Every product planner should read this

You know that the only reason I wrote this is to finally get those features in my tablet running, right? ;)

And yes, while many product planners probably will go and say "naa, I don't need that, we are not going international" they do neglect the fact that 'we' do shop international now - once that paypal and credit cards took care of stupid international banking fees and companies like UPS, DHL etc did bring us goods in few days (still overpriced, but working on that).

It was difficult to plan on selling world wide a few years ago, but not anymore. The earlier you start living with that fact, the sooner you can start thinking about expanding your business.

I lived in Japan a few years ago. Email is much less popular there than in the U.S. While some of that difference is probably due to the fact that less people there have computers at home, I think it's also that the standard computer keyboard is made for typing in English. While you (obviously) can type in Japanese, it isn't as fast (efficient) as typing in English which makes email take longer. Also mobile phones there have greater functionality and I found it easier to compose messages in Japanese from my phone's interface that from a computer keyboard - I thought it was easier to type and select the words I wanted to use.
Just a thought...

Software planners (at least at Audi, MS, and MOT) are generally aware of the existence of non-US markets, and the feature sets needed to support these audiences. The issue isn't awareness or intelligence, it's about opportunity, prioritization, and costs.

Lots of sw developed in the US is, naturally enough, initially focused on the US market. The pressure to get to market with a v1 offering is particularly intense, and feature/quality/timing considerations play a strong role in 'internationalization' features being punted from v1 in order to enable core features to make it in to product. The ususal line of thinking is 'the underlying platform has some basic internationalization support, once the demand in non-US markets picks up, we'll be able to build off this base.

'Internationalizing' a product after v1 has left the gate is usually pretty tricky, with a significant feature cost. In many cases, adding support for other languages isn't as easy as just hooking into language support in the plaform.

This creates a catch 22 situation, where support for non-English markets means jumping over a significant feature hurdle, and competing with the top 3/5/10 feature candidates in v2/3/... and these top features will form the basis of the value in these subsequent releases.

In a competitive market, your entire customer base may be at risk with each release, so making sure you keep your current customers and win new customers in your currently served market is usually the primary goal of a relase. Here's where the catch comes in - the US (and English speaking markets) are large, and attractive. If my product has to 'win' with each release, before I expend dev resources on new markets, I'm going to make sure there's enough new value to satisfy the current market. In addition, winning new customers costs money, servicing existing customers is much less expensive.

Winning Germany isn't a win at all, if in the process I lose share in the US to competitors. Winning China might be a net 'win', but only if the addressable market is truly larger than in the US - but with fixed dev resources, this would be a huge gamble.

Fortunately, product teams with successful products generally get larger over time, increasing the likelihood that internationalization and localization features make it into the release.


I can empathize with folks in other countries that don't get to use the latest/greatest English-only software - not being able to use the latest/greatest products in your native tongue must be a real pain.

As an FYI, MS has done a great job of enabling really broad, community-based localization with the Local Language program: http://www.microsoft.com/industry/government/Lo...
There really aren't many vendors out there that enable customers to create their own localized versions. Kudos to MS.

And don't forget that international doesn't just mean a nationality other than American. For a significant group of users, being able to do things in more than one language is important. My father writes in four languages, sometimes in a single document. For him, WordPerfect is still a massively better word processor than Word, because it is much more straightforward to configure across multiple languages. Native English speakers are usually monoglots; native speakers of other languages think it normal to be polyglots.

Something as simple as a form on a website can be an insuperable obstacle: there are still far to many sites which try to force my address and phone number into US formats which are irrelevant to me.

Jason: Absolutly true. It is why I usually take the example of "if you built a house with out a basement, there is no way in hell you will built one later, because it is so expensive". And it is okay if you are snobish and say the small US market is big enough for you.

Especially when you have to compete for every time you release something. Again and again. Why look outside where the others are not competing? ;)

And Marek gives a good example too - I notice that I have to force myself to just German or just English even when I take my own notes - because the processors really don't like mixed languages. This is okay for the moment, because I do realize how much effort must have went into programming this logic.

