Use a dashboard when driving your marketing

Despite all evidence to the contrary, the belief that a single number can be used to assess marketing performance is persistent. Some say that top management can only handle a single number, or silver metric, so we must choose the least bad one. Others believe that ROI is so standard as not be to worth challenging. Others again claim modernity for customer concepts such as customer equity, customer lifetime value and Peppers and Rogers' new 'Return on Customer'. Yes, they are new and, yes, they have value, but these measures are not the silver metrics their promoters claim them to be.

It is hoped that a silver metric will justify marketing expenditure in much the same way that shareholder value measures the firm's performance as a whole. Marketing is seen as a slippery area with constantly moving goal posts. However badly a performance is judged by others, any CMO can find some metrics to show it was really a great success. 'Let's find one measure of success and stick with it' is the cry, and it is attractive.

This article takes a closer look at four silver metrics – three financial and one non-financial, and shows why they should not be used for marketing performance assessment. The metrics considered are listed below.

• Return on Investment (ROI), or ROMI (Return on Marketing Investment) or ROME (Return on Marketing Expenditure), targets sub-optimal profits and cash flow and does not allow for changes in brand equity.
• Discounted Cash Flow (DCF) is a single technique with many names, for example, 'Customer Equity', 'Net Present Value', 'Brand Valuation' and 'Customer Lifetime Value'. These all discount future cash flows, and maybe risk, back to a single present-day cash value. This technique can be useful in strategy and planning but the question here is whether a DCF metric is valid for performance assessment.
• Return on Customer (ROC) is 'a firm's current-period [net] cash flow from its customers plus any changes in the underlying customer equity, divided by the total customer equity at the beginning of the period'. This attractive idea links the short-term change in cash flow with a long-term indicator, i.e. the DCF of profits from customers. Unfortunately it does not do what it says on the tin.
• Net Advocates: Frederick Reichheld, Director Emeritus and Fellow at Bain & Company, has advanced the theory that the number of net promoters, i.e. the number of people who recommend your company minus the number of detractors, is the one number you need to grow. Does this concept stand up to analysis?

Once one accepts that no silver metric will do the job, the question becomes how many metrics are needed and which they should be. This article concludes with suggestions for selection and presentation on a market dashboard.

Return on investment (ROI)

Return on investment (ROI) is an appropriate measure for capital decisions where each investment is made once and results in cash flow in subsequent years. ROI is a fraction, namely the net return divided by ('on') the investment. There are at least six reasons why ROI should not be used for evaluating marketing performance.

1. Marketing expenditure is a continuing expenditure, not a one-off 'investment' in the original sense and not treated that way in company accounts.
2. ROI is a ratio of profit to expenditure, whereas other bottom-line performance measures consider profit or cash flow after subtracting expenditure.
3. Pursuit of ROI causes underperformance because the point of maximum ROI is reached at the point where returns start diminishing, not where they stop. In other words, profits keep growing beyond the point of maximum ROI.
4. Calculating ROI involves knowing what would have happened if the incremental expenditure had not taken place. These baselines are rarely available.
5. ROI has become an 'in term' for marketing productivity and used to describe any justification for marketing expenditure. The American Marketing Association White Paper on marketing accountability, for example, identified six 'ROI measures currently used', none of which, remarkably, was return divided by investment. The term 'ROI' has become meaningless.
6. ROI ignores longer-term effects whether represented by brand equity or any other proxy.

Exceptions exist and one can get around these objections. Some companies have gone to elaborate lengths to do so, but then 'ROI' is no longer ROI. No amount of pretending a cardboard image is a house turns it into a house. The totemic use of ROI does not keep accountants at bay nor show understanding of accountability – quite the reverse.

Discounted Cash Flow (DCF)

The suitability of any tool, be it a spade or a market metric, depends on its intended usage. A spade may be a good spade but it is not much use for raking gravel. When comparing alternative marketing plans, forecasting cash flows and risks, and discounting them back to net present value, is good practice. Note that the external variables such as interest rates or economic growth can be standardised across the alternative strategies to highlight the differences arising from the managerial variables, because the forecasts are all being done at the same point in time.

