Edward Baer Roberts.

Marketing and engineering strategies for winning R & D contracts online

. (page 1 of 2)
Online LibraryEdward Baer RobertsMarketing and engineering strategies for winning R & D contracts → online text (page 1 of 2)
Font size
QR-code for this ebook


mmimf^^'^'i^Wm



ALFRED P. SLOAN SCHOOL OF MANAGEMENT



Research Program on the Organization and Management
""■ of Research and Development

MARKETING AND ENGINEERING STRATEGIES
FOR WINNING R&D CONTRACTS



Edward B.^Roberts



luly, 1965



#128-65



MASSACHUSETTS

INSTITUTE OF TECHNOLOGY

50 MEMORIAL DRIVE

CAMBRIDGE, MASSACHUSETTS 02139



Research Program on the Orpanization and Management
of Research and Development

MARKETING AND ENGINEERING STRATEGIES
FOR WINNING R&D CONTRACTS

Edward 13. Roberts

July, 1965 //128-65



Presented at the
Fourth Annual Management Conference on Marketing in
the Defense Industries
May 27, 1965



Sponsored jointly by the Defense Marketing Group, Boston Chapter,
American Marketing Association, and the Bureau of Business Research,
Boston College



This study is one of a program of researches supported in part
by a grant from the National Aeronautics and rt)acc Administration.
The findings and views reported therein are those of the author and
do not necessarily reflect those of the sponsoring agency.



Marketing and Engineering Strategies for Winning R&D Contracts

Professor Edward B. Roberts
Associate Director
Research Program on the Organization and Management of
Research and Development
M, I„To Sloan School of Management



Dr. Kantrowltz, Ifether Joyce, Dr. Vaughn, ladies and gentlemen!

Today's topic, adapting for sixrvlval In the defense industries, has been
in recent years the subject of much private and public debate. One of the most
controversial documents in the continuing exchange of ideas is a widely circulated
publication of the locally-based consultants, Arthur D. Little, Inc.-'- Of their
conclusions, with which I strongly concior, a keynote is that for assuring sur-
vival we are going to have to focus more and more on problems of marketing for
research and development and managing research and development, 2 Thro\igh in-
creases in the funding and effectiveness of company and government sponsored re-
search and development, corporate and economic growth can be maintained.

Before describing some of my research results on the R&D contract award pro-
cess, I should like to tell you about a newly elected governor and his aide who
were touring the s-tate prison. As they walked through the corridors with the
warden, they heard noises off In the distance. Walking closer they realized that
the noises were the sounds of laughter. Then, as they got even closer, they heard
somebody call out, "Sixty-five I", and there was loud laughter. Someone else
hollered, "Thirty- two!", and there was loud le.ughter again. Then someone called
out, "Twenty- fivel", but there was dead silence. The governor and his aide turned
to the warden and said, "Whafs going on here? This is the state prison. Why all



ISuperscripts refer to entries in the annoted bibliography at the end of this paper.



the laughter?" And the warden explained, "Well, you know, governor. These men
have been here a number of years, and it's pretty dreary here. So they tell
stories to amuse each other. After a period of time it gets bo that they know
each other's stories. To get in more jokes dviring an evening they put numbers
to these stories. This way, they can just say the number, and then everybody
laughs." This sounded reasonable to the governor but he asked, "VOiat about that
guy who called out 'Twenty-five'7 There was dead silence." "Well," answered the
warden, "that was old Joe — he never could tell a good storyl"

Later that week the governor was addressing a banquet, and his aide was
sitting beside him. The govemor thought that his prison tour would provide an
arousing story for the people assembled at the banquet. He got up and started
talking about his trip with the warden, and he was coming up to the point of des-
cribing the laughter they had heard off in the distance. The governor suddenly
stopped short. Turning to his aide with a puzzled look, he asked, "What were
those numbers?"

In case you feel the same way at the end of this talk, I should like to
begin by telling you ray conclusions first, and hopefully you will remember these.
Then, if later you forget the numbers on which the conclusions are based, that
won't be so bad. Actually, I have only one main conclusion, but I want to develop
it for you in two different ways. The conclusion is simply that more than most
of you now believe, and more than your firms act as if they believe, the award
of research and development contracts is not a case primarily of company proposal
preparation followed by government proposal evalviation. Rather, the award of R&D
contracts is a case, much more, of person-to-person contact, principally technical
contact; person-to-person information exchange; and person-to-person development
of confidence and trust. I intend to draw this conclusion for you, and to show
you the evidences that lead to this conclusion, in two ways: first, by looking



at government and at the kinds of activities that our research demonstrates under-
lie the award of government contracts; and second, by looking at industry, and
examining the results of research studies that ve have conducted on companies
attempting to win government contracts « From both perspectives, the government
evidences and the industry data, we shall be led to this conclusion about the
essentiality of the informal person-to-person aspects of the research and devel-
opment marketing situation, in contrast with the lesser importance of the formal
proposal preparation-proposal evaluation aspects.

