much ado about the 1099 worker

I was at a dinner last night hosted by Button (an amazing company, go check them out!) about the on demand economy. Of course, the topic on everyone's minds and a major source of discussion was the Uber ruling.

In case you missed it, the California Labor Commissioner's office ruled that a driver for Uber was an employee and not an independent contractor. It subsequently said that Uber should pay her expenses to the tune of $4152.20. (see SFChron or NYT)

I was planning to write up my thoughts on this but at this moment, I don't think I'd write something markedly more original or thoughtful than several of the opinion pieces I've read. Instead, I'm going to present my personal roundup of the best articles and opinions on this issue:

I also highly encourage people to read the official rulings of judges on the separate but related potential class-action cases against Uber and Lyft. You can see them at https://docsend.com/view/y4v488s and https://docsend.com/view/6spj76v.

going global, part iii

Quick recap: to scale internationalization (i18n), you need strong tools and processes in place. These posts go through an overview of those tools and processes and how to approach them. 

The first post of this series went through expansion strategy. The second post covered content and communication. In this third and final post, the focus is on some areas of operational i18n that impact product. At the end of the day, every part of the business is impacted by i18n. I'm focusing here on the ones that I believe need the tightest collaboration and providing some examples of what to think about with regards to:

  • content
  • legal
  • payments

 

CONTENT

I'm referring to content here as structured content that is part of running the business. For example, every market in the world has different car manufacturers, categories and models. A van is not the same thing in every market in the world. That means as you open up new markets, you have to bring in new content sources. For a long time at Uber, we had people manually entering this information and we made a small optimization to simply duplicate original data when creating new cities in the system. Of course, this was not particularly useful for our international cities because the base city that we duplicated was San Francisco. Oops. If we had to do this again, it would have been particularly helpful to rely on standardized sources of information and proactively update that with new sources relevant to the expansion markets. This data will vary by business but as you think about known data, consider what those sources will be and how to get them internationally. Some examples to consider include:

  • financial markets (stock symbols)
  • food (recipe ingredients, measures)
  • addresses (postal code formats)
  • telephone numbers (formats, country codes)

Work closely with your operations teams to understand what data they need and use on a daily basis to run the business. Even if these data sources aren't integrated into UIs from day one, having the reference will ultimately lead to cleaner data in your systems.

 

LEGAL*

*It should go without saying that I am not a lawyer and this is not legal advice but I'll say it anyway. Please don't take this as legal advice in any way, shape or form.

Regardless of industry and market, there are regulations which require compliance. At the most basic level, that applies to terms and conditions (Ts&Cs) and privacy policies. Corporate structure can also play a role in this, i.e. where your company is headquartered according to your articles of incorporation. For example, if a user with a US credit card who legally resides in Canada goes to France and uses your company's app to do something, what laws govern any potential litigation? I have no idea and you probably don't either. Fortunately, there are people in your company who will be thinking about this, aka your legal team.

Different parts of the world also have rules with regards to how you display and collect information on websites and apps, including the much maligned EU Cookie Law. How and when to display this information, especially if it changes can have real consequences on design and marketing. Some examples of questions that you may need to ask:

  • If you're required to ask for compliance at certain points, do you need to use specific language?
    • How prominent does this notice need to be?
    • Who does it apply to?
  • When and how do you need to announce major changes to your user base?
    • What defines a major change?
    • What is the actual user base to which this change applies?

Work closely with your legal team to understand what will affect the product on an ongoing basis. Laws and regulations change and being aware of potential changes can help you avoid nasty surprises and unnecessary delays to market and product launches.

 

PAYMENTS (¥$₡₱£€)

This is one of the most complex parts of going global, so it's not entirely surprising that a collection of currency symbols looks like a cartoon version of swearing. Here are just a few of the things that might come into play:

Credit card processing, taxes, exchange rates, foreign fees, international credits, new forms of payment, security requirements, merchant accounts, interchange fees... etc. etc.

Without even going into detail, it's clear that there are many things to consider with regards to international payments. If I tried to cover each of these things, I would either give up and this post would never be finished or you'd give up reading partway through. Rather than do that, here's a structure to consider in defining your international payments roadmap:

  1. match up your expansion plan with the primary payment methods
    • if there's a large addressable market, but only 3% credit card penetration, how will that affect the order in which you place that market?
  2. understand the regulatory needs for online transactions
    • do you need to collect information with each transaction, e.g. asking for CVV?
  3. understand the needs for merchant accounts, physical offices and bank accounts
    • how long in advance do you need to get these in place?
    • are these required to operate in the country?
  4. iteration is always possible
    • consider launching with only credit cards to start (which is relatively easy) and build on this at a later time

Work closely with your F&A teams to understand the financial regulations and timing requirements that will apply. This can be one of the most complicated areas even in just a single market and at Uber, the Money team is absolutely critical in managing transactions and payments for consumers and drivers globally. In the pithy words of the Wu-Tang Clan:

 

On that note, this is the last and final blog post of the series on going global. If you missed the earlier posts, you can get to them here: part i, part ii. I hope you found these useful, and if there are any future areas you'd like to see covered, feel free to let me know in the comments.

going global, part ii

Quick recap: to scale internationalization (i18n), you need strong tools and processes in place. These posts go through an overview of those tools and processes and how to approach them. 

In the first post of this series, we went through pre-planning, i.e. figuring out what parts of the world suit your expansion efforts. In this post, we're going to cover content. In this context, I'm defining content as language, marketing, and design. Fundamentally, it's about how you communicate with people. In real life, communication is full of miscommunication and it's no different online. As George Bernard Shaw said,

"The single biggest problem in communication is the illusion that it has taken place."

Language translation is usually the first thing that people think of when it comes to i18n.

From a purely technical perspective, there are several things to keep in mind. You'll need to support multi-byte character sets, typically Asian languages. As you continue to add languages and dictionaries, your file size, binary size and loading time could also increase and lead to a bad user experience. Depending on whether or not you go there, you'll also need to support RTL script which might have a big impact on design and layout.

The quality of your translations is also critical. There are many hilarious examples of poor translations - see 1, 2, 3. However, regardless of the publicity, you probably don't want your company to show up on a list like this. As with many things, you get what you pay for. Google Translate and other free services are machine-generated and that's what often leads to the bad translation examples shown previously. To get the best translations today, you need people. Translators will charge you on a per-word basis - sample range is 15-25 cents/word. Generally, good Asian translators are much harder to find than good European translators. Getting good translations is time-consuming so plan well ahead of your product launch.

Of course, it's one thing to translate your content once but as your products keep changing, so do your translations. Building out a good process requires extensibility and a deep understanding of the tone and voice of the local market. How do you make sure that as you make changes, they automatically flow through to the translations? And if those changes flow through, how do you confirm that they are both true to the brand and reflective of the local market?

For example, in the US, many companies use first names in a business context but it's considered inappropriate and potentially rude to address people by first names in other parts of the world. When people can develop a personal connection with your products, it deepens user engagement and loyalty. To do so, every word of copy needs to be thought out carefully even if it's not a traditional translation, e.g. UK English v. US English. Idiomatic usage and slang can become especially problematic. For example, the distinction between pants and trousers may not matter in the US, but in the UK, pants typically refer to underpants. Small word, big difference. While you can build systems to automatically update site and app translations, building out a core marketing message is usually manual and needs to be worked into the product development process. So, as you build your marketing communications, whether blog post, email or social media, it helps to have a comprehensive view of how product and marketing work together.

Finally, visual communication is another form of content and it's equally important. Think about whether or not your app needs to look different in different parts of the world and the level to which it differs.

If you open the Uber app in different parts of the world, and you’ll see different maps, different icons and even different car types. However, there’s a consistency that immediately gives you the comfort of knowing that you’re using Uber and everything that it stands for.

In the examples shown here, you'll see what the Uber app looks like in London, Beijing, San Francisco, Bogotá. While there are certainly differences, there's a cohesion throughout. 

 

 

 

 

 

 

Another great example of this is Plants vs. Zombies by PopCap Games. Here are different versions of the opening screens of the game:

 

As you can see in these examples, there's a lot of introspection involved to develop a foundation on which content comes from. When you have multiple people at a company working on the product, it can be easy to end up with a patchwork quilt. Creating a reference for people to understand what you're about can make i18n much easier. Here are some questions that can give you a place to start:

  • what are the values that your company stands for?
    • this determines the style, diction and tone of your copy
    • think about the adjectives and words you use to describe the company
  • what things are immediately recognizable about your brand?
    • what are the things that people would say in a word association game?
    • how do you represent the company in a visual form? 
  • what are the design elements that inform the above?
    • e.g. color, shape, movement

Understanding who you are informs how you develop content and ultimately, how you internationalize it. Know thyself.

Coming up in the next (and last) post of this series - what are operational i18n areas that affect product and technology? If you have specific areas that you'd like to hear more about, feel free to comment.

going global

At a growing company, there’s a point where you want to expand internationally, usually to increase revenues and users. There are a number of factors that go into making that decision but let’s assume that you’ve made it. So, now what?

At Google, I took the ability to expand internationally for granted. Why wouldn’t we would launch every product in 40 languages and have it work seamlessly around the world? It was only when I left Google that I realized what a strong foundation of processes and tools existed. For example, the first time that we translated uber.com, we realized that every change in content required multiple code changes and redeploying the entire website each time. When you have to build everything from scratch, you really come to appreciate what’s involved in scaling internationally. To that end, I wanted to highlight some learnings from my experiences.

This is far from a comprehensive plan. It outlines some major areas to consider and how to approach them. Also, not everything needs to be done from the outset but will be important to ultimately scale. At Uber, we launched our first international city, Paris, in 30 days. There was a lot of manual work to continue launching in other countries and languages while we didn’t have a core set of international systems  - we had to charge everyone in US dollars for several months. In parallel, we built out the foundations and kept moving pieces onto the new infrastructure, which allowed Uber to keep momentum and still scale.

Good planning and fast execution are a delicate balance. Or, in the words of George S. Patton:

“A good plan today is better than a perfect plan tomorrow.”

There’s a lot to cover with expansion, so rather than addressing it all at once, I’ve broadly grouped things into three areas, which will be addressed in separate posts:

  1. pre-planning - defining your geographic expansion strategy

  2. content - marketing, communication and design (COMING UP)

  3. operations - payments and legal (COMING UP)

 

PRE-PLANNING

To start, it’s important to set a structure for where you want to go. You don’t turn on every corner of the world overnight. Rather, you work your way through different geographies in some priority order. I’ve suggested countries as the geographical unit because it’s large enough to allow for expansion but small enough to be targeted - this may vary depending on your company model. It doesn’t and shouldn’t be weeks of planning but it’s worth taking the time to do some research. Broadly, it helps to group countries into three tiers:

  • tier 1: as soon as possible

  • tier 2: within the next 6 months

  • tier 3: everything else

Each company will need to come up with their own rubric for evaluation based on their user mix and company goals. It helps to have a general understanding of population, literacy rate and internet penetration regardless of industry to understand the growth potential. However, digital advertising companies building publisher tools will obviously have different considerations than a consumer company building out social networks. Some questions to consider as you define this include the following:

  • on what devices do people use your product?

    • mobile v. desktop

    • smartphone penetration

    • OS and device mix

  • connectivity

    • 2G/3G/LTE

    • wi-fi cost and speed

  • what is the segment of users you want to address?

  • if your product isn’t free, can you price it appropriately?

    • average salary of your segment

    • comparables (e.g. with Uber, compare price to taxi)

  • are there barriers to entry?

A great example of understanding the market is Facebook’s expansion into Asia and Africa, where the majority of people typically don’t have high-end phones. There, Facebook is testing Facebook Lite, which is only a ~250KB download and specifically built to address low-quality 2G connections. At the same time, it’s a native Android and takes advantage of push notifications and the camera, remaining true to how people use Facebook and what the product goals are.

A diametric example is Netflix’s expansion into Cuba. In Cuba, it’s estimated that 26% of people have internet access and it’s often not at high-enough to stream video. It’s also priced at $7.99 in a country where the average monthly salary is $20 and requires a credit or debit card which most of the country doesn’t have. This is a strategic play - a forward-looking view on growth in the country anticipating the political change in US diplomatic relations.

As you can see, different companies have very different approaches. The key here is to figure out what matters for your company. Create a strategy and be ready to adapt when things don't go as you plan, as they invariably will.

Coming up in the next post - how will you create content that is in the right language and resonates with the local market? Until then, adiós, sayonara and au revoir.