In essence, DCF techniques are fine for looking into the future (planning) because the raw data is indeed future cash flows. Using future cash flows to evaluate past performance is quite another matter.

There are at least five objections to using any DCF technique as a silver metric for performance assessment.

1. Doing DCF calculations at different points in time mixes up performance with variances in external variables, like the economy or interest rates or state of competition.
2. Using forecasts as benchmarks confounds the quality of performance with the quality of forecasting. The question is no longer 'How well have we done?' but 'How well did we estimate how we would do?'
3. DCF involves taking credit today for marketing activities in the future (which have neither happened nor are being evaluated). Given the affection of marketing for 'hockey stick' forecasts ('things will get a little worse before they get a lot better') and rapid job transition, including the benefits of future performance in present evaluation of past performance is not just illogical but unwise.
4. The future can be forecasted in various ways but cannot be measured.
5. Those generating forecasts that will later be used to assess their own performance will inevitably tend to reduce them if they expect to be around or raise them if they do not. The final forecasts are therefore the outcome of a political process, not an objective assessment of the future.

If a firm had perfect foresight, then cumulative improvement in DCF would be, with short-term cash flow, powerful metrics. Unfortunately, the future is not that certain. Nevertheless, at least one attempt, described below, has been made to combine ROI and DCF techniques for performance evaluation.

Return on Customer

Don Peppers and Martha Rogers claim that ROC is the key performance indicator because maximising ROC also maximises both current period and future profits. Their 2005 book about ROC became an instant best seller.

The Peppers and Rogers definition is 'ROC equals a firm's current-period cash flow from its customers plus any changes in the underlying customer equity, divided by the total customer equity at the beginning of the period.' Customer equity is, in line with current academic papers, the net present value of future cash flows from customers, and therefore has the DCF problems discussed in the previous section. In CMO Magazine (September 2004) Peppers and Rogers also defined ROC with a simple equation virtually identical to an ROI equation even though they emphasise that ROC is much better than ROI.

The Peppers and Rogers equation has been taken apart in a working paper available for free download (Ambler & Roberts 2005). The algebra boils down to something very different from the Peppers and Rogers claims. The top line of their formula turns out to be just the difference between the actual and forecast cash generated for the period under review, plus the difference between the two forecasts for the same outward years. Thus if the forecasts are accurate and consistent, the return is zero.

However good or bad the actual performance, ROC measures the quality of forecasting. One could argue that the forecasts are valid benchmarks of performance, but this is not correct. A forecast represents the expected performance. An objective forecast for a poor marketer will be lower than that for a good marketer in the same position. ROC will report good performance from the poor marketer even when it is actually worse.

In other words, this ROC formula does not measure return on the value of the marketing assets: it measures the excess of the return for the current period (compared to forecast) plus any increase in forecast, both taken as a ratio of customer equity. Thus it is not a measure of performance, but a measure of forecasting: the accuracy of the first-year forecast and the consistency of that for the later years. In short, ROC does not do what it says on the tin.

Net Advocates

In a 2003 Harvard Business Review article entitled 'The one number you need to grow', Frederick Reichheld argued that the number of people singing the praises of your company, or your brand, less the detractors, is the single indicator to watch. The top ranking question, which was apparently by far the most effective across all industries, was 'How likely are you to recommend X to a friend or colleague?' Whether this predicts actual advocacy behaviour is beside the point; if this silver metric varies with marketing activities and predicts profitability, then it is good enough. Reichheld claims that the only [my emphasis] path to profitable growth is for 'loyal customers to become, in effect, its marketing department'. This sheds new light on what marketing departments should be doing.

One must be impressed by any consultant practising what he preaches and Reichheld is certainly a strong net promoter of his own loyalty brand. The problem is that his thesis does not withstand analysis. In a letter to HBR, Neil Morgan and Rego Lupo used their own research to challenge Reichheld's theory. Their conclusions raised five objections, from Reichheld's confusing growth with profitability to the need for a multi-faceted approach. Sharing data is standard academic practice, but Neil Morgan and Lupo Rego were unable to obtain them from Reichheld for comparison.

Reichheld's response, in a further letter, fell back to arguing that the number of net promoters was a better metric than customer satisfaction. That may well be the case, as customer satisfaction has been shown to have tenuous correlation with profitability. Net promoters may well be a useful indicator but it is a long way from being a silver metric.

