London Economics (2012)
The Role Of Packaging Imagery On Consumer Preferences For Experience Goods was a study comissioned by London Economics for Phillip Morris International.
They questioned 3,000 people to determine the relative effect of different 'product signals' (e.g. brand names, packaging imagery, price) across a range of products (e.g. cigarettes, water, crisps, toothpaste.) The experiment was specifically designed to determine the effect of the removal of the product signals on preferences for premium, mid-range and low-range brands.
The baseline for comparison included all signals, and the experiment produced systematic tests with 5 different scenarios which removed product signals. Each round of the experiment for each participant involved them ranking the displayed products in order of preference from 1 (most preferred) to 6 or 7 (least preferred.)
The brands and products
Two brands each of 'full-flavour,' 'medium-flavour,' 'low-flavour' (10mg, 5-9mg, 1-5mg) and menthol cigarettes were chosen, across low-range, mid-range and premium brands (2x4x3 = 24 brands in total.) The brands chosen had the largest market share in the UK.
Six brands each of bottled water, beer, crisps, ice cream, toothpaste and chocolate were also chosen for the experiment, two each of low-range, mid-range and premium (6x6 = 36 brands in total.) Low-range brands included at least one supermarket-own brand in each product group.
Participants were not provided any information on whether the brands were considered low-range, mid-range or premium.
Six scenarios were presented. Products were presented with all, some or one product signals present:
- Name, Branding, Product information, Price, Advertising/Market share (Baseline)
- Name, Branding, Product information, Price (Removal of Advertising/Market share information)
- Name, Product information, Price (Removal of Advertising/Market share information, plain packaging)
- Name, Price (Plain packs with just price)
- Name, Brand (Branded packs, no price or information)
- Price (Just the price)
1,000 non-smokers and 2,000 smokers (no minimum consumption level, but the level was recorded) were interviewed online during July and August 2011.
All participants had to have purchased at least five of the six non-tobacco products in the previous three months.
Non-smokers were presented with all six non-tobacco products, smokers were presented with five (random) non-smoking products and the tobacco products.
Each participant went through 6 rounds, featuring one of each product and scenario. e.g. they'd have scenario 1 on beer, scenario 2 on ice-cream, scenario 3 on toothpaste...
Smokers were asked their preferred brand, and if their brand was one of the 24 present they were presented with six products in the tobbacco round. If it was not present, then seven products (to include their brand) were presented. Their selection of preferred brand also skewed the selection of brands presented to match the smoker's preferred flavour category (full/medium/low/menthol.)
- The removal of the advertising or market share signal has only a small effect on consumer preferences (Scenario 2)
- The removal of brand imagery from packaging shifts consumer preferences away from premium brand products towards low-range cheaper brands (Scenarios 3, 4)
- When brand imagery and brand name are the only information signals in the market, consumer preferences tend to shift towards premium range brands. (Scenario 5)
- When price is the only signal in the market consumer preferences shift towards low-range cheaper brands. (Scenario 6)