MARK LANCASTER - SDL INTERNATIONAL PLC (SDL.L)
CEO Interview - published 06/12/2000

DOCUMENT # KAE031

MARK LANCASTER is the Chairman and Chief Executive Officer of SDL
International Plc.  After graduating with an HND in Electrical and
Electronic Engineering, and later studying for an MSC in Software and
Computer Architecture, Mark Lancaster established an early career as a
design engineer at Satchwell Control Systems.  He then progressed his
career as Project Manager at Lotus Development Corporation and later
held a senior management position with software developer Ashton Tate as
International Development Director.  Lancaster founded SDL International
in 1992 when he identified the need for a high-level service provider
for the globalisation of software.  Mark was a founder member of LISA,
the Localisation Industry Standards Association with which he continues
his involvement.  In 1997 he was a Southern region finalist for the
Ernst and Young Entrepreneur of The Year Award.  Mark is married with
three children.

SECTOR: Multimedia Software

TWST: I suppose a good place to start is with a picture of SDL
International and maybe a little history just to set the context

Mr. Lancaster: We have about 500 people now.  We have offices in most
areas of the world: Japan, China, Sweden, Spain, Italy, Germany, France,
a number of offices in the USA, and a number of offices in Ireland,
Hungary and Romania. Most of the production is done in the UK, Ireland
and Japan, and most of those offices I've mentioned are production
offices but they're satellite production offices.  They're not just
little sales subs with one or two people in; they generally have
anything from five people to 30 people. The purpose of the business is
to deliver globalisation solutions to our customers.

TWST: What globalisation solutions?

Mr. Lancaster: There are three prongs to it; the first one is probably
the consultancy arm, and that would be consultancy on getting the
customer's product - typically high-tech products, web pages etc -
internationalised, to get it to a state so that you can actually take it
into different languages.  That's the consultancy part of the
organisation.  Then there would be the actual product side. There are a
couple of mainstream products that we market.  One of those is
SDLWebFlow.  That's a multi-lingual content management product which
allows  companies to keep their English web pages synchronised with all
their multi-lingual web pages, so any changes they make to their English
or German web page, all the other web pages would automatically be
synchronised.  Then there's the SDLX product, which is translation
memory software; this is a sort of computer-aided translation technology
which just means you minimise the number of translations you have to do
because you can re-use words and phrases that have already been
translated.  Then there's the actual production itself, which comprises
project management, a lot of engineering, and obviously translators and
graphic artists.  Most of our customers are typically in the high-tech
sector; clients like Hewlett-Packard, Xerox, Canon, and Sybase.  There
are also the other client sectors such as Sony Disney; and then the web-
type companies such as betmaker.com, LastMinute, e-Bay.

TWST: What size is the market for multi-lingual solutions for
translations?

Mr. Lancaster: I think it was Allied Business Intelligence who estimated
it was around about $17 billion dollars.  It's at least 12 with the
advent of all the content on the web.  It's very likely 17, and if you
really want to be optimistic you could say it was about 30.  But I'd
have at least a 50% confidence that it was 17 or bigger.

TWST: How is the market evolving, how rapidly, and where is it growing?

Mr. Lancaster: Obviously the web is where most of the growth is,
primarily because everything is converging onto the web; the content is
still there, but it's just in HTML or ASP format.  So that's the biggest
growth opportunity. There's also a lot of growth opportunity in the
product side as well.  The great thing for us is that as
telecommunications continues to grow and the technology continues to
evolve, then obviously people are able to be global more easily.  The
web has just accelerated that. Ten years ago, a company had to be
turning over $50-100 million before they started thinking about going
global.  Now you get someone turning over a couple of million or five
million and they can be global immediately just by creating a web page
and globalising it.  So the entry level's gone down.  The larger players
have to be global or they just can't compete; and there's the
opportunity for the smaller players to be global quickly and at
relatively low cost.

TWST: What about the competitor environment?  Who else is involved in
the business?  How is the competitive landscape moving?

