It is more than a decade since the classical model of
a biotech company ruled unchallenged: the glass-fronted building on the
gleaming modern science park, with the best part of a hundred doctoral level
scientists working seemingly double shifts to deliver an entire pipeline of
high value product candidates. The
central principle underpinning the investment case for such a model was that a
team of that kind, once assembled, could find and develop multiple assets that
could be sold one after the other to big pharmaceutical companies. The key word was ‘multiple’ – no single asset
could justify the vast cash drain such a model demands.
Finding such opportunities, which can genuinely
sustain this “company” model, has proven to be tricky. Entrepreneurs would “pad out” a single high
quality asset with several earlier stage opportunities that turn out to have
considerably less value. After all, it
is virtually impossible for any asset owner or research institution to generate
genuine pipelines – large pharmaceutical companies with billion dollar R&D
budgets can hardly even manage that feat.
Experience has taught many investors that the only
exception to this rule is the platform technology. A few exceptional technological breakthroughs
have the capability to act as the goose that lays a clutch of golden eggs
(humanized antibody libraries have been an obvious example of such a
technology). Presented with such an
opportunity, the glass-fronted building may still be the right solution – but
you better be sure the platform is all it is promised to be and not a sterile goose
with golden feathers.
For the rest, the answer is “project” financing, not
“company” financing. Instead of seeking
to balance risk at the level of the individual portfolio company, larger investors
are increasingly seeking to balance risk across their whole portfolio. Provided each asset gets less money invested
in it than would have been the case with the big, old fashioned “company”
model, then a proportion of those assets can be allowed to fail. Looking now from the perspective of the
investor, the game has changed: the “fail early, fail cheap” mantra of big
pharmaceutical companies has replaced the “more than one egg in the basket”
mentality that underpinned the old “company” model of biotechnology investing.
Typically, these “project” financed companies look
very different from their predecessors. The
liturgy now consists of capital efficiency and of an unswerving dedication to
moving to the next value-creation milestone even if doing so exposes the asset
to the risk of outright failure.
The other big difference is that progressing a single
asset through the product development pipeline (and it doesn’t matter whether
we are talking about a therapeutic, a diagnostic test or medical device) needs
lots of different skillsets. In a large
company (even one the size of the old “company” model of biotechs) the majority
of these competencies could be accommodated – the “reusability” of these
capabilities was part of the justification for putting several assets through
the same development pathway. But with
only a single asset incubating in each nest, less than a full “unit” of each
capability is required. Each skillset in
turn is required, very intensively, but only for a relatively short
period. In therapeutics, for example,
preclinical development of a single compound might take nine months, in which
period expertise in safety pharmacology and toxicology will be heavily in
demand. Upon moving into the clinic,
however, the emphasis quickly shifts to trial design, statistics and the like.
The solution is to out-source these functions. The advantage of doing so is obvious: you
only pay for the capability when you need it.
You can have three highly skilled and experienced individuals working on
your regulatory toxicology studies while they are on-going, without wondering
what you are going to with them once the study is finished and reported. Even if the cost per unit time of accessing
skills this way is higher (and it surely must be, since the contract provider
has to secure a sufficient profit), the overall cost is lower because you are
only paying for the amount of time and effort actually required. Lower total cost, higher capital efficiency.
A less obvious, but equally relevant, advantage is the
possibility of accessing better, more experienced and more capable, individuals
on an out-sourced basis than would have been possible, let alone financially
viable, on an employed basis.
What about the disadvantages? Consultants by their very nature are more
difficult to incentivize than employees.
Their association with the company is transient, and their recompense
primarily consists of fees payable irrespective of the final outcome of the
project. Too often consultants have
their focus divided between several (or, for the most in demand, many) projects
and the quality can suffer. Faced with
two simultaneous demands for urgent attention, they will always have to choose
one. With employees, managers can do
just that: manage. They can decide which
of two competing demands is the higher priority for the overall health of the
business.
So for a modern, “project financed” biotech the
question of the moment is simply this: just how virtual should I be? Is being virtual, semi-virtual or fully
staffed the “ideal” to be aimed for?
Virtual tech companies have proliferated in recent
years. Many are virtual through
necessity – without sufficient cash to progress their assets, their virtual
existence is tantamount to suspended animation.
