Ed Whiting – Lego, Meccano and Sticklebricks

Companies such as Comtec provide a travel meta search-type service to our clients, the travel trade and its customers. We send user requests for travel to several suppliers simultaneously and we present the results in a single comparative display. All straightforward enough – or is it? The title of this article is intended to highlight…

All straightforward enough – or is it? The title of this article is intended to highlight the significant underlying challenges and obstacles that hinder providing this kind of service efficiently.

Providing a quick, accurate response to travel enquiries online is akin to building a Formula 1 toy car from a mixture of Lego, Meccano and Sticklebricks. Not altogether easy – due to the convoluted layers of systems, interfaces, bandwidth restrictions and standards that power those beautifully designed websites.

The result is that the industry finds itself having to make significant trade-offs for usability and performance reasons.

To help illustrate these complexities, let’s take the example of a dynamic packaging search on an operator’s website. Fulfilling this request will require searches against several layers of suppliers. A bedbank may be consolidating content from several accommodation suppliers. Several types of airline inventory may be queried – scheduled, low-cost and charter. The speed of response to the user will be determined by the slowest result through the layers. In other words, the chain is only as strong as its weakest link.

However, the connectivity issue is just the routine part. It gets more complicated when users want a price comparison-driven search. This is where we get into the area of trade-offs.

The challenge with pricing is that many suppliers want to use yield management techniques, such as dynamic pricing, in response to supply and demand. There is also more than one pricing model – low-costs cut prices near departure but charters increase them.

The first trade-off is that the only way to employ dynamic pricing is to conduct searches in real-time. However, users want fast responses and due to the complex layers real-time searches can be slow. The best way to provide speed is to cache both availability and prices. Cache is where frequently accessed data can be stored for a period of time by an intermediary for quick access. The risk is that data may not be up to date at the time a search request is made.

Another trade-off is often made due to the fact that price comparison is far easier for simple travel products – due to issues such as layering and different pricing models. This is therefore the most common approach taken by price-comparison websites. The trade-off here is that consumers increasingly value flexibility and want to package their products in different ways. They may want to compare a pre-packaged holiday with a package created dynamically using a low-cost flight and bedbank. Aggregating the right products in these circumstances becomes more complex, slow and less accurate.

Even if an intermediary at one level in the chain is clear about which trade-offs are required, it should be remembered that another intermediary or supplier may have made a different value judgement. As we can’t control whether caching or real-time searches are made at different levels we end up back at our point of weakest link all over again. The challenge now facing travel companies and technology suppliers is finding ways to reduce and eliminate the trade-offs. This is the business Comtec is in.

Through experience, clever technology and by developing effective industry partnerships, real progress is being made. The end result is that the search offered by Comtec is faster and more accurate today than it has ever been. However, reducing trade-offs is a process, not an event. We’ll therefore continue to focus on tackling these challenges tomorrow, next week and next year.

Ed Whiting is product director for Comtec (Europe)