Content uploaded by Tim Ambler
Author content
All content in this area was uploaded by Tim Ambler on Sep 17, 2015
Content may be subject to copyright.
Marketing Review St. Gallen 2I2008 Marketingeffizienz
Eine Zeitschrift aus dem Gabler Verlag 70610
Marketingezienz
Die neue – Marketingfachzeitschrift für Theorie und Praxis | www.marketingreview.ch
2-2008
How Important is Marketing Efficiency? + Marketingerfolg messen: Optimale Kennzahlensysteme + Der Einfluss von Marketing
Assets auf den Shareholder Value + Mehr Marketingeffizienz mit Kundendaten – Das Beispiel Migros + Marketing und Sales
Performance Measurement – Das Beispiel Dell + Chancen und Gefahren von Werbetests aus Sicht der Praxis + Messung der
Werbeeffizienz – Ein zweistufiger DEA-Ansatz + Wie Kundenintegration effizient gelingt + Wirksame Unterstützung: Strategische
Kontrolle mit dem Analytic Hierarchy Process + Wertorientiertes Dialogmarketing: Kampagnencontrolling auf dem Prüfstand
4
Grundlagen & Evaluierung
Marketing Review St. Gallen 2-2008
The best practice, key indicator of marketing performance is widely claimed to be Return on Investment (ROI).
This paper takes issue both with ROI and the relative level of attention given to marketing eciency. It argues that
achieving integrated marketing and corporate goals is far more important, namely marketing eectiveness. The aim of
marketing is the sourcing and harvesting of cash ow the cash ow which is the lifeblood of the business. The intangi-
ble asset, brand equity, is the store of created demand not yet converted into sales and needs to be taken into account.
Thus marketing diers from other types of expenditure and should be treated accordingly.
T
he increasing demand for marketing ac-
countability is recognised but does this
equate to measuring only marketing ecien-
cy? Marketing is responsible for achieving its
required goals as economically as possible but
even within the context of marketing ecien-
cy, should Return on Investment (ROI) be the
dominant objective or performance measure?
is article examines some assumptions be-
hind marketing eciency, usually quantied
as ROI, and challenges their importance. In
working with marketers over the last ten years,
many have told me that they know ROI is a bad
metric but it is a “given” in their companies.
“Better” they tell me “to frame budget requests
that go with the ow.” Are they right?
This article begins with the common
ground. Other things being equal, we would
all like the same return for less expenditure
or greater returns for the same expenditure.
In reality, other things are rarely equal and
hard choices have to be made. Increasing
marketing expenditure hardly ever increas-
es ROI so we can discount that approach for
the moment and review the more common
issue, namely the extent to which market-
ing costs should be cut to increase ecien-
cy. e second section steps back to exam-
ine what marketing is supposed to contrib-
ute to the organisation. e third section
lays out some of the problems with market-
ing eciency metrics and ROI in particu-
lar. e article concludes with practical ad-
vice on the measures rms should adopt for
their three main purposes:
How Important is
Marketing Eciency?
T i m A m b ler
”Before we can measure
marketing, we have to know
what it is supposed to do.“
How Important is Marketing Eciency?
5Marketing Review St. Gallen 2-2008
1. Assessing marketing (or corporate)
performance to date.
2. Planning, i.e. deciding what actions
to take now.
3. Linking planning and past performance,
i.e. creating the business model.
Common Ground
Most practitioners can testify to marketing
expenditures which proved entirely unpro-
ductive, promotional material delivered late
for example. Some of these might have been
identied as unproductive ahead of time with
better analysis, and some seemed like good
ideas when they were conceived and only ap-
peared unproductive with hindsight.
CMOs and CFOs use their experience to
identify and eliminate waste from the plans
presented to them for approval. If the squeeze
is on, they compare the expected returns from
the various expenditures in the marketing
budget. Of course, that is easier said than
done since one needs to allow for synergies
between the activities, competitive reactions
and the fact that tomorrow is not yesterday.
Let us take those one at a time.
Marketers have long known that some ac-
tivities work better together than either does
separately. It makes sense, for example, to
have in-store promotion at the same time as
an advertising campaign. Reminding cus-
tomers at point of sale makes the ads work
better and vice versa. It is not always pos-
sible to identify their separate contributions
and, even if it is, it would not be sensible to
run one without the other.
On the second point, the most frequent
“loss leader” amongst marketing activities is
price promotion. Few marketers nd them
attractive but when the alternative is to be
delisted by a major retailer, or merchandised
on the lowest shelf, the promotion may seem
preferable. Where the retailer passes on the
full, or even more, price advantage, sales can
perk up well. Depending where one takes the
baseline to be, there may even be incremen-
tal prots. Nevertheless, price promotions are
typically unattractive and brand owners con-
tinue with them only because the alternative
is worse.
is is the Prisoner’s Dilemma: not accept-
ing the promotional slot means the retailer
may allocate it to a competitor and the con-
sequences would be worse than accepting it.
First prize would be to agree with competitors
not to accept these slots, or not on the retailer’s
terms, and stick to that. Even if such an agree-
ment would be feasible, it would be illegal.