Mr. Lancaster: There are a couple of areas.  Because we're in products
and services, there are three prongs that either attack us or we attack
them.  On the service side, there's companies like Berlitz, Bowne,L&H -
that's Lernout and Hauspie - and Lionbridge.  Those guys all provide
globalisation services, so they would provide the actual execution side. 
And then there are companies such as  Globalsight and Idiom who compete
with us on the  multi-lingual content management side. And then there is
the Anderson Consulting, Ernst & Young, KPMG consultancy arms; they
would, in a sense, be partners rather than competitors, but they offer
globalisation solutions in the broader sense.  If you say
"globalisation" it pretty much means anything these days, and we would
typically work with many partners in this area, but we probably provide
the broadest services around.  For example, we competed with Ernst &
Young on a bid a few months ago; but they would typically just sub it
out to somebody like us. No single company has the lion's market share. 
We're around $40 million turnover, whereas the other three or four that
I mentioned first are probably more like between $50-70 million
turnover.  If you take Lernout and Hauspie, Bowne and Berlitz, they're
all globalisation divisions of larger organisations. Berlitz is mainly
training and they've gone into globalisation because of the opportunity
they saw and they've acquired a load of companies and glued  together,
as have Bowne, and Lernout and Hauspie. We are completely dedicated, as
are Lionbridge.  Lionbridge was created by a management buy-out from 
Stream  It gets considerably more fragmented below those companies I've
just mentioned with a larger number of very small companies.

TWST: How exactly was SDL International formed?  Where did it come from?

Mr. Lancaster: Prior to working for SDL, I worked for Lotus Development
Corp. and then Ashton Tate Corp. I was on the other side of the fence. 
I was then looking to get the products at Ashton Tate, for example, into
Europe and internationally, much faster than I was able, due to lack of
outsourcing opportunities.  And so in 1992, I decided to set up a
business that was able to help companies go global.

TWST: Let's then talk about the products that you mentioned and the
specific services.  What segments of the market are they aimed at?  Can
you give me a more detailed description of exactly what they do?

Mr. Lancaster: If we take SDLWebFlow, that product is aimed at anybody
who seriously uses the web as part of their business.  A global tool to
do business on, to do business with.  They have to have a solution to
synchronise their multi-lingual pages with their English page. 
Typically a web page will be changing on a continual basis.  If it is
not, it is unlikely it will get repetitive visits and therefore it won't
be a viable proposition.  It's typically the low-value transactions, and
there'll be a high volume of those transactions and the sort of customer
that will look at those probably won't speak English; and that will be
in the leisure field that that happens, so companies like Leisureplanet,
LastMinute, e-Bay, auction sites, the finance field.  Those are the sort
of verticals, and then on the horizontals in terms of size we would
target British Airways, any sort of transportation company.  We would go
for the big globals through a direct sale, whereas with the smaller ones
we would do that through partners, partnering with, for example,
Interwoven, and content management systems or interactive web
developers, where they would develop a web page and their client would
have a need for going global.  But we would typically target the web,
the dot-com start-ups that have to grow and go global very rapidly, also
the big corporates who have longer lead times but possibly more
stability.  And then on the other side, in terms of sector, the sectors
we go for as I said earlier would be the leisure, transportation,
business to consumer, typically.

TWST: What exactly does SDLWebFlow do?

Mr. Lancaster: Essentially, it has a little program that sits on the
client's server, and all that does is monitor changes.  When there is a
change that's identified, it communicates with a web flow server
obviously through the web, and we'd have our web flow server sitting
somewhere - they could be anywhere in the world, but typically they'd be
with the client's master Website servers - and when a change is detected
it communicates with the web flow server.  The web flow server would
then identify whether that change had relevant content and needed to be
translated through workflow.  It would, on a parallel basis, download
all the assets from the main site, be those ASP, static HTML or dynamic
HTML, and it would then run the necessary changes through translation
memory.  The translation memory will have stored all the existing
English and translated sites material, and so typically there will be
changes; there would be a new product going up on the site, a changed
profile of a product, a new hotel that had been added, some cheap seats,
etc.  So a lot of the text would be similar and there would be changes. 
Translation memory would draw from all the existing text on the site or
in the memory; and it can be a massive memory.  So if this is new, but
I've already got 100% matches, we can just automatically translate it. 
There could be a similarity; it could be a news feed site - it could say
the yield on a coffee field in 1995 was 30%, and then that could have
been updated to say, the yield from the coffee field has now moved in 
2000 to 40%.  It would find the old piece of text which said it was 1995
and 20%, and the translator would just see those two pieces of text -
here's the translated old, here's the new English.  It's virtually a
match so they just have two words to translate.  It would automatically
compile all that information, and translators could then log-on to the
system and would automatically be able to see the web page - the old one
and the new one - and they'd be able to perform the translation of just
the little bits that had changed by using the memory.  You then press a
button, it will send all that information back to the mirror site, and
you're able to view the mirror site to make sure the text is aligned or
if it's a graphic there might be more work and it might have to go to an
engineer. Once that has been done it would then be sent along to an
engineer/reviewer to check everything's compiled properly.  The engineer
would then log-on to the site - this is all through web browser - the
web browser will then approve or disapprove.  Once they've approved they
will automatically put the notification in via the Website browser and
then that information will automatically go up to the site.  And that
could be done in an infinite number of languages concurrently, because
the translators can be sitting anywhere in the world. So essentially
you've got a Website that's completely synchronised in multiple
languages.  All the files are kept under control, all the archiving. 
You've got almost instantaneous translation and concurrency.  If you
imagine a large site and there's little pieces of text, Reuters for
example, with various newsfeeds, and that text is changing, to do it
outside of a workflow system without translation memory and computer-
aided translation would be a nightmare.  So it's quite valuable.

