[back]

The New Travel Intermediaries

Speaker Transcripts

Gary Hance
Director
Seaforths Travel/ Ticket Window

My name's Gary Hance and I'm Director of Seaforths Travel. Seaforths, for those who don't know, is a UKL40 million a year business travel specialist based in East Anglia. We also operate in Aberdeen and in Bracknell. Traditionally our business has come from the North Sea oil industry and as of now, our customers come from all aspects of corporate travel. What I'm going to do this afternoon is give you a four fact story on Ticket Window. Why we did it, what went wrong, what it does today and where we think it will go.

Let's go back about four years to the dim and distant past. There we were sitting in the office one day, well actually we were sitting in the pub. In Aberdeen, where I'm based, our business works 24 hours a day. We were pretty sophisticated. We actually had a mobile phone. Not Hoggs, they actually had people working the office so we had to do something to get a bit of advantage back. We had the next best thing. To start with we get into the car or taxi and head off to the office. Now for long haul destinations people like Tunisair, Nigeria Airways, 5.31, they're gone. All you can do is go back to the office, fire up your system and go to the airport. That's what we started to do. Along came an anorak with some interest in computers and what they could do. I had put some word processing software onto a Sabre PC so I fancied myself as the next Bill Gates. Of course, until Sabre started getting aggressive in the market and putting in proper PCs, Galileo had made do with dumb work stations. With the appearance of the Sabre PCs, Galileo came up with the plan to install PCs in agency offices. These duly arrived along with the pirated Grand Prix games, golf simulations and other dodgy software. Also, we got some legal stuff, fares databases, word processing and so on. Despite the limitations which Galileo put on agencies breaking into their expensive Olivetti 286s, we did manage to open up some and even install PC Anywhere in a network to communicate with the on-call person. That worked and although we've replaced PC Anywhere with a dedicated Galileo emulation with a host and remote client, the principal of one PC taking over another still operates and is still quite successful in the business travel market. In fact, we actually sell such an emulation. It's UKL500 and there's the order form!

The quantum leap was taken when we decided that if we can use Galileo, why can't the customers. That was the start of Ticket Window. We told Galileo what we wanted to do, they gave us the specification and the communications protocols and quietly laughed up their sleeves, never expecting to see us again. Around six months later we had the first working version. Duly impressed, Galileo certified it. That's it. It looks like a viewdata screen from here actually. We did make the mistake in that we originally launched the system as a character based product in order to have the maximum potential users. Originally users didn't need anything more than a PC and a modem. We didn't have to hand out any software. All they needed was a terminal emulator, proprietary pieces of software or shareware that came with the modem and then they could go into our system. We thought that would be the right way to communicate with the great unwashed rest of the anoraks out there as well as the corporate market.

Don't get me wrong. The mistake wasn't in the way the program was written or in how it performed, the mistake was in failing to realise we needed to make far more of an issue of the system and everyone was expecting something based on the Windows operating system. This slide is actually supposed to be Bill Gates planning for world domination but legal advice told me not to put it up there so this is Ticket Window's plan for world domination instead.

What were the issues the original system came up against? First of all, our naivete I have to say. Here we were with what we thought was the greatest thing since sliced bread. All we had to do was sit back and wait for the millions to roll in. It didn't happen that way. If you build a better mousetrap, believe me, you need to allocate at least half a million pounds to launch it, market it and get the brand known around the world. And even then the world probably won't beat a path to your door because there's a better mousetrap. We made at board level a conscious decision not to make Ticket Window available to other travel agents. This also proved quite interesting and we sat in the office with a well know industry figure in the city here. He had a travel budget in the UK of 20 million and worldwide of 500 million. After the demo he said: would you make this available to me and my already appointed travel agents? After I said: no, it's for Seaforths' customers only. The interview didn't last much longer.

If we converted 20% of the business then I'd be rich now and probably be sitting in the Bahamas not here. Unfortunately we had two problems. We had a product that was good, it worked and it was better than anything else our competitors were even talking about. This is a few years ago remember. One of our customers, particularly here in London, didn't know where East Anglia was far less who Seaforths were. The second problem was that Ticket Window as a concept at that time was so radical that people did not believe we could actually deliver it. At one major city bank the Managing Director spent five minutes looking through the CD ROM that we were using to actually produce the live availability on a flight from London to Tokyo. He only believed that we were live when we pulled out the phone line and the system stopped responding. We did actually get that business but we didn't get the one where we sat in the travel purchaser's office of a large South London computer company. Right next door was the implant from a major travel agency. It was 11.00am and he'd supposedly spent the morning trying to book a hotel in London for an important guest arriving that night and failed. Using our system, we got the travel manager herself and booked a hotel within five minutes. That was the only booking we ever got from that customer. So even when we can prove it works we couldn't always crack the account.

