possible? Yes, Lumin can spot deviations in our data as they happen with fully automated ‘nudges.’ Lumin’s anomaly engine synthesizes millions of data points in seconds to inform business users of significant changes as they happen. It detects what had shifted, when, why this occurred, and by how much.

Lumin has a robust setup of autonomous nudges on sales value. It configures this information to look out for anomalistic behavior in sales value across the product hierarchy comprising manufacturer, brand, and sub-brand. In this use case, Lumin proactively detected the sudden increase in sales in Bitterroot-Aspect-bamboo 87, as seen in Image 9. Lumin discovered that a higher rate of sales caused the increase. Additionally, the multi-variant has contributed the most to the rise.

Image 9: Autonomous nudge detected the unexpected increase in Jan 2023 in Aspect

Effectively, autonomous nudges proactively surfaced the anomaly for which we had to otherwise ask multiple questions on Lumin.

Get a sneak peek into future sales using predictive analytics

Last but not least, it will be crucial to understand how Aspect’s sales value will trend over the next few months. This will help businesses to be better prepared (in terms of reserves) to manage a possible increase in spending. Let us ask Fosfor Lumin to forecast the sales of Aspect.

What will be the sales of Aspect in the next two months?

Image 10: Forecast of Sales of Aspect

Image 11: Explainability of forecast

Lumin by Fosfor understands that we are asking it to predict the sales of Aspect. It runs multiple multivariate forecasting models under the hood and presents the most accurate model as an output. For our question, Lumin selected the multivariate time series ARIMAX model for predicting Sales, and it also identified the part cost and labor cost as the key drivers impacting the forecast.

As explained before, Lumin provides a detailed explanation in Image 11 of the input data. It provides the sanity, the prediction models it applied (including accuracy), how we interpret them, and the reliability of those models. This provides transparency of the model used and explains to the end user why to rely on model accuracy.

However, as a business, if we are aware that the rate of sales and price per volume are expected to increase in the next three months, it would be important for us to factor that increase into our forecast to make it realistic. Since we know that the price per volume is supposed to increase along with the sales rate, let us increase them to $125 and $700 in the next two months and simulate the output.

Image 12: Forecast simulation

As you can see, Fosfor Lumin factored in the increase in the drivers and reworked the forecast for us. The forecasted values have increased compared to the earlier forecast. This is more realistic and closer to what the business would expect. Lumin also helps end consumers understand the impact of variables, which could be changed by changing the drivers.

Create, curate, and collaborate on your business insights story

The entire analysis in this blog not only asked questions but gave an insightful data story that can be shared with the relevant stakeholders across your organization. Lumin, through the workspace storyboards, enables us to seamlessly create, curate and collaborate with our colleagues in the organization.

Image 13: Workspace storyboard

As we traverse through our insights exploration journey, we can create a personalized workspace and add our insights to that workspace. Once the analysis is complete, we can visit our workspace to refine or curate our story, add comments, etc., before publishing it to the stakeholders. Once published, the workspace is refreshed every time the underlying data is refreshed. Isn’t that amazing?

Lumin: The category analysis solution

As we showcased above, Lumin is a Decision Intelligence platform that delivers value across the category analysis and cause-effect in the sales journey of the CPG industry by providing category, brand, and marketing teams with a unified analytics experience across data sources. It can curate faster answers to what happened with a Google-like conversational interface. It explains business metric changes and how to improve business outcomes by understanding the drivers of those behaviors.


Shashank Pushp

Senior Specialist, Solutions Fosfor (Lumin), LTI Mindtree

With over a decade of experience in Retail and CPG analytics, Shashank is known for assisting large organizations in the spaces, get more out of their data. He is the Solutions Architect of the Decision Intelligence platform, Lumin at Fosfor. He is a data enthusiast and product evangelist, empowering customers with actionable data stories. When not at work, Shashank explores new travel destinations, reads drama, and watches stand-up comedies.

Latest Blogs

See how your peers leverage Fosfor + Snowflake to create the value they want consistently.

5 ways to foster data curiosity in your business

Just as early humans survived on curiosity to discover fire and invent the wheel, today’s organizations built on data need to know their data terrain well to survive and stay on the top. They need to understand what data resources are available to them and what challenges they face from disruptors, who always keep them on their toes.

Read more

Accelerate your production ML journey with Refract

As we all know, production ML (Machine Learning) is more engineering than machine learning. Building a prototype in machine learning has become very simple nowadays, all thanks to different open-source projects like sci-kit, TensorFlow, Keras, etc. But operationalizing that model to get the insights from the model which can be used in day-to-day business decisions is challenging and needs more engineering knowledge than data science knowledge.

Read more

An introduction to drift in Machine Learning

Read more