THE MARKET DASHBOARD

Claiming, as we must, that marketing is responsible for bringing in the cash, leads inevitability to the idea that marketing performance should be measured by cash, and if not cash today, then cash discounted back to today. That in turn leads to the search for a financial silver metric and we risk going round in circles. Three issues allow us to break out of that circuit.

Model Past Performance

The first issue is that we need to let the metrics do the predicting and not do it ourselves. In other words we need to see which of all the metrics we used in previous years predicted the performance this year, and how well. We can reliably measure the inputs (e.g. the marketing tactics and expenditures), the intermediate variables (e.g. intention to purchase, perceived relative quality, top box satisfaction, net advocates), and behavioural metrics (e.g. share of category requirements) over the past three years or so to the present time. Modelling those metrics against subsequent years' performance will also reliably give us estimates of their predictive values.

Contrast that with the often wish-fulfilling long-term cash forecasts involved in DCF techniques. The approach described above is empirically grounded and limited to predictions up to known outcomes; the second takes us into the wild blue yonder.

In this scenario, performance reporting still uses short-term cash flow, the common ground, but also the current brand-equity metrics, along with some indication of their reliability for predictive purposes. We report the facts, not speculation.

Devise a Company, not a Marketing, Dashboard

The second issue is the need to distinguish the role of the marketing department, which differs from company to company, from the marketing undertaken by the company as a whole. Marketers must gain acceptance that they are driving towards the corporate goals, and expect performance to be measured on that basis. We are looking at indicators on a market dashboard for top management as a team, not some separate marketing dashboard.

Take Note of Corporate Issues

This takes us to the third issue, which is the recognition that the metrics, or indicators on the dashboard, should be driven by corporate issues, namely goals, where the business now is in relation to those goals, the speed towards the goals, risk etc. The dashboard metaphor is exact; a company needs the equivalent of a speedometer, a fuel gauge and a miles-to-travel indicator. And it needs two levels of indicator: those that are visible all the time and those that switch on only when there is a problem needing attention, like the oil level.

Our research on US dashboard usage in 2001 and 2003 revealed that dashboard users were more able to assess the productivity of marketing expenditures, had less waste and better revenue increase, compared with non-users.

CONCLUSION

Managers need to integrate the key metrics used for performance evaluation with those used for planning. This means that one of the DCF metrics, such as customer lifetime value, customer equity or brand valuation, does deserve a place on the dashboard both to evaluate alterative plans and to track how circumstances drive expectations. DCF techniques help develop the business model by observing the way performance differs from plan.

The central point of this paper is that a firm needs a dashboard of key market metrics, not just one. Each metric has particular uses, strengths and weaknesses. No silver metric will do.

While DCF and some measure of loyalty, such as top box satisfaction or net advocates, do earn a place, ROI and Return on Customer do not. ROI is misleading and ROC tests the accuracy and consistency of forecasts, not performance.

Driving a business is not in principle different to driving a vehicle. You need a dashboard and only one pair of hands on the wheel. Marketers are there to help navigate the best way through the market in order to secure the best long-term cash flow. They should be judged by the milestones as they go by, not speculations and forecasts.

This article featured in Market Leader, Summer 2006.

REFERENCES

Ambler, T. & Roberts, J. (2005) 'Beware the silver metric: marketing performance measurement has to be multidimensional', London Business School Centre for Marketing Working Paper 050709, http://facultyresearch.london.edu/detwp.asp?id=7392

Clark, B., Abela, A. & Ambler, T. (2006) 'Using dashboards to align marketing and the organization', Marketing Management, forthcoming.

Morgan, N.A. & Rego, L.L. (2004) 'The one number you need to grow', Harvard Business Review, 82 (4 April), pp.134– 136.

Peppers, D. & Rogers, M. (2005) Return on Customer: creating maximum value from your scarcest resource, Doubleday, USA.

Reichheld, F.R. (2003) 'The one number you need to grow', Harvard Business Review, 81 (12 December), pp.46–54.

Reichheld, F.R. (2004) 'The one number you need to grow', Harvard Business Review, 82 (6 June), p.133.