Let’s work together

15 July 2008

Getting IT systems to work together is a major hurdle but managing the global design teams and process is the biggest challenge writes Rob Jamieson

Over the last few months I’ve been involved in looking at large companies and how they go about designing products. I suppose some of what I saw I expected - that is in some cases there was a good level of integration in design systems and in others a compete lack of. As with any large company there is a tendency for a level of internal politics to interfere with processes. This could be due to the structure but often it is geographically related, a case of different people from different countries just doing things in different ways.

So if you are a big multinational company with multiple teams how do you manage them? Often there is global design that appeals to a global audience but with teams that are used to designing for local markets. When a large company tries to take a local team and target a new market it does not always work. British motor bike manufacturers in the 70s targeted the US market but the average US male was a little taller than the British male. The term “toy bikes” was used by the targeted audience in the US as they thought the bikes were too small and the campaign was not a success.

How do you guard against this sort of mistake today? Forming a global team by recruiting people from all over the world and placing them in one place is very expensive and often over time you lose the people’s connection with the local market and hence their original advantage. You can use technology to bring a global team together and this is the absolute minimum you should do. The problem is that due to acquisitions over the years often big companies are made out of smaller ones and for historical reasons use different design packages. Data exchange is better today but you generally only get the base geometry and not the design intent. Yes you can edit the base geometry changing hole sizes but if the model was linked to linked parametric tables this never comes across the way it was intended.

As soon as you perform a data exchange the data is locked and out of date. Many notable designs have fallen into this trap which can cost a company a lot of money. Physically sharing the data at a regular interval is a major challenge. Today there are a lot of file viewers with abilities to view geometry but an export is needed to produce this and by its very nature it’s a process that needs to be done by the designer so the design intent is delivered. Of course, all of the team then needs to be able to view this data and often just viewing a 3D data file is not enough. This then entails a presentation which sometimes is made by the designers but often by other people translating where they think the team is at that time.

Design systems don’t make mistakes on design; it’s the designers and the leaders not under-tanding the issues

One of the biggest issues is where you have different teams from different disciplines exchanging design intent. If you are a stylist for cars you would start with a 2D sketch, then perhaps move to a 3D stylist package, then to clay. This data is then passed to mechanical, aero dynamists advanced surfaces etc. These teams then change the design to suit their own ends. To gain the packaging requirements and aerodynamics the original stylist’s design is often then compromised. I’m not saying this is wrong as getting things to work is essential, but you often see concept cars with radical design ideas but when the car is made the design is a lot less radical. Sometimes this is for practical reasons but how much of this is because it is designed by committee?

Each geographical region has its own requirements. In the case of motorbikes in the 70s it was size, today it might be down to which side of the road you drive. Door, and mechanicals placement are affected by this, and even today not all cars are made for both sides.

Each region will have its focused requirements such as densely populated countries like smaller cars but will also offer their own talents. For example, we know the Germans and Japanese are good at car engineering but for styling UK and Italy are strong. In any global company playing to the national strengths is always a good way to get an iconic design. No computer system is going to have the knowledge of country capabilities but often the most iconic designs are led by a single minded individual. In some ways globalisation has diluted some of the pure ideas we once had in small design teams of the 60s.

Of course, money is important and design needs to reflect this fact so the product makes a profit, but at the same time bean counters might not understand the criteria for design as much as they should, so should they lead a project?

I think what I am saying is that design systems don’t make mistakes on design; it’s the designers and the leaders not understanding the issues. Looking at design first then managing how the design moves to prototyping based on the assets you have in the global village is the challenge. Companies that do this will have strong products and survive any over lending by bean counters in other countries.

Rob Jamieson is a marketing manager at AMD. In case you don’t believe that 17-inch laptops and standard class don’t mix, just ask the poor guy sat next to him on BA0285 a few years back. This article is his own opinion and may not represent AMD’s positions, strategies or opinions.
.(JavaScript must be enabled to view this email address)

Comments on this article:

Leave a comment

Enter the word you see below: