Off-shoring and globally distributed Agile teams have become a fact of life for many of us working on Agile software development projects.
However, many in our industry feel that Agile and offshoring go together like orange juice and toothpaste. If you have business analysts in the UK, testers in Russia and developers in Singapore surely this flies in the face of the Agile principle of a single common room where the whole team can communicate face to face regardless of the benefits of Skype and video conferencing.
Without doubt the Agile offshore model presents a number of challenges. One of the greatest being the cultural challenges you will face on a regular basis. Managing projects is difficult enough when everybody shares a common culture. When people in the globally distributed teams have different cultural backgrounds serious misunderstandings can arise.
Distributed Agile Teams
For instance, most Agile methodologies prescribe ways of working that embraces change, accepts the element of uncertainty inherent in product development and promotes creativity and self-organisation. But what if the offshore team you are working with comes from a culture that struggles to embed those very concepts? These challenges often increase when the offshore team are more than just a body shop (how I hate that term) for development or testing. When the offshore element of the project team is tasked with wider responsibilities that bring a greater level of autonomy this tends to generate far greater challenges.
In my own experience I found it hard to understand and quantify these cultural challenges and even harder to resolve them until I stumbled across some writing by Professor Geert Hofstede on ‘Dimensions of national culture’.
Hofstede’s Dimension of National Culture
Working for IBM in the 1970s as a manager of personnel research, Professor Hofstede job was to travel the globe and interview employees, asking about such things as to how people solved problems and how they worked together and what their attitudes were to authority. Over time Hofstede was able to develop an enormous database for analysing the ways in which cultures differ from one another. Today ‘Hofstede’s Dimensions’ are the most widely used paradigms in cross-cultural psychology.
In his research he found five value dimensions that are used in order to explain the cultural diversity in the world.
In my experience as a project delivery and test manager I have found the following three dimensions to be particularly salient in helping to understand some of the cultural challenges that inhibit teams working together efficiently when trying to follow an Agile course.
- Power Distance Index
- Individualism v Collectivism
- Uncertainty Avoidance Index
Power Distance Index (PDI)
People in societies exhibiting a large degree of Power Distance accept a hierarchical order in which everybody has a place. In societies with low Power Distance, people strive to equalise the distribution of power and demand justification for inequalities of power.
If there is a higher degree of deference to authority on your project then the ability to get open and honest answers could be difficult.
Individualism v Collectivism
The individualism-collectivism dimension identifies the extent to which people in a society are integrated into groups. In an individualist society, there is an expectation that individuals look after themselves and connections between individuals are loose; while in a collectivist society, individuals are integrated into strong, cohesive groups, which may often involve extended family. Unsurprisingly, the US scores high on the individualism index (and is the only industrialised country in the world that does not provide universal health care).
From a project perspective in Individualistic societies, good candidates are people who are outspoken and express strong opinions. In Collectivist societies, good candidates are people who are relatively modest and less likely to speak out or rock the boat.
In an agile Scrum environment where members are expected to be self-managing does high collectivism presents a problem?
Uncertainty Avoidance Index (UAI)
Uncertainty Avoidance expresses the degree to which the members of a society feel uncomfortable with uncertainty and ambiguity. Individuals from a high uncertainty avoidance cultural background may feel a strong need for a definitive prognosis, time line, and outcomes expectations. Individuals from a low uncertainty avoidance cultural background may feel more comfortable with the unknown and have less need for a definitive prognosis. Strangely, Belgium and Denmark are neighbours but at opposite scales of uncertainty avoidance. According to Hofstede’s findings the Danes have more in common when tolerating ambiguity with Jamaica than with Belgium.
The ramifications for the Agile principle of being able to respond to change are obvious where the UAI is high.
In my most recent project I worked with a large offshore team consisting of multiple scrum teams. The bulk of the issues we encountered could be traced back to a few core areas:
- Poor communication – due to heavy a layer of middle management ‘controlling’ peer to peer communication channels
- Working to the letter rather than the spirit of a requirement – imagine a text box where it is not possibletoaddanyspacesbetweenwords because the requirement stated alpha-numerics but did not explicitly state the use of the space bar or backspace. Yes that really happened.
- Failure to apply course correction
- Difficulty in getting a true picture of progress – On our project we found that we could not get a straight answer from scrum team members as it was not their place to contradict the message from the managers that everything was ok.
Reading Hofstede’s work I could map his dimensions against the culture I was working with which helped me understand some of the underlying issues. In the case of my project the offshore location was India which scores in the middle on the Collectivism chart (although arguably higher in the workplace), high on the Power Distance Index and mid table on the Uncertainty Avoidance scale.
To quote the Hofstede School:
‘If one were to encapsulate the Indian attitude, one could use the following words and phrases : dependent on the boss or the power holder for direction…People generally do not feel driven and compelled to take action-initiatives and comfortably settle into established rolls and routines without questioning.’
There are just as many, if not more, positive traits displayed in workplaces which foster Indian values, but I could immediately align some of the index scorings to the issues we faced on our Agile project.
I’m using India as my most recent example but I could apply the same understanding to other offshore centres I had worked with. And of course the awareness of these national dimensions works both ways. Appraisals are a good example of how the extension of Anglo-Saxon models to different cultures can go completely wrong. As a result of their low PDI (and in combination with high individualism) the US and UK promote the idea of providing frank, direct feedback, as “the right way” of improving performance anywhere on the planet. This notion fails to acknowledge that in Collectivistic cultures with high PDI (which are present in much more countries in the world) such procedures are seen as disgraceful and disrespectful.
Solutions for the Distributed Agile Team
What we did on our project to mitigate some of these challenges:
- We sent some of our key personnel to the offshore location and vice versa in a form of cultural exchange.
- We made an effort to get out of the office. We had a darts and pool night and went for meals. This helped break down the hierarchy and form relationships.
- We directly appraised individuals in the offshore team, recognised achievements and made an effort to learn everyone’s name not just management.
- We hired resource who originated from offshore region. Most Western countries are quite multi-cultural so there shouldn’t be a problem finding a developer, tester or Project Manager etc from India, Russia etc.
One needs to tread carefully around any suggestion that an offshore project may fail due to cultural discordance between two teams unless one wants to be tainted with accusations of generalisations and stereotyping. But it is something we should not avoid discussing when engaging with different cultures and Hofstede’s dimensions helps facilitate that discussion.
Many teams prefer to pretend that culture is not that relevant. They think that managing projects is the same all over the world. In order to be more effective, we need find out more about the values of the culture(s) in which we may be operating. If culture considerations are not aligned with the Agile approach, the implementation of an Agile strategy will not be successful.
Hofstede dimensions will not help address all the issues such as time differences, language skills, poorly incentivised staff, or inept members of the project team. Individuals within all cultures vary based on differences, preferences, values, and experiences. However, understanding the key national dimensions is a way to take advantage of cultural knowledge as an important success factor for globally distributed Agile teams.
Finally, it should be noted that this blog post focuses on one side of the relationship – the challenges the onsite team experience in their engagement with the offshore supplier. I look forward to hearing feedback that provides perspective from the offshore experience.