The Net Works

I believe that our only real asset is our network. Yet networks are still poorly understood and networking still has a bad reputation.

Allow me a quick personal anecdote. When I was 16, everyone in my school had to do two weeks work experience and most of my friends got their parents to sort them out with interesting placements. However much to my annoyance at the time, my dad steadfastly refused on some ambiguous – to me at least – moral grounds (something about not propping up ‘old boys networks’, standing on my own two feet etc – you know the kind of thing). Anyway, I was left to the mercy of our school geography teacher/careers adviser, who sorted me out with a placement at a solicitors office, despite never having expressed any interest in law. Needless to say it wasn’t the most stimulating of experiences yet I did end up reading Zen and the Art of Motorcycle maintenance, in between ploughing through divorce papers, so it wasn’t completely wasted.

“Networking is just one letter away from not working.” Chris Powell

Despite admiring my dad’s meritocratic principals, and a certain get up and go spirit it may have instilled in me, I think the world has changed profoundly in the last 20 years due to the web, so that I think it’s now possible for reputation and ability to be broadly aligned in a way that wasn’t the case a generation ago. Therefore, this post is an attempt to encapsulate what we’ve learned about networks, and how we use them. And even though I’ve blogged about some of these themes previously, I wanted to capture as much as I can in one place with a hope that it might spark some thoughts and discussion.

Strong Ties and the Dunbar Number

The Oxford based sociologist Robin Dunbar has studied all kinds of communities from all over the world and concluded that we have an upper limit to the number of strong social connections we can maintain at any point in time. This number is approximately 150 and is now known as the Dunbar number. These ‘strong ties’ usually tend to be friends, family and close colleagues. And at it’s most basic, for communities smaller than about 150, everyone tends to know everybody (to some degree at least) which means stuff gets done. Whereas for larger communities we tend not to remember everyone’s name, or what they do, so in an organisational context this leads to the sense of ‘a bunch of people over there who we don’t know what they do’ which can be divisive and unproductive.

Dunbar’s work has been so influential that organisations such as Gortex use it as an organising principal in their business. I’m told that in Gortex, when a business unit grows beyond 150 people, they break it up into 2 smaller units as this is more efficient. On a personal note, this was made real for me in my last company (with about 100 people) where it really was possible to make something happen by walking around the office and chatting to people, in a way I hadn’t experience previously when working for larger organisations.

Whilst Dunbar’s work is fascinating and highly influential, I can’t help but feel that Robin Dunbar hasn’t quite grasped the impact of the web and social networks yet. In an article for the RSA last year, he stated that “Social Networking sites [such as Facebook, twitter…] are very good for slowing down the rate at which relationships decay.” but went on to argue that they aren’t good at forming new relationships. However I profoundly disagree with this statement based on personal and professional experience of using social media to forge all kinds of new relationships and opportunities. Which brings us to Gravenotter and the more controversial area of Weak Ties.

Weak Ties and Innovation Capacity

In 1973, a US based sociologist Mark Gravenotter developed a theory on the spread of information in social networks known as “The Strength of Weak Ties”. There seems to be some debate as to the upper limit of Weak Ties that people can maintain, and there almost certainly is one, but it’s certainly in the 1000’s so for purposes of this article, essentially infinite. These are relationships with friends of friends, people we know by reputation, or people we only see very infrequently (and in my case often forget their names). In the past we were ill equipped to manage our weak ties however with the rise of social networks we have in many ways the perfect tool to build and maintain many more relationships with people we don’t know well.

Building upon this, Ron Burt (and of course many others) have looked at the impact of our weak ties and concluded that there is a correlation between the size and diversity of our weak ties and our innovation capacity. In other words, the more people we know, and the greater their diversity, the more likely we are to be innovative. This has been demonstrated powerfully in some studies of the social networks in two very large organisations. Broadly, Burt and his team mapped the social networks of employees in – one an investment bank and the other a large defence company – and in both cases the the larger and more diverse their social networks, the more they earn, the better they are perceived as innovators by their bosses etc. This is quite startling in many ways but intuitively feels right to me. If all your friends know each other then what tends to happen is you all recycle the same information (the same jokes, the same facts etc), whereas the more people you know from different backgrounds you get exposed to more and different information, and this has a direct correlation with your pay, bonus and innovation capacity.

“You don’t own your reputation. It lives and breathes in those that interact with you.” Ron Burt

Burt goes on to argue that open networks create bandwidth not echo. In other words, closed networks create exaggerated opinions or reputations, whereas open networks create innovation capacity. And your reputation is not your property. You can’t control it, only influence it directly, through the stories other people tell about you. And because the internet is written in ink not in pencil (i.e. is a much more permanent record) I like to think that this means that we all need to behave more responsibly because our collective memory is being written in real time.