About four years ago, the M. loT. Sloan School of Management, with the strong
encouragement of Mro James Webb, Administrator of N.A.SoAo, determined that it was
of major importance that we become more actively involved in research and teaching
on the management of research and development*-^ As part of our research program,
I started in I962 a study of the R&D contracting system. The contracting system
regulates the principal portion of our spending in research and development in
the United States, and it seemed important, both in 1962 and now, to find out what
determines the award of R&D contracts, the terms of the contracts, the industry
response to the contracting system, and the like.

We started our research with a series of studies of government agencies and
their practices in awarding research and development contracts , Thus far^ although
we have examined only three large government procuring installations, we are already
tiring of finding the same results With numerous indications from letters, visits,
and discussions that the results in other places will not be very different, prob-
ably soon we shall draw this phase of data collection to a close.

The research began at one NASA installation where my research assistants and
I looked at a series of ten large awards - all above one million dollars in size,
with an initial range of one to forty million dollars. (As you might expect, in-
cluding changes, growth, and the like, that forty million dollar award is now



about 150 millions.) These large contracts were of interest and fiirnished an
informative pictirre of what happens in the award of R&D contracts. But the situa-
tions studied did not provide sufficient qiiantitative results of the sort that
managers of marketing and engineering activities would like to see before drawing
conclusions on new company policies. Me then turned to two Department of Defense
field centers where we attempted to gather more statistically-oriented data, data
that I am going to present to you this morning. In the first of these installa-
tions we studied forty-one competitions. The competitions resulted in awards of
research and development contracts, ranging from $100,000, the minimum award size
that we chose to include in our study, up to eight million dollars in size. In
this first installation, the awards were made during the period from January, I96O,
to June of 1963. Our second study in another DOD field center covered forty-nine
contract awards, ranging from $100,000 to in this case $2,000,000 in initial size;
and these contracts were in the time period from May, 19^2, to June, 19^^^ the
time that we started gathering the data on this last study. Thus our data are
drawn from fairly current cases, the latter study reflecting whatever influences
the McNamara regime has had on the R&D award process.

The next five figures describe some of the results of these studies. Four
of them show data from the two Defense Department organizations, the results of
the first installation at the top of the page, and organization two down below.
We shall compare the award structure in the two studies on the basis of several
different dimensions » In Figure 1 we look at that measure which most people
think is the most important in R&D awards, or at least it is the most broadcast
as the most important. That measure is the evaluated technical rank of the com-
petitors for an award., What we have done here is to plot a frequency distribution
of the award of contracts as a f\inction of the evaluated technical rank of the
individual companies receiving the awards. In the first organization, the graph



o
a:

<

<



z



40



20



\



^^.



S 1 >



in ^5 out of U9 award cases, better than 90^> following the technical evaluation
no more than three companies were possibly in the running for the award o^ the
contract. You sec from these data the power of the technical evaluation situa-
tion as determinant of who is going to receive the award. We shall return again
to this point, tliat it is the power of technical evaluation that you should remem-
ber in considering marketing strategies to be adopted for winning research and
development contracts.

Wc find in Figure 3 what happens when the award winners are ranked on the
basis of their cost positions, looking at how much each bid relative to his com-
petitors for the award. It is important to note that all the awards were either
CPFF, CPIF, cost reimbursable or cost sharing in some sense, that is in no case
was there a formal incentive for a low bid. There were no fixed price competi-



to

Q
QC
<

<



O

z



20


\


^ALL PROPOSALS




1


iO


-\










1


^^^-r-^t—


— u-






3



II



Q

q:

<

<

u.
o

d

z



20


r~


V


z

"TECHNICALLY ACCEPTABLE"
PROPOSAL ONLY
(-f 16 MORE IN WHICH


10




\


ONLY I WAS "ACCEPTABLE")






1


• 1 1 1 1



\



3 5 7 9 11

LOW BIO HIGH BIO

RANK OF PROPOSED BID AMOUNT



AWARDS AS FUNCTION OF BID AMOUNT



Figure 3



-8-



tions in all of oiir 90 avards, and as ycai well know, the cost gro^rth that takes
place in cost-plus contracting is sufficient to v;holly distort any initial bids
that are presented. Yet we still c^t this perhaps sxirprising picture, that
looking at all the proposals in organization #1 (even those that were not tech-
nically qualified) and in organization v^2 at the technically acceptable proposals
(cost data was not kept on file for those not regarded as technically acceptable),
the low bidder received more awards than any other bidder. This shows up even
though for the second group studied I dropped the sixteen cases in which only
one company was ranked acceptable. (l did not know how to treat that one com-
pany, as a high bidder or a low bidder.)