We decided to go with the flow and change the whole thing to a Windows application and started that process early in 1996. In retrospect it's easy to say that we should have gone with Windows from the start but that would have delayed our launch between six months and a year. I think that the operational lessons which we learned in the original system would have put us much further behind now. Also, financing the programming with absolutely no revenue would have been hard.

Let's talk about finance for a minute. I did say a few moments ago we had two problems. We actually have a third. The programmer who actually does most of the Ticket Window software writing was employed by a PC manufacturer who expected a far faster return on his investment than Ticket Window was able to offer. From a position of a 50/50 partnership, which was the only way I could sell it to the Board in the first place, it simply came clear that Seaforths was to take complete responsibility for everything. We did that, we bought out the partner and financed the complete development. It did cause a few sleepless nights, particularly for me, as I'd sold the whole thing to the Board in the first place. So briefly that's the history, why we did it and what's gone wrong.

So now the Windows version. It's finished. It was completed last September mainly using Delphi programs and Windows NT. It's a client server application but for those who are not technical, a two-part thing. One part resides at our office - that's the server - and the other part is on the user's PC - that's the client. The client serves as an Internet browser. That means that customers can use the Internet as a communications method without allowing their employees access to the Web. This really is the opposite of what was being said earlier but I'll carry on anyway. It's fully encrypted with a constantly changing key which means the client and server tell each other which encryption method they are going to use and change it regularly during each session. This prevents unauthorised access. You may wonder, those of you who wonder what I'm talking about, about the wisdom of producing a client server in these days of all conquering Java. There's two-part logic behind this decision. Firstly, producing a client means that lots of the screens and images can be downloaded onto the user's PC. This saves considerable time over the normal Internet method of sending every screen every time you connect. As Simon was showing earlier with the graphics, it does take time to download graphics. We only send the data to populate screens using our own data-compression.

The second point is with the World Wide Web itself. Customers are concerned that their employees, given access to everything on the Web, will spend their time at playboy.com or whatever the female equivalent is and I guess from the previous presentation, it must be cosmopolitan.com. Using our client, the customer gets the benefit of Web communication at a low cost without the associated risks of time wasting and net surfing. So what's the disintermediation factor? Incidentally, Microsoft Word 7 doesn't recognise disintermediation - it's a spelling error. So that's what the disintermediation factor is. Why is it worth contacting Ticket Window instead of going direct? Someone asked me this on the panel in Seattle last year. The question was why should customers come to your site rather than go direct to the airline's own site? The answer was the same then as it is now. Airlines have for some considerable time offered direct access to their services. All the user needed was a low cost, easy to use communication device. The airlines would even pay for the call. Travel agents had to survive in the past in the face of direct competition from the people they're supposed to represent. They'll continue to do so electronically by replicating what they do on the streets to compete. Add value, offer alternatives, suggest back-to-back ticketing, point out that the frequent flyer programme which the employee wants to use means the company is paying twice what they have to for flying that airline, provide MIS which doesn't just cover that airline's services but all the other airlines the customer uses also, confirm wait lists, suggest and implement a travel policy, negotiate route deals and introduce alternative suppliers. All these things we can do via Ticket Window because it's a travel agency system. It's not a simple access to the CRS, it's a filter and although it makes my staff much more productive in dealing with the customers, they are still there, quality checking bookings, suggesting ticketing and airline options and generally continuing to make themselves indispensable.

So let's go over a few of those extra options again. The things that we do which will ensure the client goes to my service and not that of the airline.

Fares: we freely publish consolidation fares to all users. We haven't got 118 airlines that Ann has but we've got about 35 which we put on the system. Not only do we offer economy class but again we offer business and first class as does Gold Medal. The fares are actually displayed in a seamless, cheapest to most expensive way. We actually take in the information from the IATA database on Galileo and integrate that into our own net fares database and present the customer with the database where he has no idea where the fares have come from. They can be net fares, published fares and they can read the notes and decide which regulations and which fares apply to them and go from there straight to find a seat on an airline which suits that particular price.

Management Information Systems: really, we think this is the killer application, the one that gets the most notice when we're in sales situations. We've developed a program which allows the customer to see at the touch of a couple of keys who's travelling where, when and for how much. This is ideal for route deal management, for policy compliance and also, of course, for financial managers. It's also great for individual travellers to check their frequent flyer programmes. Most recently, we've further developed this idea to include an off-line MIS system which will reside on the client PC and allow them to run local reports, pre- and post-trip, using information downloaded when they log on to Ticket Window.

This really is the difference between our system and many other systems. With the pre-trip information the client can select from a very comprehensive set of criteria and decide they want everyone travelling to New York next week, everyone travelling to New York on BA, everyone travelling to New York on BA flights 1,2,3,4. They can be specific or as wide as they like.

