The Client is a worldwide apparel company with a stable of over 30 brands. After divesting of one of its old legacy brands, the company wanted to implement a new growth strategy via acquisition. They approached Management Science Associates, Inc. (MSA) seeking a robust solution for identifying what types of brands and businesses would have the greatest future growth potential for the apparel industry.
MSA’s Market Architecture™ solution was the perfect approach to clarify the current state of the apparel industry, as well as provide a glimpse as to where the best growth opportunities are within the category moving forward.
Instead of looking at repeat purchases as a way to compute brand loyalty, Market Architecture looks at brand purchases along key consumer dimensions to estimate what this solution calls ‘choice elasticity’.
- Market Structure – How do consumers choose products within a category?
- Brand Choice Drivers – What factors influence consumers to buy certain brands?
- Brand-to-Brand Competition – What is the nature of competition between brands? Why do consumers buy competitors?
- Profitability – How to leverage brand strengths to grow sales and profits?
While consumer survey data is usually the main data input for Market Architecture, in this case, MSA worked with ‘purchase occasion’ data from a syndicated consumer panel for the apparel industry to which the Client subscribes.
The study looked at all of the Client’s current retail customers, as well as other broader retailer classifications and channels. The study also covered almost all major retail apparel lines.
The output from the study provided the Client with a clear and consistent ‘three-tier choice hierarchy’ for women and men, with just a few exceptions. In addition, MSA was able to illustrate a ‘strength of appeal’ table for the Client showing which retailers, apparel categories and brand types had the greatest appeal to the least appeal. This table was crucial to the client for assessing the profit potential of target acquisitions.
As a direct result of the study, the Client soon afterwards made two major brand acquisitions, investing nearly $1 billion.
Direct Marketing Program Management
The Client is an American consumer packed goods company. They asked Management Science Associates, Inc. (MSA) to take over their opt-in consumer database and completely manage a Direct Marketing (DM) program for them.
Their database was already large and consisted of consumers who wanted to receive communications and promotional offers from the company. The Client wanted MSA to manage further growth of their database and have certain consumers regularly receive specific coupons, surveys, and info on sweepstakes and giveaways, via DM mailers and email (eDM).
MSA created its DM Program Management solution initially for this client. MSA’s solution consists of six (6) major components and processes:
- An opt-in consumer database of users and competitive product users. In this case, MSA took over an existing database and managed its ongoing maintenance. Consumers were recruited via either company-sponsored websites, or a call center that received requests from consumers who wanted to be included in the program.
- MSA’s proprietary NaviGATE™ technology. MSA created a custom, NaviGATE user interface for the Client to access the database and be able to setup consumer target criteria for DM and eDM campaigns.
- Database generated mailing lists for DM mailers and email lists for eDM. MSA took these lists and coordinated with a third party vendor to have 10 DM mailers and 5 eDM campaigns sent out every month.
- Periodic data extracts and reports generated from the database. These outputs were given to key Client stakeholders and to some third party vendors.
- Advance DM analytics. MSA used some of the data extracts to develop custom Response Models, Loyalty Models and Vulnerability Models, so as to continually improve and optimize the DM program’s campaigns.
- Mobile Coupon app for CPG. Consumers can be served coupons once they enter certain retail locations. They are able to redeem offers by displaying coupons from their smartphone at checkout.
Within a couple of years, the Client’s opt-in database grew to over 12 million consumers. As far as coupons delivered via the DM and eDM campaigns, redemption rates were much higher (6-16%) than for those delivered via traditional FSIs (<1%). The Client saw sales lifts of 0.6 to 1.9 units per week across all markets depending on how generous the offer.
Advertising Integration Analysis (AIA)
The Client is an American beer brewing company. They approached Management Science Associates, Inc. (MSA) seeking to better understand how advertising leads to increased brand loyalty and sales by linking various measures of advertising quality and effectiveness to key sales indicators for their flagship light beer.
This data heavy project involved integrating multiple data sources; ad tracking, behavioral tracking, copy tests, GRPs, PR and marketing mix coefficients, across brands, ethnic groups, age groups, gender and geographic regions.
- Create a large dataset by integrating ad tracking and behavioral tracking data based on age groups, gender, regions and ethnicity. Then tabulate and graph all of the variables to get an overall understanding of data trends.
- Create 2-dimensional maps to visualize which ad tracking and behavioral tracking variables are key drivers of ‘brand attachment’ vs. ‘behavioral motivations’ and how they relate to each other. Create 2-dimensional maps for the category, key brands, and for ethnic groups, age groups, gender and geographic regions for the subject brand.
- Use Factor Analysis to consolidate the data and help manage the correlation between similar variables.
- Use Structural Equation Modeling (SEM) to quantify the relationships between measures, such as awareness and loyalty, and to see how they vary across demographic groups and brands.