For instance, when I googled the name of a former boss of mine, the top search result on Google was an Australian dating website (he used to live there) called ‘’ which described in glorious technicolour why he was not worth dating, which amused me greatly as I didn’t particularly like him, but that’s another story. Needless to say we are all googled to a greater or lesser extent and we want to make sure those first results don’t return ‘red flags’, as in his case, and the only way to influence that is to behave responsibly.

Conversations first, then relationships, then transactions

I believe the one and only rule of networks could be summarised as ‘Conversations first, then relationships, then transactions’. I’ve blogged about this before (here) and other people seem to have run with it more that I have (notably Viv McWaters), but I’m a convert to the idea that innovation today is much more like playing Poker rather than Chess. In other words it’s a numbers game – whereby you have lots of conversations, of which some will become relationships, and some of those become transactions. This process cannot be reversed, without breaking some fairly fundamental social norms, and can seldom be predicted. If you knew exactly where a conversation was leading then it wouldn’t be a conversation. And I’m inspired by work by the RAND corporate and others on distributed networks.

The image below contains 47 dots connected in three completely different ways. One on the left is total centralised, and is similar to the transport infrastructure in the UK (everywhere is well connected to London, but try getting from Manchester to Nottingham on a Tuesday morning and you can pretty much forget the rest of your day) and the one of the right is distributed and similar to the way the internet is structured. In fact, I’m told that the only phones that worked around the World Trade Center after the 9/11 attacks where the VOIP ones (Internet Telephony) as there were multiple paths for the signals to travel. And if the world increasingly works primarily through distributed networks (on the right) why are most organisations and institutions still structured more like the centralised network on the left?

Trust and Hierarchies

The glue that makes networks work, is Trust. Trust takes time to be earned but can be eroded very quickly. Karen Stephenson, an expert in social networks from the University of Rotterdam, makes the point that Hierarchies and Netorks are the two dominant forms of organisation within and across networks, however one is very visible (the org chart on your intranet is one manifestation) however the invisible networks and potentially more powerful at any point in time. And if you are working across organisations, and let’s face it all the most interesting problems fall outside of the remit of even the largest of organisations, then it’s harder to resort to the hierarchy to make decisions and get stuff done. Rather, decision making happens much more through building consensus, and through trust.

“Trust is the glue that holds the network together, which at any point in time is more powerful than transactions and authority” Karen Stepheson

The 90:9:1 Rule

Duncan Watts, now at Yahoo, has done a whole series of fascinating experiments around networks which I won’t describe in detail, but his work and books are definitely worth checking out. One thing that has cropped up again and again in all sorts of circumstances is the 90:9:1 rule – otherwise referred to as the participation inequality – which I believe is actually attributed to Jakob Nielsen. Namely, for any community of 100 people, there is approximately 1% who are lead users or extremely creative or innovative, usually through awareness of some unmet need, and a larger minority of approximately 9% who are synthesisers who package up the learning of the top 1% and make it accessible to all. For instance, for every person who uploads a video clip on You Tube, there are approximately 9 who will comment, or vote, and on average 90 people who simply consume it. A similar breakdown occurs on Wikipedia and other online communities. And we’ve had our own reinforcement of this ‘theory’ in projects we’ve done with EON and Virgin where approximately 1% of the communities we’ve build/worked with have been most engaged/active and creative.

Personally, I would put myself primarily in the ‘synthesisers’ category, and without wishing to be ‘too meta’ this post is intended to be exactly that – synthesising other peoples work to hopefully make sense of it for a wider audience. And the question we often like to ask, is ‘Who is your top 1% and what motivates them’. This is interesting because in somecases your top 1% is very visible (for Virgin Atlantic it was their frequent business travellers i.e. customers who use the product/service the most) but in other cases they are much harder to find (for EON is was often customers who use the least of the product/service i.e. the most energy efficient). And in terms of what motivates them, it is almost never money initially, rather an awareness of a particular unmet need which they feel compelled or motivated to do something about. I just want to be clear that money is important, but tends to be a secondary driver.

How I networks and how they use me

At 100%Open we run several networks including a professional network called the Union where we bring senior innovation professionals together once a quarter for an evening event and we use informal techniques such as Pecha Kucha (short pitches of a particular need or opportunity). This is a free event, though we keep it reasonably close knit to peers.

In terms of social media, on a personal/professional level, here is roughly how I use them now:

Twitter – I love twitter – it’s my favourite place to hang out online. I use it to follow anybody (currently about 1500 people) who are interesting to me personally or professionally. I do this for several reasons – to ambiently keep in touch, conduct mini surveys, to learn and build relationships. I use it to share personal updates about what I’m thinking/doing (within reason) and professional updates about events, projects. Have both a personal (@rolandharwood) and professional (@100open) twitter account and try to mix up the posts to both. We also use twitterfeed to repost other peoples open innovation challenge tweets to our @100open account. I find the blurring of my professional and personal lives on twitter interesting and helpful thought don’t always get it right and some of my closest friends complain about the volume and relevance of my tweets (but I guess I don’t care enough to change) and like the fact that they know a bit more about what we do.