Travel Policy: more flexible and more enforceable than the CRSs own systems, our travel policy system allows the administration of the client company to state for individual users exactly which airlines can be used for any city fare or any country fare. It can also be set up so the user cannot overwrite policy or if the user's allowed they can overwrite the policy. A couple of examples up there - from Aberdeen to Bergen I can only use Air UK and I can't overwrite the flight. From Houston to LA, which is number 7 on that list, I have to route via Phoenix. Although there are direct services, they won't show on my display and I must use America Airlines and number 8, London to Milan in Italy, I can only use Alitalia.

Other Features: rail ticket ordering, non-CRS car hire and foreign exchange. Because we're acting as a filter for the CRS, not simply an access method, we can add much more to the front end data. Effectively we're creating an e-mail system allowing the user to communicate electronically with the travel agent.

News and Information: we can tailor the way that we present the news pages including a client company travel policy, how they allow expense claims, special news on their own specific route and destinations and much, much more. This can be specific for any customer. Effectively this creates an Intranet for the company to use internally with its link to the outside world allowing full access to the system. There's another benefit too. One that we talk about a lot when we're in the sales situation. That's the unbiased view which the system gives. Call an airline, either by phone or on the Internet, ask them for the next flight to New York. The flight they offer will be one of their own, or at a stretch, one of their partners or co-shared airlines. It's highly unlikely to be the service of one of their competitors. With Ticket Window, unless the client has already defined a travel policy, he's sure of an unbiased response and knows whether he's calling the travel agent or the airline, he's going to get every flight, not just the flights that that vendor wants to sell.

I've connected to the Internet's ISP and my software, the client, will only access it in Windows. It doesn't allow the user to go anywhere else. So we're now live on the Windows system and I'll go to flights and availability. Let's say we're in London and we're going to Oslo. If I'm a user who has no particular travel preferences then all I need to do is put in my departure airport, arrival airport and the date I want to travel. You can see that took about two seconds and we've now got a live response back from the CRS with all flight options from London to Oslo. Let's say we want more information on the BA 762 and click on that flight. It tells me that it's a 757 200 series, it takes two hours and there's at least nine seats available. If I want a seat map from BA, I wonder if I can do that. I'm trying for a seat map for business class and you can see that row 3 is already sold out and row 4 has A which is by the window. To make a booking, just click on the flight. It goes to my name as the default because I logged in. It's obviously designed for the corporate travel market so I can actually list all my travellers up here. Let's say I'm taking Bob Houser with me so I put in Bob's name. Put proceed. Now my specific profile has been set up so that I can put in a purchase order number. The benefit of this is that it keeps the management information so that when the customer reads their MIS later they get all this information back. Let's say we want centre economy and click okay. We'll book our return on BA again. It knows whose travelling so it goes straight to the flight and we confirm that. One of the main features of the system is that it will re-book or offer re-booking to the best available fare. It doesn't matter which class the customer chooses. They can actually see what the best available fare is. That's what it's showing us right now. Economy is UKL424, the best is UKL205 and I can read the notes on the lower fare or I can just re-book with a single key to get into that better fare. If I wanted to read the note and see what the cancellation fees and so on are, I would just click on there. All the Ticket Window screens can be printed out at any stage. Because we're actually replacing the travel agent in this part of the loop, we have to make sure that the customer is aware of all the regulations as far as passports, visas and so on are concerned so on every booking we do put up errata pages and let them read information as to health, visas and so on. This is the screen it's showing me now. If I wanted to read the visa notes I would simply click here and up they would come. For the time being I'll click confirm. It's asking the same thing for the second passenger. If I went over here and clicked, this passenger needs some help with passport application, visa application. I click up here and it shows me the itinerary, down here I can print out the booking on my local printer so it saves the travel agent having to fax me my itinerary. Literally, you can do everything you do in Galileo. If you change your mind and you want to change the date, pre-book your seats, pre-book a particular meal - all the meals are in there. Services are the same story. You get all the stuff you get from Galileo - wheelchair, sports equipment, ships' crews and so on. You then go to end booking. You would normally get four options which are store the booking, send it for ticketing, send it for ticketing on departure or reservation only. All those have the same effect, they all queue off to the travel agency service desk with different queue numbers for each option and then obviously a travel agent picks up that booking, does a quality check on it and decides whether it needs ticketing, whether the client needs a phone call to say have you thought about back-to-back, whatever.

