Marketing Science Based Shrinkflation

Marketing Science Based Shrinkflation

Over the weekend the AFR warned consumers to get used to the growing trend for FMCG manufacturers to reduce pack size but keep price at parity. The trend, a by-product of long-term downward price pressure being exerted on manufacturers, sees unit prices increase in an effort to preserve margins.

The article quotes several well-known examples – ranging from Arnott’s recent reduction of Tiny Teddy’s multipack sizes from 10 to 8, to Smith’s 2014 chip pack shrinkflation, to Magnum’s length snip from 117mm to 107mm.

If you’re an FMCG marketer considering a pack size reduction, with a non-commensurate or non-existent price change, conducting consumer research is a smart move.

However, you’ll need more than a simple ‘question & answer’ approach – we all know that homo economicus will tell us they don’t want a negative result. Whilst it’s worth getting an impulse read on the magnitude of possible consumer backlash (read – you’re probably aware of some epic high profile fails in this space which resulted in a reversion back to the previous pack size!), this alone is not enough and won’t go far in helping get empirical backing to the business case underpinning the matter.

Our first suggestion is to start with some basic fact finding. To what degree do consumers accurately understand your current pack size, price and unit price? To what extent do they perceive your brand and product to offer value for money? Information of this nature is simple and easy to obtain, and goes a long way to understanding the basic permission you have to embark on a pack size overhaul.

A marketing science experiment is also worth investing in. Whilst scanner panel data is the obvious data asset to leverage, this typically proves fruitless because it only contains pack sizes and price points that have already been on shelf historically.

An alternative is volumetric choice modelling combined with so-called Monte Carlo simulation. Choice modelling is a research technique that allows us to answer ‘what if’ type questions about future market scenarios – scenarios that have never before existed. Model results calibrated against known MAT data allow us to predict the share, volume and revenue impact of all the pack size / price permutations being considered by the manufacturer. This can then be supercharged when combined with COGs, to produce profit forecasts of numerous future world scenarios.

Because choice modelling is based on ‘point in time’ cross-sectional data, it has a natural limitation when it comes to understanding how pack size and regular price scenarios impact long term volume – given that so much grocery volume today is driven by a range of promotional mechanics that vary week to week. To this end, Monte Carlo simulation can augment choice modelling results to provide a powerful longer term predictive indicator of in-market response to all elements of the in-store mix.

Studies of this nature can readily be completed in less than 6 weeks, and the value of the investment relative to downsides of failure make it well and truly worth doing.

If you’d like to know more about how marketing science can give you confidence in your business decisions, drop us a line. We’ll be happy to share some case studies, and have a confidential chat about the problem you’re tackling.


要查看或添加评论,请登录

社区洞察

其他会员也浏览了