By Marc Wright
“The average age of people in PwC is 28 and a few months and we found that more and more of them just could not cope with the way we wanted them to work. They told us that the way we did things was too old-fashioned and too difficult. So rather than argue against them, we asked them to help us make it better by turning them into advocates,” explains Simon Levene, Global Knowledge Channels Leader at PricewaterhouseCoopers (PwC).
“We had a long list of possible technology solutions and narrowed them down to two possible platforms and put them into a sandbox where people could play with them. Jive was the clear winner with 97% of users preferring it. The ones who liked the other system tended to be librarians. It was clear what we had to do.”
They started off with 30,000 licenses – but this rapidly accelerated to 100,000. “I believe that the reason why we have had such high adoption is that we have converted the complainers. If you can turn the complainers to your cause then they become fantastic salespeople.”
Levene has been doing knowledge management by his own admission for far too may years – and was an author of the British Standards document on Knowledge Management back in 2001.
“I believe that Knowledge Management cannot depend on culture change as changing any culture is very hard and usually fails. So we had to work with the way we do things at PwC. It is difficult to encourage people to put things into databases on the off chance that someone might want to find it there. Adoption rates for Gateway which is our 7 year-old Knowledge Management system runs at about 14% of the population – and we think that is really good by industry standards. But it depends on a federation of people whose job it is; we do not put things into the system naturally.”
So what was the culture that prevails in this group of 180,000 people that Spark could complement? Levene recalls an occasion when he was contacted at 11 at night by a person in Brazil who had noticed that he had some expertise in turning around failing housing projects in London. The PwC colleague – whom he did not know – was going into a client the following morning on a similar project and needed help.
“Because it was a person-to-person interaction it fitted our culture for me to talk for an hour at night telling him everything I knew about an arcane subject. PwC by its own nature is a fantastic knowledge-sharing organisation that depends on person-to-person connections. We swap knowledge on a one-to-one personal basis at the drop of a hat. “Now if that is our culture then it would be crazy to introduce a system that did not exploit it; we are just introducing the tools that fit our way of working. The problem was that in the old days if I wanted to access some deep knowledge about a complex tax matter, I would have to find the expert and visit them in order to get that knowledge – and that just is not scalable. What we are doing with social networking is connect people and – at the same time – share those conversations with the rest of the world in a very open way. And this defragmenting of our organisation really fits our culture.”
Simon has looked back at the proposals he put forward in the past for portal-based systems that failed. He now realises that the business aims of those projects were very similar to those of Spark – it was just that the technology was not ready: “It’s only in the past two to three years that the technology has reached a maturity where it can meet the business promise.”
So how does PwC use the Jive platform? The core is each person’s profile. These avoid complex structures and controlled taxonomies – people can describe their expertise and interests informally. What’s interesting is that on Spark they have made a distinction between ‘friending’ colleagues and ‘following’ them. So for instance Simon might follow his senior sponsor Moira Elms, (Global Leader Brand Communications who among other things was in charge of knowledge management), because he is interested in what she is doing. But she might not be interested in the detail of what he is doing. So ‘following’ works in one direction – unlike ‘friending’ which is bi-lateral. It shows the importance of making some simple but hugely strategic decisions in the setting up of a social platform and making this key distinction between ’following’ and ‘friending’ has been crucial in Spark’s success.
The only areas where Spark met some resistance in PwC was in the area of confidentiality and IT security. Risk & Quality and the Office of the General Counsel wanted reassurance around how Spark dealt with issues around data privacy and client confidentiality in an age where clients are increasingly concerned about data security, particularly when their data is in someone else’s hands. In the US there is a great deal of legislation around e-discovery which means that they had to put safeguards in place to make Spark compliant with legal requirements both in the US and Europe. There was some resistance from IT as it was not a Microsoft solution and because it was hosted outside of the firewall, albeit in a dedicated PwC environment. They have managed to address these concerns and the system works on mobile devices even with a single sign-on. In the end the objections were not insurmountable and indeed now PwC advises its own clients on Risk Assurance issues in the area of implementing social business.