These results still do not fully explain the R8JD award process. However, we
first had technical evaluation determining so much, and secondly, cost evaluation
apparently explaining much of the rest. But there is one catch. All of these
formal evaliiations, of course, occur after proposals have been solicited, after
proposals have been prepared, after proposals have been received by the govern-
ment agency. But things do take place even before these phases. And one of the
things that takes place before proposal solicitation, preparation, receipt, and
evaluation is that the technical initiator in the government agency prepares a
procurement request (PR) on which he indicates among other things a list of sug-
gested companies he has in mind for doing the job. In our research studies we
went to this source of data, the procvirement request forms in the government files,
and we produced the results shown in the next figure. We have plotted for both
agencies the frequency of awards as a function of the winning company's position
in the list of suggested companies on the procurement request form, made up by
the technical initiator six months to a year prior to the evaluation of proposals.
We have excluded any alphabetical lists, because we do not believe that a company



-9-



20



o





(r


\


<
^ 10


A


u.


\


o


• N


o

z


m



'-\ 4-



10



15



20



10 r



V)

o







fm


a:

<




I




<


5




o




\


\


d'






^^








r-^^ 1






1


5 10
POSITION ON PR LISTS



AWARDS AS FUNCTION OF PREFERENCE INDICATOR

ON PR LISTS



Figure h



• 10-



name beginning with "A" is more favorable than one starting with "z" for getting
government business. (l must admit, however, that a recent study that I have con-
ducted of spin-off companies from the M»I.T. Instrumentation Laboratory shows
that TOfo of the twenty-eight spin-off firms have names starting with "A" through
"D". But we shall assume that alphabetic position is not a prejudicial factor.)
Looking at these non-alphabetical cases, we make a very simple hypothesis - that
where a firm's name appears on this list prepared by the technical initiator is
a good indicator of the initiator's preferences. When we plot, as in Figure ^,
the award of contracts as a function of listed position on the PR form, we find
that indeed the data support this hypothesis in both organizations.

We can take one more step with our data. If a company preferred by the
technical initiator and so listed, did not bid (there appear to be numerous slips
of this sort), it is obvious that this preferred company could not win. Revising
the lists by dropping these no-bidders, and looking at the frequency of awards as
a function of those who were listed and did bid, we produce the sharper curves
of Figure 5-

The key question is, "Are the indicators shown in Figures k and 5 important?"
Our answer is obviously, "Yes I" The data shown here are drawn from procurement
request forms that precede by six months to a year the technical evaluation mea-
sures shown in Figure 1. It is almost always the government engineer or scien-
tist who prepares the list incorporated in Figures k and 5 who also prepares the
evaluations that produce the set of technical evaluation curves. Whatever pro-
duces the technical prejudice, the feelings of confidence and trust in one parti-
cular organization rather than in another, that is built into the initiator at
the time that he prepares the procurement request, seems to stick with him
throughout the entire formal competitive process. It is apparent that if you
want to win research and development contracts, you win by competing prior to

-11-



to
o

<

<

li-
o

d

z



20



10



ri-



_\



J' NO- BIDS"
DROPPED



• • v«



10



15



20



-NO -BIOS"
DROPPED




POSITION ON PR LISTS

AWARDS AS FUNCTION OF PREFERENCE INDICATOR

ON PR LISTS



Figure 5



■12-



the preparation of the procvirement requests, not dxiring the period of time of
formal proposal solicitation, proposal preparation, and proposal evaluation. The
proposal solicitation, preparation, and evaluation are responses to a decision
by the technical initiator to undertake a set of technical acts under contract,
It is clear that he generally enters into that set of acts already committed,
at least in his own mind, to one or two companies.

