William James Griffiths graduated from Imperial College London in 2004 in Chemistry with Management, he spent several months as a scientist at UK biotech company Celltech, but realized that life behind the bench was not for him and has since invested his time in developing the ChemRefer.com website. The site offers quick access to full text chemistry literature that is available free to those in the pharmaceutical and chemical industries.

You have a fairly conventional science education, what made you opt for chemistry with management for your degree?

After completing the 3 year BSc Chemistry course, I could either choose to graduate at that point (which would have left me with a straight BSc Chemistry) or do an extra one year 'add on' course in Management and then graduate. I was skeptical about Management at first. I thought it was a lot of jargon over nothing, but it did teach me many of the skills that I used to start ChemRefer.com and it was very worthwhile in the end.

Generally, I think chemistry teaches you a lot of very technical skills but, since most people learn it in an academic environment, they are not necessarily encouraged to consider the advantages it gives you in a commercial situation. This is what the Management Course succeeded in doing.

What did the entrepreneur project in your degree involve?

That was actually a group project. We had to design a product and plan a marketing strategy for it. Our product was an ultra-thin, pressure sensitive device which could be laid on top of a tennis court and act as a line calling aid. It would change color when a pressure (e.g., a tennis ball bouncing) was exerted on it. This color change would be reversible making the device reusable for an indefinite period. Piezochromic materials have the necessary property of being able to change color reversibly. There were several technical problems though. Such materials are UV sensitive and, since most tennis courts are outside, this meant some way had to be found to shield them from UV rays. They also tend to undergo the reverse color change very, very quickly. Too quickly for a tennis player to walk over and check the mark! Our research suggested that combining the material with some sort of polymer might slow the reverse color change process, since the polymer would obstruct the molecules (which were perturbed by the impact of the ball) from returning to their original orientation.

Is the invention progressing towards commercialization?

I presented the whole idea to EPIS (Edinburgh Pre-Incubator Scheme) and they were very enthusiastic about the idea but they were unfortunately unable to match me with a suitable supervisor so it couldn't go ahead. Nevertheless, I felt that I got a good reception from them and it encouraged me to go and start ChemRefer, Ltd., shortly thereafter. I would certainly encourage anyone thinking of starting a company on the basis of a technical product or service to pursue that through EPIS.

You spent some time working for biotech company Celltech, I take it you didn't enjoy working in the lab.

Unfortunately not. It wasn't the job for me. On the other hand, had it worked out I would never have set up ChemRefer so, in the long run, it probably pushed me into thinking about what I really wanted to do. I wanted to work in science somewhere and I wanted to keep the option of lab work open. It really would depend exactly on what kind of lab work since it's not fair to classify it all in one category. Naturally, I prefer to work for myself doing something of my choosing since it's far easier to get motivated. Watching ChemRefer develop over the last few months has been a good experience and I wouldn't swap it for any other career.

ChemRefer is an intriguing concept, how did you come up with the idea?

The problem with manufacturing-based business, like the piezochromic device, is that you need to convince others to back you before you can proceed, since the initial capital outlays are considerable. This got me thinking about chemical information (which requires zero capital outlay) and eventually I came up with ChemRefer, a company I could start without anyone else's say so and with a budget of zero. I also had enjoyed working on the literature report as part of my chemistry degree and so it seemed an obvious avenue to go down.

But, I actually never planned for ChemRefer to be a search engine. ChemRefer.com was initially a single submission form ("Enter your email address" and "Enter your research query") and I would research people's chemical problems that they would submit through this form. Only problem was, when traffic began to exceed 10 visitors per day I was swamped with queries and began to look for automated ways to operate ChemRefer. Converting it to a search engine was the obvious choice. Also, having researched so many people's queries through the old system, I had an understanding of where to find good quality literature, so it wasn't a problem to locate resources for ChemRefer to 'refer' visitors to. The whole thing was unplanned and evolved entirely organically. If ChemRefer continues to flourish, it will be because of this built-in adaptability.