Another key accelerator are the 1,500 advocates; Philip Mennie who works in Risk Assurance is one of them:
“Through the grapevine I heard there was a thing called Spark. They were asking for people to become advocates. You did not have to be chosen – if you want to become an advocate then you just join in. Our job is to drive Spark in the business units. This means I can talk about the benefits in a way that is relevant to the people in Risk Assurance. In my team we have a group of five advocates and we have a conference call every week to discuss ways that we can implement Spark deeper into the Unit.”
Mennie wears a badge: ‘Ask me about Spark’; so what are the tangible benefits?
“In the past if someone had a highly technical question about a piece of accounting software then they would email the whole group and someone somewhere would have the answer. But then that knowledge is forgotten – so a few weeks later someone will ask the same question and the solution is hidden deep in an inaccessible inbox. So we decided to put such discussions into Spark instead of email and so the next time someone wanted to know the answer to a previously answered question, the solution was sitting there for them already, complete with a contact and what to do about it.”
“The advocates across the world can feedback on any changes and improvements, and they are the best people to help the more senior Partners get involved through reverse mentoring. PwC has 21 lead territories – these are the strategic regions where they make most of their money – and for each of those 21 territories they nominated someone at Director level to be responsible for driving the engagement of Spark. It was a truly viral process that depended on local adoption with strong sponsors rather than a top down global approach. Other territories then just followed suit through word of mouth.”
The benefits have been tangible and in some cases revolutionary. In the San José office one team inverted the way that projects get resourced with talent. The classic model is to win some business and then assign people to the task. At San José they allow people to bid for the work that they find interesting once it has been brought in.
“Another Team has combined Spark and Office Connector for proposal development,” explains Levene. They found that they saved 80% of effort on document versioning compared to the old system. By having the proposal in one place and sharing it with everyone, they managed to get proposals out far faster with a dramatic saving in all the hassle that is necessary when many minds are concentrating on one proposal. This was a big win for us.”
Interestingly Levene and Mennie do not rely on heavy data to prove the value of Spark:
“PwC is full of accountants which means that they distrust any metrics that the we might produce to prove a return on investment. If we gave them details stats they would just want to audit our figures.” So instead Simon concentrates on success stories and anecdotal evidence.
On the technology front they have a policy of ‘buy not build’ which means that they do not customise the Jive platform but just use what comes out of the box and can be scaled without major code writing. They do have a number of recommendations that other companies could find useful:
Groups over Spaces: In Jive you can choose to have ‘Spaces’ that mimic the way companies run projects, with leaders and levels of hierarchies. Leven recommends avoiding them unless absolutely necessary as Groups are more flexible and user-friendly with less hierarchy and structure so you do not have to be at a certain level to post, comment or sign off on material.
Non-work Groups: Spark encourages social groups that have nothing to do with work such as somewhere to share pictures of last year’s Olympics. There is even a David Hasselhoff Appreciation Club. Levene argues that this helps drive adoption by making people familiar with the tools in a no-risk environment.
Bubble Tree: Spark has a graphical interface using BubbleTree that allows you to navigate 8,000 groups in a visual way. It has allowed them to categorise and help people to manage the Groups themselves.
Preparation: What Levene would recommend to other companies on this journey is that “You cannot invest too much in an advocate programme and I wish I’d spent more time with the legal and governance community before we launched”.
And the results?
In Levene’s view: “We have managed to defragment the organisation through a common platform. We had come from a fragmented situation and everyone was fed up with that. And we have helped make a large firm feel small.”
Typically he ends with an anecdote: “Colleagues in Germany were looking for two weeks for someone with a particular software skill set who could speak German but could find no one. Spark had not yet been rolled out in Germany due to concerns from the Work Councils there. So in the end they phoned an English manager to put the question on the network. Within 20 minutes he had 5 people from the US come back who had the requisite skills and spoke German. The Germans were amazed and that helped to push the business to launch Spark in that territory.”
It is an interesting lesson to other organisations how a company with so many accountants have bought into social media so intuitively. “It’s hearts rather than wallets that win the day,” points out Levene.