Retail Furniture Chains

Retail Furniture Case Study

Challenges

A furniture retailer has a POS system that tracks multiple brick-and-mortar stores in markets across the country, as well as e-commerce. There’s a wealth of data, but the marketing team can’t easily filter and synthesize that data to create a marketing strategy by chain, market and store. Instead, they’re sending the same direct mail, email and social messaging to every customer nationwide. Spend is up and ROI is down.

“Having worked on both sides of the Atlantic with data and database experts, I’ve found Bonnie to provide a breath of fresh air — outspoken, pragmatic, clear, and insightful in an industry full of jargon-speakers and shysters. She’s as honest as you’ll ever want to meet, will tell you if she can or cannot help, and adept at asking the ‘right’ questions to get you on the right path. Truly a trusted partner to have in 2010 and beyond.” Paul Woolf, Wild Onion Marketing

Solutions

One-size-fits-all has not been an effective marketing strategy since the advent of the personal computer and customer relationship management systems (CRMs). Modern marketers know:

  • who their customer is, using detailed customer profiles
  • what they’re buying on their first, second and third visits
  • when they’re likely to buy again
  • how all of that is changing over time

Massa & Company created a marketing analytics platform to find the answers to these questions within the client’s data and to translate it into a series of actionable, visualizations.

Process

Step 1: Use data to identify customer segments. We started with the client’s last 5 years of POS data for every customer, including e-commerce purchases. We cleaned records and appended demographic data: customer age, race/ethnicity, education level, presence of children, household income, social media achiever scores and more. This gave us a first look at customer segments that the retailer could use to develop a marketing plan.

Step 2: Build interactive, visual dashboards. Using this data, we built charts, maps and other visuals to show the marketing team what was happening by store, market, customer group and purchase category. The client’s marketing team gained 24/7 access to:

  • An interactive map overlaying purchase history, showing what each customer segment was buying by chain, market and store
  • Weekly sales by chain, market and/or store cumulatively compared to previous year
  • Revenue by SKU, by top purchase categories, by customer age, home value, etc.
  • Direct mail, email and social campaign results by age, ethnicity, etc.
  • Website use, email and social media metrics

The client could now see who their customer was and what they were buying, and therefore the marketing team could develop a strategy by customer segment. The dashboard was updated weekly with new customer and sales data, directly from the client’s POS system, website, social profiles and other marketing platforms.

Step 3: Build a predictive model. A predictive model is a marketing algorithm that uses existing data on purchase histories, location and customer profiles to predict customer buying habits. Predictive models inform marketing spend, campaign type and audiences to target.

The 5 years of customer data informed the initial model and showed how purchase histories changed over time. Massa used historical and new weekly POS data to build a predictive model that identified actionable trends in customer purchase behavior. For example:

  • Data shows a trend among 35-49-year-old couples in Ohio. Those who buy bedding on their first visit are most likely to buy bedroom furniture on their second, within 90 days. A preliminary email campaign shows that this group responds best to traditional décor.
  • A predictive model identifies similar customers in the marketing database. The marketing team sends a highly targeted campaign only to these customers who are ready to buy, displaying the pieces or room group they’re most likely to buy. The model is also used to upsell to customers who have not yet made that second purchase.
  • Throughout the campaigns, results are captured and used to modify the predictive model and inform the next steps in email, direct mail and digital marketing strategies.

Results

With careful data analysis and the predictive model, this client saved marketing dollars. Direct mail, email and digital were used like lasers to target only the customers likely to buy with messages and imagery they would find engaging – instead of a widespread net of wasted creative, postage, unsubscribes and unfollows. At the same time, the model increased conversion rates and sales. It sounds like the marketing dream – and it is!

Why Massa

Massa has worked with 7 furniture retail chains that realized a 30% to 50% increase in response to direct digital marketing over the traditional outlets used by each chain in-house. We have also helped develop direct mail acquisition programs that yield an average 1.5% response because we built the customer profiles by chain, market and store.

Even with a robust team, many in-house marketing departments simply do not have the analytic skill nor the tools to build and utilize a predictive model. (Nor is marketing analytics an assignment for IT, as they are usually consumed with mission-critical responsibilities for managing the company data systems.) By outsourcing your marketing analytics management to Massa, you save on staff costs and other overhead while getting the information the entire marketing team needs to succeed.


Call Bonnie Massa today (312.463.1050 x701) to talk about your direct and email marketing challenges and the tools we have to that can increase campaign response rates and revenue!