- Create a second dataset integrating the ad tracking data with GRPs, copy tests, marketing mix coefficients and PR data. Use this secondary data set to model the impact of TV advertising on awareness via Regression Analysis.
The study identified nine key measures for the Client’s brand that are influenced by a combination of both advertising and product experience. MSA calls these “bridge measures”. More importantly, MSA was able to provide the Client with absolute clarity on the intricate web of interrelationships between these nine affinity (brand image) and involvement (repeat purchase) bridge measures that comprises the brand’s purchase funnel.
As a direct result of the study, the Client was able to refocus brand monitoring on just these nine bridge measures. Whenever they saw any significant shifts in these key measures, they knew they had to look into making changes to their communications and/or other marketing levers. Not only did this immensely simplify their brand monitoring activities, it also provided the Client with a tactical marketing early indicator mechanism for their most important brand.
Baseline Sales Analysis
The Client is a leading manufacturer of cosmetics, skincare and haircare products. They approached Management Science Associates, Inc. (MSA) wanting us to analyze the performance of their trade promotions. One of their reasons for coming to MSA was that they were specifically not satisfied with the base vs. incremental sales volume estimates they were receiving from their syndicated sales data provider. These baseline sales estimates were often higher than total sales volume, resulting in unrealistic, negative incremental sales estimates.
Members of their Trade Promotions Team felt they were wasting a lot of time on trying to come up with alternative calculations for better estimates of base sales volume. They knew more accurate baseline vs. incremental sales volume estimates would result in better ROI calculations of their trade promotion investments.
MSA’s goal was to provide the Client with a consistent, accurate approach for measuring sales volume driven by trade promotions (i.e., incremental), while excluding brand marketing and temporal factors (i.e., base).
MSA tested and compared the results of several modeling methodologies to determine which approach would be best for the three years’ worth of weekly data available for the Client.
In the end, MSA used a model-based approach that decomposed sales into short-term incremental volume vs. base volume driven by loyalty/momentum, seasonality and marketing factors. For the modeled baseline estimates, MSA also considered trend and cyclical patterns, as well as factors related to distribution and competition. The models produced estimates by sub-brand and account.
MSA shared the modeled results with the client who then compared them to a syndicated sales data provider’s estimates. The client concluded that MSA’s baseline estimates were more consistent over time and more intuitive.
The modeling solution MSA designed is very flexible. The definitions of ‘baseline’ and ‘incremental’ are adjustable based on a Client’s industry and product category. For instance, FSIs can be included in incremental volume along with trade promotions.
The Client is a U.S. government Cabinet department. They approached Management Science Associates, Inc. (MSA) seeking a comprehensive examination of the long-term demand for steel within the U.S. The study needed to consider the expected demand for steel-using products and services, the impact of imports of steel-containing products, and the substitution of alternative materials by steel-using industries.
MSA’s comprehensive demand forecasting approach was the perfect solution to tackle this challenge. For demand forecasting, we look at deploying both Causal and Non-causal approaches:
a) Causal – Statistical models are used to isolate and determine the impact of key variables on demand to come up with forecasts.
b) Non-causal – Straightforward, time-series statistical techniques are used to forecast demand.
Our initial analysis showed that, historically, growth in GDP correlated very well with growth in steel demand. However, we also saw that more recently this correlation had broken down, indicating that GDP is not the only predictor of steel demand. Therefore, we concluded we could not apply a straightforward Non-causal approach in this case.
As such, we focused on a Causal approach. For this study, U.S. demand for steel was defined by the simple equation of shipments of steel products from mills, plus imports of steel, minus steel exports. Through extensive analysis, we determined that the primary drivers of steel demand that needed to be incorporated into our forecasting models were GDP, exchange rates, market sector-based demand for steel, the steel required in manufactured goods, and imports of high steel content goods.
We then conducted a series of quantitative analyses to estimate yearly demand for steel over a 10-year period. The estimates assumed both a cyclical outlook for the overall economy, one that assumed two recessions over the 10-year period, as well as a non-cyclical outlook that assumed no recessions.
The study determined that the most likely scenarios indicated that domestic steel demand was expected to grow at an annual rate of between 1.2% and 1.7% over the 10-year period.
We also determined that while growth in steel demand would be driven by consumer demand for steel-containing goods, growth would be offset by continued growth in imports of high steel content goods, the continued growth in plastics as a replacement to steel, and increased usage of alternative, lighter weight high-strength steel in the auto industry.
Ticket Sales Forecasting
The Client is an internationally acclaimed symphony orchestra with over a 100-year history. They approached Management Science Associates, Inc. (MSA) seeking a reliable solution for estimating future ticket sales, especially subscription-based sales, to help better manage their internal budgeting process and program management. They had many years’ worth of historical data of concerts performed with which they wanted to see if MSA could leverage to compute relatively accurate ticket sales forecasts.