TWST: So there is still a considerable human interaction in the workflow
management?

Mr. Lancaster: Machine translation is not here yet.  We've looked at
that in quite a lot of detail, and people who have got serious web pages
are not going to be able to go to complete automation.  But the
translation memory element, the computer-aided translation, minimises
that significantly.  There is effort in checking, but then there always
will be, just because any language is pretty ambiguous.  And so I would
say that there is human interaction, yes.  I wouldn't say that is
significant because an awful lot of the automation is taken care of by
the computer-aided translation technology.  But yes, there are defined
checks and there is translation of sentences that are completely new and
a similar sentence structure can't be found within the memory.

TWST: That's one product, SDLWebFlow.  Was there a second product?

Mr. Lancaster: SDLX is the translation memory, so that probably more by
accident than by vision plugs straight into SDLWebFlow.  Some people
might not work on the web and they just might want to have a translation
memory product that they keep and use for their manuals, so that is sold
separately but it plugs into WebFlow.

TWST: What are the legal aspects regarding multi-lingual translation
services?

Mr. Lancaster: It can be very simple or very complex.  Obviously, if
you're translating a newsfeed then you have to make sure that it's
accurate, and you've got the compliance with local law.  For example, in
Germany there are different advertising laws to the United States.  And
there are also the actual cultural differences.  That's not a law issue,
but you're not going to sell your product in Japan in the same way,
using the same examples, as you would in the United States. In the
United States, you can condemn the competitors heavily.  In Germany
you're going to get sued if you do that.  So there are legal
ramifications there.  If you're selling, making commitments and selling
plane tickets on the web, then there's various transaction laws that you
have to abide by and loads of small print which has to be complied with,
and that has to comply with local law.  So there are legal
ramifications, and one does have to be aware of those.

TWST: Where are the new opportunities for you and what sort of products
will you introduce to capture those?

Mr. Lancaster: That's a big secret!  I'll give you some clues as to
where we're going. We will continue to evolve our computer-aided
translation technology.  We've recently acquired a company called ITP,
which is based in Ireland, and they have some very interesting workflow
technology, and so we will be integrating that technology to provide an
even more extensive workflow system.  The trouble with workflow is that
it tends to get incredibly complicated, so it's only applicable to a
very small number of people.  At the moment, the market is only just
realising they need to be global anyway, and so essentially as that
becomes more readily known, providing the pricing barriers are correct,
then smaller companies will need a simple, low-price product.  WebFlow
is about $50,000, and then you've got maintenance etc.  There's no way a
small company is going to invest that sort of money.  We're looking at a
cut-down work flow solution for those guys.  But primarily that's the
marketing opportunity on the low end, and that will be big.  Then on the
technology side, more computer-aided technology for  translation.

TWST: Let's talk about how you're going to achieve the goals that you've
set there.  First of all in terms of balance sheet; are you sufficiently
strong to carry out the investments?

