What kind of content silos were you experiencing at Sony?

Kennan Rossi: Our information silos all relate to operational silos. We expose silos to all of our customers. We used to have three or four different business units that our partners had to deal with in each of the three regions, so they had to speak with a dozen people and go to a dozen silos to get information. Very frustrating!

Stephen Townsend: A lot of different teams were supporting the same group of people in their own way. So they’re providing content that’s important to their business units and important for them to communicate…but then that partner ends up getting too many channels. It’s like a firehose and they’re not sure where the best place is to get support. It was really hard for them to self-serve their issues. Our audience don’t want to submit a ticket - they want to self-serve - but when stuff is so spread out and they don’t even know who to ask they just get lost. We wanted to solve this problem.

What challenges do these silos pose to the organization?

Stephen Townsend: One of the biggest challenges was teams creating content and publishing it in a way they felt was best for who they were delivering it to - the specific user of their application or tool. So when you come in and say you want to streamline that process, what you’re really saying is “We’re going to force you to think about this in a different way.”

Stephen Townsend

Senior Manager, Content Strategy and Information Development Manager at

But we had to say your content is going to be living alongside a lot of other content…what does that mean in terms of how your content is structured? What does that mean in terms of how we tag it with taxonomy, in terms of how we organize it in terms of categorization? That was going to teams and setting that tone and saying “We’re going to have to rethink your content in this broader sense.”

What did it take for leadership to decide to solve the issue of content silos?

Kennan Rossi: The executives were well aware of the operational silo problem and dissatisfaction of the partners, so the effort to consolidate all of the various business tools that we expose to the partners under a single PS umbrella came from the top.

I heard there was this project going on. I invited myself and found that they had selected a documentation solution. So they were going to use a support knowledge base for all of their documentation and we actually got stuck operating that for a little over a year. Can you imagine..your publishing process is you open a KB article and you copy and paste your content that’s gone through a review process and press publish. Lather, rinse, repeat. It was PAINFUL.

This is where the bottom up approach came in. We said, “Excuse me but WE NEED TO CHANGE THIS!”. So we did. It took a while to get the attention of the organization, to get all the various program management, product management and engineering around the table to say we need a new documentation solution.

Stephen Townsend: If you have research or feedback from customers, that trumps anything else. If you say this is what our customers want, here’s feedback from them, someone’s opinion or preference of what you should do instead isn’t going to beat that qualitative data. Funny story about the flat KB. When we had moved to our new docs portal we published the same content we had on a KB basis and were doing user research on whether it was usable and findable. We had people clicking through the content and would find these help topics and were saying “This is great! I would have loved to have read this. I didn't know this existed!” It was the same content, just organized in a better way.

What were the organizational or technological obstacles you had to overcome to start solving the issue?

Kennan Rossi: Organizationally, we got agreement that we needed to do something. The question of what to do was the challenge. The culture here is “We’re a technology company - we can build everything. The answer to the whole ‘Build vs. Buy’ thing has always been build, build, build! But we were like “We don’t have 10 years!”.

So we did bring a couple of vendors to the table and encouraged other participants in the program to do the same and went through a whole vendor selection process. Now, we pretty much knew at that point what the answer was going to be since we’d done all the research, but you have to let people come to their own conclusions. So the biggest challenge was taking the right approach to affecting change because it’s a strongly consensus-driven organization.

Stephen Townsend: From the technical side, there was a pretty big learning curve in terms of the Zoomin platform for our team. We have all this content, you’re promising to publish it all in one place for us, but what work do we have to do? What conceptually do we have to understand? The first thing we did was a content audit. What is all our content? What aren’t we aware of? It wasn’t just the content of our team that we wanted to streamline and combine, it was a lot of other content as well. So we had workshops and content audits of what we wanted to publish. Then we had to start thinking about the sort of taxonomy we wanted to create. Do we either let users browse or search for content? What ways do we want customers to find and discover content?

The biggest challenge with our partners is there’s thousands of them with varying levels of experience with different purposes to their roles. Within game development you not only have different roles but you have different specializations and our content has to be discoverable and usable for all these different types of users. It was a challenge trying to figure out how to construct something where people could easily find what they’re looking for, without us having any prior knowledge of their role or what it is they’re looking for.We floated the idea of organizing stuff by apps and tools but then thought that's not a great way of doing it. So we decided to organize it by the overall life cycle of developing a game, since that's the same for all of our partners. The user knows where they're operating within that framework so we went with that. We user-tested it and it’s worked pretty well so far. But that’s all stuff we didn’t know going in and that’s stuff you need to know if you’re making decisions about what to propose.

When did you start getting feedback that this was helping customers and having a positive impact?

Kennan Rossi: We first got the sense we were going to make a difference even before we went live. In our UX research studies we were A/B testing it against our previous solution, and it was night and day. There was no question about it.

We are seeing an uptick in NPS and other metrics but we’re still not down to one site. We went from multiple to two sites, but two is still multiple. Partners are very much less unhappy than they used to be but we still have a way to go. In terms of content types we’re about 80% complete with our migration. In terms of content volume, we’re about 10% complete because we still have massive amounts of developer content that’s still authored in Word! We need to get all of that into structured authoring content, that’s our phase 2.

Kennan Rossi

Director, Global Technical Information Group at

Stephen Townsend: I think the value came out when the organization realized we had a clear place to publish content for our audience and started sending our content strategy team a lot of requests around content our partners wanted. We were able to publish it in a way that made sense. Now we’re tracking case deflection but we are getting good analytics in terms of fewer searches with no results, and better general feedback.

What’s been the internal impact of this initiative?

Kennan Rossi: Operationally, it’s night and day. For our group the efficiencies have been huge - easily 10X. Other teams are coming to us which makes a difference and upper management are noticing there's a good thing happening here.

Stephen Townsend: When you talk about a site powered by Zoomin, if you have the resources and the team and the appetite, you can do pretty much whatever you want and create whatever experience works for your users. Depending on what your publishing system is today you may not have that much control over the presentation or how it’s organized - you may be limited by the system.

We were always asking engineering to add or change this and it never made it into the sprint somehow. We had enough, so we went with this system where we operate it and control it and do what we need for our audience. We listen to them and react to what they’re telling us. It may sound obvious but we were not in that position before and now we very much are. It’s very liberating to know it’s within your control and the only thing you’re limited by is your understanding of your audience and their ongoing needs. As our business changes there’ll be changing needs and innovation we’ll have to provide. Controlling that was really big in us choosing a platform like this.