Indeed, how do you see ChemRefer evolving, presumably, you'd like it to be the open access equivalent in chemistry of PubMed?

I am currently looking at various options. It all depends on who wants to invest and how much they are prepared to back ChemRefer, since these factors will drastically affect the resources available to ChemRefer. That does not mean I will be less ambitious without investment, just that different avenues will have to be explored that require fewer resources. Naturally, ChemRefer aims to allow users to search as much chemical and related literature as possible, but this must not be done for the sake of having millions of searchable articles. The quality of search results must be high and I am aware of the comparison with PubMed, but it's not a reality just yet. Careful and constant expansion is the idea. ChemRefer is also commercial and this throws up entirely different opportunities/risks since publicly backed projects have an assured stream of funding.

So, what problems have you faced so far and how have you overcome them?

Web design, configuring the search engine, setting up hosted exchange e-mail, and most of all, marketing ChemRefer have all posed problems. Marketing is expensive and, without money, only a limited amount can be achieved unless you can conceive a brilliant viral marketing scheme. Contributing to forums and blogs and trying to help with other's research has been very effective as people can then judge the ChemRefer service for themselves. Contacting chemical companies by phone and e-mailing related websites is useful but you have to be careful. As for web design, I got around that by using a free website builder since I know little about HTML. I think I did okay with it though. I have generally good feedback especially about the images on ChemRefer.com.

In terms of the OpenAccess movement itself, I assume you consider it important, but can you tell me why you think so?

Relative to other subjects, chemistry is expensive to teach and practice in and therefore unattractive to finance. It makes no sense for governments to shell out for this only to have the fruits of their expenditure (the published research) withheld by those who did not pay for it. If this burden were removed, companies and especially libraries could then direct more attention to improving scientific educational resources. Anyone could open a library on a shoestring and quickly make the most authoritative content available to a local community.

What do you think are the most important trends in chemical information today?

Reduced barriers to entry into the chemical information industry. Anyone with a good idea, some knowledge and determination can now make an impact. ChemRefer's entire marketing expenditure so far has been 96p (about $1.50). Would that have been possible a few years ago or even last year? Overall, this is a healthy phenomenon because it means that competition is maximized and it follows that this leads to maximum innovation in chemical information services. There is still an 'elite' chemical information industry, e.g., Thomson Scientific, SciFinder, etc., and they seem to escape the effects of this by being so comprehensive that in order for anyone to match them, a significant capital outlay would be needed. But even this may change in five years.

e-Molecules (formerly Chmoogle), PubChem, and others offer searching using structure and sub-structure from drawn chemical structures. Might you be able to incorporate such a facility into ChemRefer?

I have made enquiries about this very issue. The ChemRefer search engine can have a structure search add-on fitted to it without any major overhaul. It would require something that would translate a graphical chemical structure into a string of text that can be used as a search term (even if that string of text is behind the scenes) and then the existing search engine would do the rest. Also, for each page with a specific chemical compound, a diagram could be assigned to each page and this would appear alongside the search results, just like a Google image search. Obstacles to this could include unforeseen adverse effects on the quality of search results from structure searches. But these problems could probably be ironed out.

Do you think there is space in the shadow of Google Scholar, Live Academic, and the likes of chemindustry.com for another niche search engine?

Specialist search engines are always vulnerable because they cannot spread their risk over different niches (since they have only one or a few niches to start with). As long as ChemRefer covers a unique specialty (i.e., full text chemical literature only) and can adapt quickly to change as it has done over the past few months, things should continue to progress. Websites can literally adapt to increased competition within a matter of hours and ChemRefer will certainly have to do this many times in the future to survive.

If an offer came along, for how much would you sell ChemRefer?

An investor is really what ChemRefer needs most. If that meant selling a share of ChemRefer Limited (minority or otherwise) then, for the right price, I would be very receptive! An ideal investor would also be a co-contributor to the ChemRefer Project, helping to develop the service.

Finally, what's the long-term plan?

Making a success of ChemRefer is the priority for the next few years at least. Maybe diversifying into new projects as well. Apart from that, it's hard to predict the future!