Mr. Lancaster: We're a public limited company listed on the London Stock
Exchange.  I think post-rights issue we will have about $19 million
dollars in the bank.  We actually tend to focus our efforts a lot more
on technology and development investment rather than marketing spend. 
Our marketing spend tends to be limited to people like me speaking with
you - it's so much more cost-effective!  You just get more truthful
coverage. Marketing is what really eats the dollars, and so we don't
have high burn rates We tend to invest those in people and getting our
message across through journalists.  In terms of acquisitions, we
obviously have paper and cash to play with; we did all the acquisitions. 
We had $7 million in the bank and we decided rather than go for finance,
to have a rights issue.  We got some very good institutions supporting
and following us, and so even though there were very bad market
conditions at that time, we were able to go to the market and raise $22
million on a rights issue.  I don't think we saw that many institutions. 
We probably only saw about 15.

TWST: In making an acquisition, it's not to achieve critical mass in the
particular market place that you're in; it's more to fill in gaps on the
technological side? 

Mr Lancaster: That's where we're going now.  I'd
say that you need to be of certain critical mass to be on the radar
scan.  If you're not on the radar scan - which we are now on but we
weren't on two years ago - then you do miss things and have to go into
advertising, so I think it is important to have some critical mass.  At
the stage we are now - 500 people, $40 million turnover - I'm pretty
comfortable that we're OK.  Our acquisition strategy is much more to
fill in the technology gaps that we might have and also to fill in the
global gaps that we might have.  For example, I'm looking at Korea and
Latin America at the moment, to maybe open offices there.  But unless
there's a change in market conditions we're unlikely to go and acquire
another big company; that is a massive distraction.

TWST: How has the acquisition gone in terms of integration?  What have
been the challenges?

Mr Lancaster: Mergers and acquisitions provide many unique management 
challenges. We've done a lot before, but this is quite a big one.  I'm into 
integration very rapidly.  A few big hurdles to overcome at the beginning, fine, 
and then you get things integrated and you move on.  If you don't do that 
then you're in deep trouble.  I think we've probably done more structural 
changes and advanced the business we've acquired in the time that we've 
been in there than they've moved in the past two years.  So I'm very pleased.  
I've got good buy-in from their management team, a really nice, good set of
people who know what they're doing.  If you have that as a cornerstone,
good buy-in from the management, then providing the vision is shared,
things are going to go forward very nicely.  We are paying a lot of
attention to the customers, making sure they're all happy as we further
improve the services and just basically broaden the whole thing.  I'm
very pleased with the people.  We've lost a couple of good people, and
we've also let some go; and so that's been very good.  So I'm pleased
with the way it's going.  It's hard work, but I'm pleased.

TWST: One other item that I noticed in your growth strategy is joint
ventures or partnering arrangements, I think the most recent being with
Interwoven.  Can you tell me what exactly this involves for you? 

Mr Lancaster: Interwoven sell their content management system into the big
corporates, and essentially they are going to be talking to a lot of
people that will either already be multi-lingual or will want to be
multi-lingual.  And Interwoven will need to have some sort of solution
for that, and essentially we plug in very nicely and they have to solve
the problem.  We would literally integrate our product into Interwoven's
product.  It's not difficult to do, because they have a database back-
end technology, we have database back-end.  It works quite nicely, it's
quite easy to integrate the two products.  We also do that with a
company called Media surface as well, they are also a content management
system; and we would also partnership in the next six months or so with
the Interactive Developers.  I would see us going on sales courses with
those guys, with both the Interwoven's and the Interactive Developers,
and I would see them calling us in, or on the other side just literally
selling our product or our services through without being involved so
literally just one-to-many marketing.  We would have a license or
royalty that they would pay us every time they sold our product with
their product.

TWST: Your business is split between products and services and that's
the business model.  Do you expect it to change over the next two to
three years?  

Mr Lancaster: It'll definitely move more and more in terms
of revenue proportions to products.

TWST: What's the profit margin structure on the two halves of the
business? 

Mr Lancaster: On the services, you can only do so much,
because you're always going to have to get more people; you can make
things more efficient.  But we've been growing at 50%, and if you're
growing at 40-50% then you're going to be ploughing an awful lot into
infrastructure.  You're going to be running at a profitability of 8-10%,
whereas if you get to a more sensible, stable rate, then you're probably
going to get your underlying profit up from 8-10% to probably 12-14%. 
It's a very healthy business.  You could even go up to 17, and
especially because we have the automation element as well - that's very
powerful.  On the products side, it's difficult to tell but you would be
looking at fairly significant margins because we can sell so much of it
through our services.  So you've got a sales infrastructure and you've
got a partnership infrastructure and you've got the big company name. 
So the profit margins there will be significant.