MSA took much of the available historical data and developed a series of custom, algorithmic and regression-based models to calculate ticket sales forecasts for each upcoming performance, both subscriber-based and single ticket sales.
We then embedded all of the models into Excel® and developed an interactive, user-friendly, scenario-building tool with which the Client was then able to input various upcoming concert details. The tool visualizes sales forecast results as table summaries as well as in interactive charts. The tool also tracks model performance over time to determine if any model adjustments are necessary.
This symphony orchestra has used MSA’s ticket sales forecasting solution successfully for over five years now. Predictions for annual sales have been within a +/- 2-4% range of historical actual sales. These scenario-based forecasts have been providing this Client with previously unseen clarity to their programming and have allowed for significant budget savings over the years.
Media Viewer Segmentation
The Client is an American global mass media and entertainment company. They asked Management Science Associates, Inc. (MSA) to conduct a custom segmentation analysis of their TV viewing audiences. They wanted unique and specific descriptors of key segments, as well as robust insights on those segments, to support advertising sales initiatives.
The goal was to create a segmentation framework based uniquely on just TV viewing behavior derived from respondent-level TV program viewing data. The segments needed to show rich behavioral and attitudinal insights, beyond just simple demographics.
- Narrow the thousands of panelists into just 250 “nodes” based on program viewing behavior using a set of custom developed, multi-stage, machine-learning algorithms.
- Take the nodes and classify them into 67 unique clusters, again based on program viewing, but using a separate set of custom developed metrics.
- Test the clusters using both the art and science of segmentation and ultimately classify all of the panelists into just 20 distinct viewer segments.
- Observe how the final viewer segments fall into four sub-groups that are predominately oriented to either Females, Males, All Kids or All Adults.
The study captured meaningful insights into consumers’ identities based solely on the TV programs they watch. It was determined that when they make their viewing choices they provide clues into their interests, life stage, household composition and socio-economic status.
The final viewer segments and their descriptive profiles were updated regularly, and the names given to the segments became internal language for the client and used when talking about TV viewers.
As a direct result of the study, not only did the Client use the insights provided at the segment level for upfront and other types of ad sales meetings, they were also used for internal and external channel/program promotional marketing and communications initiatives.
If you’re a direct-selling cosmetics company, you understand, at first blush, the importance of social networking. And if you’re one of the world’s largest direct-selling cosmetics companies, you make social media a top priority for engaging your customers and incentivizing sales reps.
This particular beauty company wanted to understand how their reps and customers use social media. They wanted to communicate more effectively with them. And they were specifically interested in analyzing Twitter.
Twitter is one of the largest and most popular social media sites in the world, and the beauty company had already built a strong following. They engaged 50,000 followers with contests, special offers, and beauty tips.
But they wanted to do more. They felt they could enhance the effectiveness of their Twitter feeds through a better understanding of their followers. They wanted to know more about their lifestyles. Their interests. Who else they were following.
They came to MSA for answers.
Using a relatively new and sophisticated product called Twitter Follower Segmentation, our Business Analysis (BA) division got to work, extracting data on all of the client’s Twitter followers. A key advantage of this technique is that Twitter provides data for free.
We looked for what ‘friends’ the followers followed, how many people followed the followers, and how often the followers Tweeted. Turns out you can learn a lot about a person’s interests and lifestyle by studying their Twitter habits.
- Beauty & Social Media Tools – These followers don’t Tweet a lot or have a lot of followers, but they over-index on the beauty company’s other Twitter accounts and for social media management tools.
- Fashion News & Comedy – These followers have average Twitter activity. Their primary interests are fashion industry news and an eclectic mix of adult humor and politics. This also suggests a more professional demographic.
- Cutting Edge Beauty – Followers in this segment are more active on Twitter, and are very focused on the newest fashion trends. They have a strong interest in trying new products and follow beauty sites with edgy, dramatic looks as well as sites focusing on samples of the latest products.
- Music & TV – These followers are big on Tweeting, and are distinguished by their interest in entertainment. They follow personalities from reality TV shows and pop music stars.
Our BA team provided the beauty company with an overview of the unique aspects of each segment. We also provided suggestions for targeting each segment.
From a strategic standpoint, our in-depth segmentation study allowed this major cosmetics company to extend the word-of-mouth element that has always been integral to their direct-sell across the vastly larger, farther-reaching world of social media.
Tactically, we were able to identify top candidates for “Get This Look” videos; reveal a surprising strong interest in dramatic, theatrical fashion looks; and suggest a need for bundled sample packs versus individual samples.
Overall, MSA enhanced the effectiveness of our client’s Twitter account by identifying key interests of the account’s followers and suggesting content tailored to those interests.