In the final stages of packaging development, tests are conducted using shelves
modelled on a few major retail chains, to validate the effectiveness of the new packaging in terms of estimated increase in the
proportion of buyers and the sales volume. An increase in these metrics would confirm that the new packaging is effective at
these chains.
Packaging validation tests described here assume that only packaging was changed,
and all gains in sales are solely attributed to packaging. As such, these tests are only suitable
for established brands, and not for new brands or new product variants.
The analysis is broken down by current buyers and non-buyers, to provide an
assessment of the risk associated with losing current buyers as well as the potential of attracting
new or lapsed buyers.
Controlled Store Test
Controlled Store Test (CST) is an in-store assessment of the potential of marketing
initiatives such as packaging, in terms of gains in market share and sales.
CSTs require two panels of stores, control stores and test stores, matched on a range of criteria including
store size, category turnover, product range, and shopper demographics. The new packaging is introduced into the test stores,
whereas the control stores are insulated from these changes.
Sales, price, display, promotions, stock-outs and other in-store causal factors are tracked over weeks before,
during and after the introduction of the new packaging.
Comparison, across the panels, of the lift in sales over the period that the new packaging was introduced, provides
a good assessment of its impact in sales volume and value.
While it can provide accurate estimates of the lift in sales due to the new packaging in a real-life setting, a
CST needs to be very carefully managed, consuming time and resources. Moreover, it requires the co-operation of one or more retail
chains, which is not always easy to secure for CSTs. Importantly too, manufacturers need to produce the product contained in the test
packaging in quantities large enough to cater to the test stores in the real market, and it may not be viable to do so.
For these reasons it may often be more practical to conduct simulated shelf tests in a virtual controlled
location setting.
Simulated Shelf Test
Simulated shelf tests are computer-assisted personal interviews (CAPI) that use virtual
control shelves and test shelves. Respondents are asked, in a virtual setting, to purchase the product category from test and
control shelves, and the shelves are rotated to prevent bias.
The control and test shelves are modelled on a few of the major retail chains, and the respondents are the primary
shoppers at these chains.
The simulator tracks purchasing behaviour in terms of items examined, items purchased, the order they are purchased,
time taken to shop, and the amount spent.
The use of screen-based eye tracking, in a controlled location setting, greatly enhances the study by providing
valuable information on the respondents’ gaze plots, gaze duration, order of fixations, and heat maps.
If no eye tracking devices are deployed, respondents participate in a visual search test where they are asked to
find items on the shelf. This methodology is described in section
Shelf Impact — Visual Search Test.
The eye tracker/visual search test serves as an assessment of the findability of the packaging.
Respondents are then shown the test and control shelves side by side and asked to indicate
which one was easier to shop from.
A key metric for validation is the sales volume/value and share. As with CSTs, hypothesis testing is used to
analyse the difference in the sales between the control and test panels, and to test their significance. If the detectable
difference is expected to be small, larger samples will be required to achieve confidence levels at stipulated confidence
intervals.