TWST: And there's no expectation or nervousness that automated
translation products will become commodotised and margins will collapse? 

Mr Lancaster: If they are, I'll be there.  I'm very happy with that
because it helps the business.  Obviously, my goal is always to do as
much as I can with as little as possible, and I've looked at automated
translation for a long, long time, and we have got 150, 200 skilled,
degree-level translators in-house.  We know the professors who develop
and work on these systems and research them, and we're just starting up
our own division to specifically just review the market yet again. We
are watching closely the development of this technology and the
companies operating in this space. The question you've asked me, I get
asked on a regular basis, because you can see the conceivable benefits
of  the people side.  When we did the analysis on it, and we really
tried to buy it, it just wasn't worth it.  Even if we only paid five or
ten million, it was not viable because it added no value.  Now, as I
say, in three years' time that might be evolving.  In ten years' time,
it might be vaguely useable in certain applications, for example
translating your e-mail.  But for translating press releases with
pictures, profiles and spins, no machine can do that.  Machine
translation works very well with structured input.  So in the automotive
industry and with translating manuals, there's a good application for
it.  In the marketing web page type area, moving away from the very
formal 'put the bolt on here and do this', it doesn't work.  But we are
looking at it and I'm hopeful that in time you will be able to apply it. 
But now you'd spend more money reviewing what you've done.  You're much
better using computer-aided translation which is where we are.  So we
think we've got the leading technology that really does minimise effort,
available to us today.

TWST: At the level of management, can you give me a snapshot of the key
management team and what they do, and what do you do on a daily basis? 

Mr Lancaster: We've got Alastair Gordon who's the FD. He looks after all
the legal side of the acquisitions, mergers, partnerships - he has a lot
of strategic input into the business.  There's the Chief Technology
Officer and Chief Operating Officer, that's Keith Mills and Christina
Lancaster, respectively.  Keith looks after the strategy behind all the
technology. He looks after all the strategy and the execution of all the
products in the business.  And then Christina runs the operations.  In
the operational set-up we have a number of divisions all run by
divisional directors who have P&L responsibility and we have about six
to seven divisions in the organisation for around 60 people  $7 million. 
All those guys have P&L responsibility. Then there's the development
arms themselves, so there's the WebFlow group that has a divisional
director who reports directly to Keith. A lot of our work's done in the
US and we have two VP's located in the US managing our sales activity. 
We also have somebody focussed specifically on our SDL Webflow product
sales and marketing.

TWST: Are there any areas that you anticipate strengthening or changing
over the next 18 months to two years? 

Mr Lancaster: We will definitely evolve, so what we will be shaped like next year 
I don't know.  I'm pretty comfortable we've got a very solid structure running forward,
because we built that obviously to ensure we've got the structure that's
capable of doing all the things we want to do.  We'll bring in second-
tier stuff' we'll bring in probably another merger and acquisitions
person. We'll probably bring in an HR Director that will report in to
Alastair in the near future. I don't see anybody else at senior level. 
We've got most of the bases covered there.

TWST: When you look five years down the road, what's the vision for SDL
International?  

Mr Lancaster: It will probably be much more product-
oriented, I would suspect.  I think we will be pretty much a leader on
the technology side, and my goal is to just make sure that we continue
to give our customers the solutions that they're looking for.  Now, if
that means that they want more consultancy then we will build up a large
consultancy unit.  Essentially it'll be providing customers with the
solutions to go global in the most effective way for them.

TWST: If one turns to the financial ratios, what are the most important
ones that express the performance of SDL International? 

Mr Lancaster: I would say that growth is important, and so I would like to 
see 30% headline growth.  I'm not looking for explosive growth in the next year
or so, because we're making sure that we are investing and bedding in
We're fortunate because we've got an established business, a pretty
solid revenue stream which can fund an awful lot, so I'm going to be
making sure the ratio between revenue and loss is not like 10 million
revenue, 5 million loss.  I would like those ratios to be reasonably
sensible, 20% max, on an ongoing basis. We would want to minimise the
sort of losses we're making.  The only reason I would change that
strategy is if I thought it was absolutely crucial that there was heavy
market penetration and that we basically just bought the market share. 
Most companies that do that actually never turn round, or they get
bought.  It's a good strategy, if you're into making money and that's
the sole purpose of your life, then doing that and getting bought out is
good.  But it typically would damage the industry more than anything. 
So we're looking for a sense of direction.  We'll continue to invest
very heavily in technology and we'll fund that technology through top-
line, headline revenues.