Let me relate some anecdotal Information as to the kinds of things that
some technical Initiators do when they are thwarted in their attempts to award
the contracts to the companies that they technically prefer. As you well know,
not every attempt to issue a sole source award results in a sole source contract.
There are a large number of contracts that are awarded by formal competitive pro-
cesses after the technical initiator has tried very hard to make a sole source
justification. One example copied exactly from an RFP in our sample of contracts
demonstrates the case of the angry initiator who was not allowed to go sole source.
He wrote the RFP so that of the twelve solicited firms, only Company A, the "favored
sole source company", would bid. A few selected lines from the RFP might provide
clues to why none of the eleven other solicited companies bid. To quote, "This
will require that the bidder, in order to qualify, need have a minimum of success-
ful experience with Doppler equipment identical with that developed and operated
by Company A." You don't have to be Company A; you only have to have the same
experience that it has.' Here is another quotation from the same work statement.
"Note: Experience comparable with that shown in the Company A XYZ studies and
major subsequent ionospheric studies will be considered satisfactory." Thus, If
you had all the same experiences as Company A and you were really familiar with
their equipment, there is a good likelihood you could proceed and bid on thisoaitiBct

Ve can provide numerous other examples of certified testimony or extracts
from official government files covering the contract awards in our studies. In
one award case the procurement officer said during our interview that the initiator

-13-



was sold a bill of goods by Company X. Company X, needless to say, got the
award. In another case, "Company Y had worked with the people who had prepared
the work statement, the initiators, had previously done a feasibility study and
obviously had an inside track." Company Y got the award. Quoting from a third
case, "The M agency had quite a bit to do with the selection of the recommended
sources and the evaluator might have been able to have justified a sole soirrce,"
according to the procurement officer. The recommended source, needless to
emphasize, got the award. Says the procurement officer on another award, "Company
N could have been made the sole source." Company N got the award. In still
another case. Company Z vras requested as sole source, but the request was re-
fused. Company Z got the award. And again our interview records state: "Agency
K -Arote the specifications and vas the major participant in the evaluation.
Agency K had a predisposition towards Company R." Company R got the award.

Fe can repeat simj.lar occurrances time and time again. We can cite not
only subtle approaches, but in some cases really malicious approaches by the
tachrical initiator to make sure that his opinion would not be thwarted by the
formal practices of procurement. These government technical people are not risk-
ing their jobs because of political pressure or due to any graft or corruption
reasons. They act this way because at the time they are ready to go out on con-
tract, they are truly convinced that one or two particular companies are the
companies to do this work, that no other company could carry out the work as well
for the good of their program objectives and for the good of the government.
They fear that the formal procurement mechanisms are going to halt them from
being able to award the contract to the organization in which they have confidence.

Before finishing this aspect of discussion, I should like to describe some
interesting results that have come from a study being done at M.I.T. on the deter-
minants of technical effectiveness.^ Using data from twenty- two R&D contract



compctitionc, the researcher hao fp-uphcd the average effort for all the biddoro
rcccivinc a fiivcn technical rank. Figure 6 illustrates these results, #1 bcir.3
the highcct technical rank. In general, the data produce a U-ohaped curve, in-



2 —



-J 3
O 4

o

< 8

q:

^ 9
<

y 10

2:

5"

UJ
H 12

13



x^ — ^



■^^



/I.



-K-



/
\



\



V




N
N



THRESHOLD OF
COMPETENCE



N



\



\



\



N



\



\



N







ir



\



0.8 0.9 1.0 I.I 1.2 1.3

MEANS OF NORMALIZED LEVELS OF EFFORT



LEVEL OP EFFORT VS. TECHJIICAL RANK
(22 R&D PROPOSAL COMPETITIONS)



Figure 6



dicating that if a company is above some threshold level of conrpetence, putting
in more proposal effort results in a higher evaluated technical position. Below
this threshold level of competence, putting in more effort apparently demonstrates
company ip;norance more clearly, liut the one nziin exception to this U-shape is
the CO].: a.;y ran]icd vl 1 teclmically, ^rho tisually is t^he winner of the award,. As
you can zee, the -„ 1 tcchnically-ovaluatod firm jjuts less effort on the average
into the proposal than its ]}2f )'i2, )'i'^, and -l,^ technically- ranked competitors.
Our own results provide the reasons for this curve. The #1 technically-ranked
company is not competing primarily by means of proposal competition. It is com-
peting largely through the pro-selling efforts that have been addressed to the
teclmical organization in general. This pre-selling has caused the company to
be ranked i,'l technically.

In closing this phase of discussion I shall make one comment in passing.
IVo months from now, at the Office of Naval Research Conference on Research Pro-
gram Effectiveness, I intend to examine the cost/effectiveness of this R&D pro-
curement system that so often seems to necessitate six to eight months worth of
foriral proposal preparation and competition by industry, apparently for little
or nought. 5 There is grave doubt in my mind that we should be straining and
stressing so much these formal procedures that our Secretary of Defense tells
us produce cost savings every time we instigate them farther in the Department
of Defense. Our research evidences suggest the counter-conclusion that the


1

Online LibraryEdward Baer RobertsMarketing and engineering strategies for winning R & D contracts → online text (page 1 of 2)