We could do a lot more. If we decided we just wanted to see a fare and then decide whether we had enough money in our travel budget we would go to the fares menu rather than the flights menu. The timetable simply gives all the options from A to B, direct connections and all the flights that are going from a particular airport. Let's book a hotel while we're in Oslo, shall we? I could have actually done this when I was booking the flight. The customer can fill in just the basics which is what I've done. If they wanted a particular hotel they can do that also. They can fill in a location site which would be downtown, airports, resorts, etc. and they can also put in the price they want to pay. All this is doing again is what we really do. It takes the Galileo screen and codes and makes them easy to understand and work with. Let's say I'm interested in the Park Hotel. Incidentally, its default is my corporate rates but it could search on for a specific client's corporate rates or just for general RoomMaster rates. Using RoomMaster it's giving us more information. It's one mile from the airport, there's the phone and fax numbers and check-in and check-out times. At least with this the customer gets the chance to use his own criteria to select the hotel he wants and of course he isn't wasting the travel agent's time while doing this.

The rail and foreign currency requests are all just fill-in forms for the customer to actually say I'd like this train ticket to Gatwick and this is the day I'm leaving and coming back. They don't do anything more than create remarks on the Galileo PNR which comes to our office and basically we've got a very cheap way of electronic mail from the customer to us.

The profiles obviously allow the customer to personalise the bookings towards his own preferences if he's got particular airlines that either he or the company have selected then they can be built in there plus his frequent flyer numbers and so on.

Let's have a quick look at the travel manager system for management information and call for a 90 day search starting from today. Now you can see the power in this database search system. You can put in the surname, destination, airline or anything. You can have all bookings or you can just have, for example, bookings that they haven't sent off for ticketing yet. This would be useful for the travel manager on a Monday morning looking for bookings for the next week who they haven't had tickets for yet. They can actually transfer this data to a local machine. We can do a search for between today and the end of September for people travelling to Oslo. This original screen gives you the routing. There's dates for each sector that takes the booking and same information with the airline and the classes. The same information for the dates of each sector. Created dates, ticketed dates, departure, return and cost centre. Reasons for travel. This is the fares and savings report so that we can see on number 8 I booked for UKL441 but I could have travelled for UKL222 so I've got some explaining to do to the financial director. Then number 9 is the summary. Eight bookings, 20 sectors. That's what I booked and that's what I could have paid.

Now I'm not the travel manager or the accountant anymore but an assistant in the sales department and I'm wondering about going off to Dubai. Let's find out when the public holidays are in ? so we don't have to send our salesman off on a public holiday. This is not rocket science, it's just taking information that's already in the CRS and presenting it in a way that's easy for the customer to understand. In my travel agency I've got 40 or 50 staff all using Galileo, all very good staff, but if I said to them now find public holidays in Dubai, they would fall off their chairs laughing. They really would have no idea where to start. They'd know they were in Galileo somewhere but it would take them a long time to get there. This is a good one. What should we wear in Dubai? As you can see there is much more to this. Passports, visas and so on.

All these features add to the basic proposition which is no risk electronic travel. There's a travel agency behind to stop them screwing it up.

So what's next? Destination related to content, more links to other Web sites, trawling of other Web sites and putting them on to our local hard disk so that those people who are using our browser and can't get anything else, can still get more information. For example, the CIA World Factbook is a good one. I don't know if you've ever used that but try it. It's got very detailed information, including maps and lots of other stuff on every country worldwide. The World Travel Guide on-line which AT & T is close to is also extremely good. Rail times for the UK at railtrack.com. Rail times for Europe at the Deutsche Bank site. Latest airport arrival times at the ba.co.uk site. These are all live sites which anyone can access using the Internet. How about real destinations and specific content controlled by us. For example, if a client books a seat for Singapore through Ticket Window. Up pops an icon offering a list of restaurants in Singapore, city map and a range of hotels. Their are revenue potentials here too of course. The hotel in Singapore might pay only a few pennies every time that's shown but it's much better for him to target that advertising than spend $10,000 on page 3 in Time.

So in conclusion, the new intermediaries in the travel equation could, and in my opinion should, be the same experts who have been intermediaries over the years. Travel agents have their contribution to make but they will have to keep working to prove what they bring. A simple London - Frankfurt booking for a travel agent whose done it 20 times before will provide little opportunity for the agent to shine. But the provision of MIS, tracking of route deals and comparison of suppliers will be the value that they can bring to that booking. If it's a more complicated booking with 10 sectors around the Far East and the Middle East, then the agent will have more of an opportunity to add extra value while still tackling the MIS, route deals and so on.

In my opinion, agents in the next few years will need to either spend their own money creating uniquely branded technology like Ticket Window or spend their money buying other people's technology. Perhaps the second solution is cheaper and less risky but it provides no USP. A major London agency recently announced the launch of their on-line reservation facility. It's booking engine is actually the Sabre site in the US which can be accessed by any agency using Sabre and being prepared to pay the fee for an overbranded front page. There's no USP there I don't think.

Our experience with Ticket Window has not been easy. Convincing people that Seaforths can provide a safe and reliable service was, and still is, a hard job but it has succeeded and it's provided Seaforths Travel with a clear future which a lot of agencies have yet to find.

[back]