TWST: Is there a time scale to consistent profitability for you?  

Mr Lancaster: We're looking at three years down the track.  Some people if
they're losing masses, 25% loss, say they're going to be profitable next
year, and I just don't believe that.  I'm being a little bit more
realistic and saying I'm not telling you I'm going to be profitable next
year, but then if you look at the losses that we're making they're
pretty small, and you can see that the history speaks the future.  So
within three years we're looking to be profitable.

TWST: When you speak to the analysts, what sort of concerns, what sort
of questions are they putting to you? 

Mr Lancaster: I think all of them understand the need.  I think it's, how will 
the market evolve; will we become acquired by other companies - which is 
not a problem to them, but is it going to be bigger fish, an IBM or an Anderson 
Consulting that will acquire us as part of their portfolio, or is this a market that
will be serviced by people like us.  They're also keen to understand the
move from whether we're service or product company.  But most of them
are fairly comfortable that that's a pretty good strategy.  That's why
we're on the valuation that we're on.  I'm sure if I said I was going to
lose ten million I don't think they'd lose a lot of sleep, although
there has been an awful lot of change between attitudes from analysts. 
Analysts have gone through a very turbulent period in these last four
months, where it was, everybody's got to invest in the Internet.  If
you're not losing money you're not there. It's almost that we weren't
losing enough money.  Now the mood has changed from analysts, and it's:
what is the business model?  So, what's the model behind it?  Is the
model really workable?  Will the product really be successful?  Is it
really needed?  Does it really provide a solution?  Those are their
concerns.

TWST: How's the stock market doing in representing the value of the
company? 

Mr Lancaster: I think one has got to be very careful how one
values the internet possibilities, because if you were to mention some
of these business to consumer companies, for example, QXL, LastMinute,
and the like, even at the values they're at now, I would say, hold on a
second - that seems incredibly high a valuation relative to my
valuation; whereas if you were to compare SDL with a bottling business
in )you would see that we have an incredible track record of consistent
growth for eight years, proven acquisition ability, proven ability to
ship products and global organisation.  That's very important.  But all
this stuff about will dot-coms go to the wall - the answer is, no they
won't.  Some will, but what will happen is that the shareholders will
force senior management into those businesses.  But if you go to visit
them right now, you'd see three or four floors of a lot of PCs kicking
around.  There was a student who got his web page bought for GBP10-15
million, which is not a lot of money in City terms but it is a lot of
money for an individual.  And so I would say based upon the
infrastructure and the people that we've got and the markets that we're
in, I'd say probably there's significant up side in value, because
there's evidence that it's not just an idea, it's that those ideas by
that management team that have been turned into reality before.  And
then you look at the strategy and people do need globalisation and the
services are required, and it's a massively growing market, and there's
a track record to actually produce it.  So there's a market need for it,
and I would say yes, it's undervalued.

TWST: Thank you.

MARK LANCASTER
Chief Executive Officer and Chairman
SDL Plc
Butler House
Market Street
Berkshire SL6 8AA
United Kingdom
Tel: +44 (0)1628 410 100
Fax: +44 (0)1628 410 505

Each Chief Executive who is the featured subject of a TWST Interview is
offered the opportunity to include an Investors Brief or other highlight
material to be provided and sponsored by and for the company.

Copyright 2000 The Wall Street Transcript Corporation
All Rights Reserved


The Wall Street Transcript (TWST) interviews are published verbatim, and TWST does not in any way endorse or guarantee the accuracy of any information or opinions expressed herein and all opinions are subject to change without notice. Nothing herein constitutes a solicitation to buy or sell any securities. TWST interviews with CEOs may include include "forward-looking statements", which are based on factors that involve risks and uncertainties. Actual results may differ materially from those expressed or implied. TWST shall have no liability whatsoever for any trading losses arising out of use of this information. Copyright 1999 Wall Street Transcript Corporation. All Rights Reserved.