These companies are often a single individual with an asset portraying
themselves as a company either (in their view) to aid in raising finance or
perhaps, where a group of individuals are involved, simply as a mechanism to
divide up the ownership in a convenient and legally robust fashion. But these are not really virtual operations –
because they do little or no actual operating.
Largely virtual but nonetheless operationally
aggressive companies, with financial muscle, are a much rarer beast. One such company is Funxional Therapeutics
(founded by the author, and where he is now Chief Scientific Officer).
Funxional is a Cambridge (UK) based therapeutics company with a single product
in clinical trials. With only a handful
of employees, mostly engaged in corporate management rather than operations,
but nonetheless supported by significant investments from blue chip venture
capital firms, Funxional is operating one of the purest forms of the virtual
biotech model. It is ‘”project
financing” taken to its natural extreme.
And for the most part it has reaped the benefits of this virtual
lifestyle. It has advanced its lead
candidate from research to the end of phase I at a remarkably low cost and at a
speed to match the industry’s best.
Whether or not the product actual works, allowing a profitable exit, or
not is in a sense neither here nor there – that simply reflects the quality of
the founding asset. From a purely
operational perspective, it has been a resounding success already, by reaching
the value-inflection points as quickly and cheaply as possible.
What has been the key to such success? And how were the pitfalls avoided? The single most important factor is the
management team, and the management structure.
Ensuring every part of the project is managed by an “inside” consultant
whose job it is to manage the contractor responsible for delivery provides the
ability to be an expert customer while being almost entirely virtual. In one sense, it doubles the management
overhead – contractors are so keen to provide project management as part of the
deal that, on the face of it, it can seem difficult to justify adding another
layer: with the senior management overseeing an expert consultant directing a
contractor’s project manager who is managing the work. But convoluted as it sounds, this is one of
the secrets to successful virtualisation.
Our own experience, in TCP Innovations, has shown the importance of these “expert customer”
roles. We have witnessed a massive rise
in requests for the provision of domain experts to perform exactly these kind
of roles for a wide range of companies. With
experts covering CMC, toxicology, chemistry, pharmacology, statistics,
intellectual property and more, the right people are out there to make sure the
ultimate virtual model can deliver success, more cheaply, more quickly and more
flexibly than ever before.
There is still some way to go before all but the most
avante garde investors regularly go all the way to this level of virtual
project management. One limitation is
their own internal benchmarks: years of analysing more conventional, less
virtual businesses has engrained certain expectations. Most seasoned investors expect management and
administrative expenses (that is, everything but the value-creating operations
themselves) to be less than a certain fraction (perhaps 30%) of the total
spend. Yet highly efficient virtual
models may not reach this benchmark.
Their failure to do so comes from several sources: out-sourcing saves
money, so the operational costs are lower than with a conventional employed
operations team. At the same time,
management gets more expensive, particularly with this extra layer of “expert
customer” consultants managing every aspect of the operations. So even though a “project financed” lean and
mean, virtual company spends less overall building its value, for superior
capital efficiency and better returns for every stakeholder, the fraction of
the expenditure going on (what has previously been considered relative
unproductive) management and administration has risen sharply.
Until investors gain enough experience to see that a
rising fraction of expenditure on management can (though clearly is not always)
be a sign of an efficient model, it will be difficult to win them round with a
business plan for a fully virtual biotech.
Worse still, asked to cut the management expenses back to a more traditional
level risks surgically removing the “expert customer” capability which is so
key to successful virtual operations in the first place.
Virtualisation is here to stay. Investors and managers alike have to learn new
benchmarks, new practices and above all a greater trust and closer relationship
with each other. Doing so exploits the cost savings of
out-sourcing, retains the ability to do things well, and, in the end, this new
model of a “project financed” virtual biotech may be the only show in town.
This article was written by
David Grainger. David is a scientist directing an internationally recognized
research laboratory in the Department of Medicine at Cambridge University,
focused on inflammation. David is a biotech entrepreneur and investor, having
founded several biotech companies, including Funxional Therapeutics, and he
runs a successful consultancy business, TCP Innovations. He is also a Senior
Partner at Total Medical
Ventures, a life sciences boutique investment group, and a Principal at ATPBio.