On the third point, we cannot assume that
the past is a reliable guide to the future. e
most protable customer last year may be out
of business next year. e imaginative and
fun promotion one year is unlikely to inspire
the same enthusiasm in the following year.
Liverpool’s sunshine last summer may be out-
done by London’s sunshine next year.
Experienced CMOs and CFOs jointly rec-
ognise practical problems of these kinds.
Their world is neither simple nor perfect
and, to some extent, they can estimate their
way around the diculties. at done, the
CFO is likely still to be pressing for ROI g-
ures for each element of the marketing mix
for comparison with alternative expenditures
elsewhere. If a new bottling line has a high-
er ROI than a new advertising campaign, he
will recommend the bottling line.
is is where we leave the common ground
and ask whether we are comparing like with
like. Before we can measure marketing, we
have to know what it is supposed to do.
What is Marketing for?
Ambler (2003) examines three types of mar-
keting denition: nancial, functional and
corporate. e nancial view denes mar-
keting in terms of expenditure which is most-
ly advertising, promotion and research costs.
When academic articles discuss the return on
marketing, that is usually what they mean, but
marketing is, or should be much more than
nancial expenditure. e traditional 4P ap-
proach (McCarthy 1996) refers to product,
price and place (distribution) as well as pro-
motion. A marketer might be doing a ne job
rening the product, the price, the brand ex-
perience and the distribution, via sales, with-
out spending any money at all. To some ex-
tent, discussions about budgets and cutbacks
(eciency) can distract the marketer from
more important aspects of the job (achieving
the vital cash ow for the company).
e functional denition of marketing re-
fers to what the marketing department, or
specialist marketers, do. at has the advan-
tage of including all four Ps but is still less than
the full concept. Furthermore, most rms do
not have a marketing department at all. is
denition may lead them to conclude they
do no marketing but that is far from the case.
Every rm has products, goods and/or serv-
ices, to sell and customers to nd and nur-
ture. To maximise sales they must not on-
ly try to get customers to accept the products
(“push”) but also create demand so that cus-
tomers want to buy (“pull”).
So the third, corporate, view of marketing
is the sourcing and harvesting of net cash ow
by the rm as a whole. is includes call cen-
tres and other aspects of the customer’s brand
experience.
Although, ultimately, marketing perform-
ance can be measured by net incremental
cash ow, the problem lies with the word “ul-
timately” because we never reach it. At any
point of time we have the cash gathered in
and immediately due, plus the expectation of
future revenue due to the demand that has
been created but not yet realised. is is rep-
resented by the intangible asset usually called
“brand equity”. We can forecast future cash
ows but we cannot measure them. Brand
[equity] valuations are generally based on dis-
counting the expected future cash ows.
Marketing expenditure matters to the ex-
tent that it both reduces and increases net
present value but it should not be the prima-
ry consideration.
Why ROI is a flawed Metric
ROI has been used as a control measure since
it was introduced into DuPont by Donaldson
Brown in 1912 (Howell 2006). Ambler and
Roberts (2006) discuss the problems with
ROI in some detail. e main objections to
its use as a performance indicator or objec-
tive are arithmetical, inconsistency with cor-
porate objectives and the exclusion of the cru-
cial marketing asset, brand equity. Let us take
these three.
ROI is a ratio. In order words, the revenue
is divid ed by the marketing expenditure as
distinct from subtracting expenditure from
revenue. In consequence, due to the law of di-
minishing returns, ROI peaks at a lower lev-
el of expenditure than does prot or net cash
6
Grundlagen & Evaluierung
Marketing Review St. Gallen 2-2008
ow (see Figure 1). Here, the maximum prof-
it is $10M and is delivered by a $3M market-
ing expenditure (point B). However, a prot
outcome of $8M is delivered by a $2M mar-
keting expenditure (point A). e ROI at
point A is 400% versus 333% at point B – but
which is better? Real life choices are more
complex (e.g. what else could be done with
the money?) and the ROI arithmetic can be
made more sophisticated but in all cases the
point of maximum ROI sub-delivers on the
corporate cash ow maximisation objective.
In other words, the arithmetic of using ROI
in place of the corporate goal means that mar-
keting expenditure is sub-optimised. An ex-
penditure of almost nil, which produced a
signicant return of any size, would have an
ROI of almost innity and it would probably
be a silly marketing plan.
Although, as noted above, ROI calculations
can be made more sophisticated, in practice
they are short-term and ignore the longer-
term implications, i.e. brand equity. For ex-
ample, some promotions may give a good re-
turn in the short term but may damage the
brand over time.
Finally, the intangible marketing asset
needs to be taken into account. In any nan-
cial period, performance is given by the net
cash ow or net prot, for the period plus
the increase in brand equity. Brand equity
stands proxy for the expected future prots
or net cash ows. If we had perfect future
knowledge and could separate out the conse-
quences of actions to date from future mar-
keting activities, we could value brand equi-
ty at the end of each nancial period and use
that to estimate the performance in purely -
nancial terms. It would be valid but it can-
not be done so we have to rely on non-nan-
cial indicators instead. I come back to that in
the next section but the point to note here
is that ROI calculations ignore brand equi-
ty altogether.