And set aside bashing others for not doing their job: Microsoft Office (more to the point Word, Excel and Visio) are one hell of an example for working localisation - not just translations, but working localizations. There are some minor things if you want to exchange csv files but in general, this is a such a hidden treasure you would not believe it.

Well most people won't as they don't use it that much or see the hidden beauty behind it. Word's spell checking for example is one of the very very few which actually work at German.

Marek, I don't know if your father knows / uses this, but he can mark parts of the text and assign the language to this, so spellchecking and everything around it will work flawlessly. You could assign buttons with mini makros for them. Makro recorder - love it.

Eric Schwiebert has done a nice post about difficulties in this area - and no contact form as well as I don't log in for comments, so let me give a manual trackback:
http://www.schwieb.com/blog/archives/16

Robert, as your blog is in English you can at least expect that your readers are not kept away by language issues. In terms of half-baked technology solutions there are just way too many examples. Here just a couple of them:

- address books: different country, different address standards. Now try to keep an address book where addresses from different countries are stored and should be printed correctly for each country.
- online forms: different countries, different languages, different address formats, different phone number formats, different date formats should be automatically supported and then normalized on the back-end
- remote speech recognition (e.g. IVR systems): There is simply no speech recognition system that can function properly when it doesn't even know upfront which language it has to deal with
- character set issues: nearly 20 years after the first Unicode drafts the problems still persist. It is not as bad as it used to be, but still noticeable

The big underlying problem is that variables are treated as constants, which reduces the technical challenge, but creates half-baked products. One part of that problem is that a lot of developers don't even realize that what they treat as constants are in fact variables.

It's more than just INTERNATIONAL. It's cultural, regional...a million differing reasons. Marketing is WORK, everyone (yes, EVERYONE) is differing and has varied needs and wants, carte blanche, labeling it all "international" and grouping segement of people, misses a million more logistical underpinnings. But international support is a big game of chess, winning other markets, could lose in markets that matter...no easy answers there.

Well I guess you could say, that Geekdom is really a foreign culture to the majority of the population, and edge-casing will eternally spin circles. OK, yes...

"What kinds of international differences are you seeing that keeps you from adopting technology? "
The VBA CSV date bug (whereby it forces dates into US mm/dd/yyyy format) still isn't fixed in Excel 2003 and its been around since at least Excel 97. I really hope its fixed in Excel 2007. It doesn't stop us using Excel, but has caused days of work unpicking financial transactions that have gone into the wrong month.

The smart companies will be making sure their software works in China ready for when its economy overtakes the US.

Some examples I've dealt with recently include:
Licensing - I can't download some stuff from iTunes without a 20.x.x.x ip address
Support - Sandisk won't let me order memory cards from the US
Promotions - I never did get that USB key from the Microsoft Licencing promotion

I'm in the business, iI can rationalise why these problems occur, and why vendors have to do these sorts of things. Their problem is they're ignoring my bank account as well, and its not as forgiving as me.

Of course I can rationalize them as well and see arguments for it. I am just fascinated by the way some reactions pour in like "all our customers have javascript turned on! this is why we can make a heavy stupid javascript site!" - "well, if they don't they cannot visit or shop, so of course they have, but how about all those customers left outside?" - "they are insignificant numbers"

And all of that without knowing about how many are left outside. ;)

Additionally: As those bank accounts are ignored by so many, it is much fuller than your average buyer in the US ;)

From the UK ... dates, yes!
But for me the most disruptive aspect now, and maybe even more so in the future, is the apparent assumption that users will have a working Internet connection, and even more so that it will be broadband. Not so, geeks. The Internet may be your lifeblood but for plenty of regular people a fast connection, or even any connection, is not a priority or even possible.
You are dividing your potential market into two parts - or even losing half of it.