By Elizabeth Pridgeon
Share this post
Understanding brand performance is critical for brand and category managers. It’s the foundation for decision-making, for setting a strategy that helps them drive growth and win shelf space.
To reach effective brand analysis, pressing questions must be answered promptly:
- What’s the overall category trend?
- How are my brands performing compared to the category and to my competitors?
- What’s causing my market share growth/decline?
- Where should we invest?
- How do we gain more basis points?
Yet, answering these questions is often easier said than done. Data stays underutilized in “data lakes” where vast amounts of data pool together. Data analysts then sort through the data to run reports and generate insights. It can take days to several weeks, depending on the scope of the question.
Even data scientists, with their advanced skill sets, encounter tedious and time-consuming obstacles as they try to translate complex analysis into actionable insights.
In both cases, it’s difficult for data analysts and data scientists to quickly and effectively communicate their findings to the business team. The insights probably won’t be current and may not even be actionable to answer the pressing questions.
This past year revealed how quickly environments can change, especially in the CPG and retail space. Businesses have seen ten years of change in one year. Customers’ attitudes shifted, supply chains were disrupted, and uncertainty spread.
Static dashboards and traditional data analysis couldn’t track the rapidly changing conditions as managers felt the pain of out-of-stocks and missed opportunities.
Accurate and timely brand performance analysis is vital to a company’s success, now more than ever. In this article, we’ll break down how to achieve brand performance analysis that overcomes uncertainty and generates proactive, actionable insights.
What is Brand Performance Analysis?
Brand performance analysis ideally occurs when brand, category, manufacturer, and competitor data are analyzed to evaluate performance and generate insights that help businesses make decisions.
The insights from this analysis should help managers and teams understand how their organization performed compared to other brands in the category over short-, medium-, and long-term time periods. Brand analysis should also reveal drivers behind brand performance, or the “why,” and opportunities to gain market share.
Drivers contributing to brand performance could be anything from pricing to distribution. Likewise, drivers detracting from brand performance could range from the channel mix to product performance or package type.
With so many factors in play, managers need to understand which ones are impacting performance the most.
There are innumerable metric combinations that could explain the root cause behind change. At most companies, brand performance analysis is not yet sophisticated enough to handle this complex analysis.
In the next section, we’ll discuss some of the shortcomings of the current approach and how to improve your process.
Brand Performance Analysis Today
These innumerable metric combinations provide many options to consider, such as pricing, promotions, out-of-stocks, and distribution. Unfortunately, there’s not enough time to analyze them all.
The manual analysis process itself is often repetitive and inefficient. Analysts, working in Excel, must pull and prepare data on repeat, test their assumptions, and build a narrative with visualizations and insights.
When business teams have follow-up questions, the process repeats, leaving little time for strategizing and planning.
Even dashboards, which may seem user-friendly to managers, aren’t built for more complex questions. Dashboards simply display what happened (“Brand A increased sales by 9% in Q2”), but not why.
Analysts can help pinpoint some causes, but the manual process prevents them from testing everything. This simple answer doesn’t help brand and category managers discover what actually caused the increase in sales.
Data scientists may have more advanced tools to dive deeper into brand performance. However, it’s hard for them to share the insights due to the complexity of the ML (machine learning) algorithms and the lack of common business language within the programs.
These difficulties result in incomplete business intelligence that’s manual, non-standardized, biased, and time-constrained. As a result, brand and category managers struggle to gain actionable, timely insights that help them make proactive decisions.
Analysis is less of a competitive advantage and more of a struggle to get a basic understanding of performance.
There’s a better way to perform brand performance analysis.
Better Brand Performance Analysis
How can companies modernize their brand performance analysis?
Instead of shoehorning brand performance analysis into an outdated and ineffective process, organizations should think about how brand performance can actually lead analysis.
ML algorithms can be tailored to the specificity of the business, compared to dashboards that remain static. Imagine if brand and category managers can ask questions like “How is Brand A doing?” An answer is available in seconds–one that fully explores the important dimensions of the brand.
AI and machine learning can automate this analysis, testing every factor to ensure no stone goes unturned. Natural language processing lets business teams ask questions directly.
Managers won’t have to depend on data analysts and data scientists to interpret results or answer their pressing questions. Likewise, analysts and data scientists are freed from addressing the same question over and over again.
To round out a strong solution, natural language generation produces insights that business people can understand. The insights describe how many basis points a brand gained or lost, and why. It also includes context for how that performance has changed over time.
These are basic components of any good analytics solution, but moreover, they actually solve the challenges with brand performance analysis:
- Prioritizing thorough, contextual analysis that actually “understands” the components of brand performance.
- Putting insights directly into the hands of the decision-makers who most need them.
To incorporate this way of thinking into existing business processes, managers should first build a consistent understanding of the brand performance use case, start to examine the existing process, and solidify the type of information the company requires.
Managers should ask themselves:
- What problem are we trying to solve?
- As mentioned, brand performance analysis should help brand and category managers make better decisions and drive growth.
- What type of insights do we need?
- Identify insights that will move the needle on decisions: key drivers, opportunities, etc. These insights are the outcomes that should lead the analysis process.
- What’s the existing process?
- Identify top analysts and see what they’re doing well. These ideal workflows can serve as the basis for automation, to be replicated and distributed to the business via natural language queries.
- What data do we need?
- Only now is it important to consider the data. Many companies make the mistake of adding a dashboard on top of data, but this leads to the cumbersome process described previously.
- Rather, quality brand performance analysis requires that the data suit the use case–meaning data need not be perfect before organizations can reap the benefits of AI and ML.
Once these questions are answered, brand and category managers can look into automating the analysis. This process will further ease the strain on business teams and data teams.
Automating analysis provides companies with a competitive advantage. In the following section, let’s discuss the specifics of automated brand performance analysis.
The Competitive Advantage of AI and ML on Brand Analysis
Incorporating automation into companies’ existing data analysis processes saves organizations time and money by providing actionable answers and unbiased insights. Additionally, it’s becoming more critical for CPGs to invest now so as not to get left behind.
According to Forbes, “Marketers use AI-based demand sensing to better predict unique buying patterns across geographic regions and alleviate stock-outs and back-orders…Having this insight alone can save the retail industry up to $50B a year in obsoleted inventory.”
Since retail marketers are using AI technology, CPGs must incorporate it into their business practices as well. This ensures their knowledge of shelf presence, market share, and customer shopping behavior is current.
Instant insights enable better retailer partnerships and the kind of proactive decision-making that helps CPGs move faster than their competitors.
The other advantage of AI is that data insights are no longer at risk of biased processes or time constraints.
Brand and category managers and business teams now have access to insights presented in plain language and clear visualizations. They no longer need to ask data analysts and data scientists follow-up questions to understand what’s truly going on.
Take National Beverage, one of the largest soft drink companies in the United States. The data and analytics team knew its manual process hindered its ability to gain actionable insights.
The company turned to automated analysis through AnswerRocket’s AI-powered analytics software. Using AnswerRocket, National Beverage was empowered to create a culture of data-driven-decision making, which transformed its existing data processes. Learn more with the National Beverage Case Study.
Ultimately, brand and category managers should consider automation for brand performance analysis to streamline existing business practices.
Ready to transform your brand performance analysis? Talk to our team.