At this stage in the argument marketers
typically become uncomfortable and claim
that when they say “ROI”, they mean some-
thing dierent, such as productivity in gener-
al. Misuse of terms is no defence.
So How Should We Measure Marketing
Performance?
We should begin with how the customer sees
the rm and its products, e.g. customer satis-
faction, and what improvements, for custom-
er and rm, could ow from that. Financial
arrangements apart, cash comes from cus-
tomers and marketing success lies in creat-
ing the demand by customers to spend their
money with the rm. As discussed earlier,
marketing is best understood as a corporate-
wide matter, some part of which may be al-
located to the specialist marketers if the rm
employs them. Marketing does two things:
create demand (brand equity) and release
the demand into sales and consumption. To
measure marketing performance, therefore,
we need a set of metrics to measure brand
equity, analyse sales and track consumption
(satisfaction). And we need to do that rel-
ative to the competitor´s performance with
their own, and our, customers.
“Brand equity” is the intangible asset which
represents the longer-term cash ow, i.e. be-
yond the present period. It includes measures
like satisfaction, perceived quality and inten-
tion to re-purchase.
Having established the corporate and mar-
keter goal metrics, we next need to ask if, on
the basis of prior experience, changes of mix
and expenditure reliably and measurably
change performance or whether the links in
the business model are more subtle. In oth-
er words, if no direct links exist from inputs
(from marketing expenditure and activities)
to outputs (changes in the goal metrics), we
need to look for intermediate metrics, such
as customer satisfaction, which directly link
with the inputs on the one hand and with
outputs on the other. In other words, inputs
correlate with intermediate metrics which in
turn correlate with outputs.
is of course is the business model and
it needs to have as many links as it needs to
connect inputs with outputs. All this can be
established either formally with quantitative
analysis of past data or informally with senior
executives sharing their experience or, pref-
erably, both.
e nal set of metrics is conventionally
now displayed, on paper or screen, as a “mar-
keting dashboard”. is is a new and exciting
development, mostly in the USA, and should
not be confused with the Balanced Scorecard
which is primarily internal with only limit-
ed reporting of the market, customers’ views,
end consumers and competitors. e two
concepts can, though, be complementary.
These intermediary variables can also
be seen as measures of brand equity. ey
may not be nancial metrics but if they re-
liably link marketing activities and expendi-
tures with the bottom line at a later date, then
they are, in combination, the proxy for future
net cash ows that we have been looking for.
ey are also the metrics that should appear
on the dashboard top management uses to
drive the business.
Conclusion
Marketing performance should be assessed
on the basis of what it contributes to the rm’s
goals, usually shareholder value, net cash ow
and/or prots. In particular, marketing out-
$M
Fig. 1 Relationship Between Sales and Prot
Source: Tim Ambler et al. 2004
0 1 2 3 4 5 6 7 8
A
B
Profit
Sales
10
Profit
Marketing Expenditure $M
Sales
How Important is Marketing Eciency?
7Marketing Review St. Gallen 2-2008
comes should be compared with the relevant
corporate goals, namely those concerning
customers, end consumers, brands and com-
petition. Marketing eectiveness is deter-
mined by the achievement of those goals; e-
ciency is only interesting within that context.
Yes, we should ask if eectiveness could have
been achieved at less cost but even for that
purpose ROI is not a valid measure. ere-
fore eciency is indeed important and less
productive activities should be dropped in fa-
vour of the more productive, but here too we
need to consider the longer term as well as the
immediate. at is why brand equity must be
included in order to take account of longer-
term marketing results.
Marketing metrics are vital in determin-
ing what actions to take (planning), assess-
ing the resulting performance and developing
the model which senior management should
understand and share in order to develop the
business.
Marketing is not an expense like any other:
it is more important than that. It is the source
of the rm’s life blood, its cash ow. Any rm
needs to worry about its health more than it
worries about the cost of the next meal.
References
Ambler, T. (2003): Marketing and the Bottom
Line, 2
nd
edition, London.
Ambler, T./Braeutigam, S./Stins, J./Rose, S./Swith-
enby, S. (2004), “Salience and Choice: Neural
Correlates of Shopping Decisions, in: Psychol-
ogy & Marketing, 21, 4, pp. 247-261.
Ambler, T./Roberts, J. H. (2006): Beware the Silver
Metric: Marketing Performance Measurement
Has to Be Multidimensional, Marketing Sci-
ence Institute, Report #06-113.
Howell, R. A. (2006): e CFO: From Controller
to Global Strategic Partner, in: Financial Exec-
utive, 22, 3, pp. 20-25.
McCarthy, E. J. (1996): Basic Marketing: A Mana-
gerial Approach, 12
th
edition, Irwin, Home-
wood, Il.
Tim Ambler
Senior Fellow at London Business School,
London
E-Mail: tambler@london.edu
Author
