Insight, Analytics and the New Marketing
How much to spend in a recession?
Two weeks ago I wrote a post about setting the marketing budget. It described the often very basic rules marketers use to make multi million dollar investment decisions in marketing. It also had data from a recent McKinsey survey showing that although 45% of the companies were planning to reduce their budgets, 27% had no changes planned, 20% of companies were planning to increase their budget and 9% didn’t know. What can we learn from past recessions that can help us spend our marketing dollars better during this one?
Over the last months professional marketing literature has been dominated by material advising marketers on what to do during a recession. A lot of that material covered rules and recommendations about how marketing budgets should be set or adjusted during recessions. Reading through it I noticed very little new thinking. Most of the articles were either reruns of material that was written after past recessions or they are articles that use the same methodologies used in those articles on more recent data.
The most popular methodology used to demonstrate what happens when advertising budgets get cut during a recession compares companies that cut their advertising to those that held their spend stable or even increased it. Performance of these companies is then compared during the recession and in a period after that.
This methodology was 1st tried out in the 1920’s by Roland S Vaile. He published his results in the Harvard Business Review issue of April 1927. Vaile compared companies that maintained their advertising spend during the 1923 recession to those that cut their budgets. He found that the biggest sales increases were recorded by companies that advertised the most.
Buchen Advertising Inc. applied the same methodology to 4 recessions in the 40’s, 50’s and 60’s. They were able to show that companies who cut their budgets during the recession not only performed worse during those recession but also in the subsequent years of recovery.
In 1982 Cahners Publishing Company published the results of an analysis of the Profit Impact of Marketing Studies (PIMS) database which then held information for almost 2000 companies. They showed that companies who increased their budgets during a recession gained on average 1.5% in market share during those periods.
Similar research was done by Biel (1991), Hillier and Baxter (2001), Buck (2001). The most recent example was presented at an IPA conference in 2008. Malik consulting repeated the analysis Hillier and Baxter did in 2001 on more recent data. The conclusion was very similar to all of the studies mentioned so far. Companies who increase their budgets during a recession show increases in market share and return on capital employed (ROCE) in the years after the recession.
While the observations in all these studies are accurate, I often feel nervous about the recommendations derived from them. Most people claim these studies make a case for increasing investments during a recession to outmuscle weaker competitors and gain market share that can lead to a sustained advantage during the recovery and beyond. This implies causality and none of the studies above delivers proof of this. Do companies who spend more during a recession perform better in the long run because they increased their spend? Or are companies who perform well before, during and after a recession better placed to maintain or increase their spend during a recession? Some of the authors acknowledge this issue of potential reverse causality. Few offer an alternative approach.
I believe a more thorough approach is required to make the case for sustained marketing investment during tough economic times, especially if we want the arguments to stand up to the scrutiny of already skeptical CFO’s and CEO’s. Three things can help here :
1. Increase the use of econometric modeling
2. Take a holistic view of the marketing plan
3. Determine the long term effects of marketing
Increase the use of econometric modeling : Econometric modeling has been around for a long and it’s power in helping marketers understand what works and what doesn’t has been demonstrated over time. Today however it’s use is still fairly limited. Only 15% of the case studies submitted for the IPA awards use econometric modeling to identify the effects of campaigns. The most often quoted paper that uses econometric modeling to make a case for sustained investments throughout a recession is probably Cutting Adspend in Recession Delays Recovery by Dyson in 2008. The article uses econometric modeling to prove exactly what it’s title says. Dyson showed that “the increased spend required during the recovery just to get back to pre-recession sales levels within a year will have to be around 60 per cent higher than the amount saved by cutting the ad budget in the first place”.
Take a holistic view of the marketing plan : Let’s say you have build econometric models that give you a clear sight into what would happen if budgets are cut. One would still need to take into account media costs and consumption. Media costs often fall during a recession. Simon Broadbent (1999) gave a very interesting example which demonstrated the impact of cost savings due to falling media costs. This in one example leads him to recommend increasing spend during the recession of the early 90’s. There is also evidence that media consumption can change during a recession. People stay home and consume more media, which could increase the efficiency of media spend during tough times. Clearly an approach like this which uses modeling and takes into account the specific context of the brand, the category and the media landscape is much more thorough and easier to defend than the one size fits all golden rules described earlier.
Determine the long term effects of marketing : Budget cuts during a recession are often the result of a short term view. Advertising efforts rarely pay themselves back in the short term. As described in a previous post on the long term effects of advertising, 5 recent studies showed that every 1$ spent on advertising on average only generates 50c in payback in the short term (Dyson 2008). However, there is growing evidence that the long term effects could be 2-6 times the size of the short term effects. Broadbent (1999) called quantifying the long term effects of advertising the most important task facing advertising researchers. He said that a lot of work still needs to be done in the area and that this work shouldn’t just be done in tough times. The understanding of long term effects is incredibly important as it can impact shareholder value. According to Barwise (1999) it is a myth that financial markets only care about short term financial results. They tend to maximize long term shareholder value. It is therefore up to the marketing community to demonstrate that long term shareholder value is not only a function of short term financial performance but also of sustained longer term investment in the brand.
References
P Barwise, Advertising for Long term Shareholder Value, Admap Magazine October 1999
A Biel, The cost of cutbacks, Admap Magazine 1991
S Broadbent, Tough Times, Admap Magazine April 1999
S Buck, The true cost of cutting adspend, WARC, January 2001
P Dyson, Cutting adspend in a recession delays recovery, WARC Online, March 2008
P Field, Marketing in a downturn: Lessons from the past, Market Leader, Issue 42 2008
T Hillier & M Baxter, How to prevent a hangover, Market Leader, Issue 14, 2001
B Ryan, Advertising in a Recession, AAAA, Value of Advertising Committee, 1991
Comments