LinkedIn – I find LinkedIn increasingly useful but it’s purely for professional contacts I’ve met or corresponded with (currently approx 1000 people). I tend to use it as an unofficial CV, but more importantly to check out people’s experience and contacts. Use it to share stuff I’m doing (via twitter feed, blog, and slideshare) and build reputation and profile and start new conversations.

Facebook – I don’t use Facebook as much as I used to. For me it’s mainly for friends and family (currently approx 500 people) and for sharing pictures and photos and so a bit more intimate and closed. I re-post my tweets as my status updates so that tends to start quite a few conversations. I also like to see what my friends are up to.

Other – We use other onlinespaces such as slideshare/vimeo/flickr/this blog etc for specific content too.

Monitising Networks

Despite starting this post saying that our only asset is our network, we’ve thought long and hard about how we monitise this and for now have come to the conclusion that we do this indirectly, not directly. The reason being as follows. If you charge for introductions or subscriptions to a network, there will be some people who find that excellent value, however networks being a numbers game (Poker rather than Chess – see above), the majority of people will tend to feel they have poor value from this kind of arrangement.

“The only thing worse than being talked about is not being talked about.” Oscar Wilde

Therefore, we’ve concluded that the network is better monitised indirectly, and that way everybody gets some value out of it and people only engage with you directly if they want to and you are easy and straightforward to work with. Time will tell if this is a smart strategy. However for companies like P&G, they pride themselves as being perceived as a partner of choice in their industry, and even though in many ways they are very similar to their competition, this means that they get to see opportunities before their competitors which has huge value. So our aim is to be the partner of choice in our industry.

Also, we’ve come to the conclusion that so much opportunity comes about through a low lying awareness of what you are doing and what your capabilities are, which is another field where social media can help massively. However in order to succeed here you need to be famous for something that people want which usually means keeping things simply.

“The world isn’t getting smaller, the networks are getting bigger” Greg Hadfield


So that’s it. This is pretty much all I’ve learned about networks over the past few years. I think networks are as important, if not more important, than our knowledge (either as individuals or organisations). And the increasing interconnections mean we all have to figure out how to harness their power and minimise the downsides of networks (which ought to be another post as this one is long enough!).

If I were to go back to University now (which I have not intention of doing by the way) I’m pretty sure it would be do study Network or Complexity Science which seem to be two really exciting areas right now. Anyway, I don’t have any grand conclusions to all of the above, but I present it as food for thought, discussion and debate, and would welcome any comments as ever. Thanks for reading and if you do nothing else, please chat to someone you don’t know today – you never know what might happen.

by Roland


  1. Love this post, Roland – a great synthesis of all sorts of interesting research. Good to see a link to Jakob Nielsen as well (sp by the way) – I used to read his stuff a lot when working in IT.
    I’m planning a research programme for this year on just this stuff, so we should get together.

  2. Thanks Louise. Really appreciate that. Have correct the spelling of Jakob – thanks for pointing that out. Yes he’s great isn’t he. Would love to hook up on this stuff. Cheers, Roland

  3. Beautifully written Roland. My mantra is that all we need to achieve any of our person and professional dreams is within our mind and within our network. Both vital resources for the 21st Century Co-Creator.

    Look forward to seeing more of the emerging science in this area in The New Works More….

  4. Great post. This is one of the chapters of a new book! I am particularly interested in the qualitative nature of these networks too. What you say to which audiences and what effect is has. Your discussion on social media is a good start as looking at what you might call the ‘communication context’ of networks.

  5. Great post, one that I will share with folks who are just now understanding the power of networks. Thank you for your efforts.

    I’ve been intrigued about how we might guide networks strategically. At Purdue, we have been developing new strategy disciplines appropriate for loosely joined, open networks. We are finding ways to accelerate civic innovation…improving schools, building sustainable communities, and so on.

    It all starts with thinking differently and understanding the power of networks. Your post provides a great introduction.

  6. Hi Roland, came here via Johnnie Moore, with whom I had a great lunch recently Viv McWaters was there too).
    Very interesting and useful synthesis. Love the 90:9:1 thing.
    On hierarchy v network, I would say it’s down to prestige and power. It’s much easier to sit on your ass in a hierarchy once you’ve “made it” to the corner office. Becoming and remaining a useful node in a network is orders of magnitude harder, IMHO.

    btw, I think there’s a missing word in one of your sentences – “In fact, I’m told that the only phones that around the World Trade Center after the 9/11 attacks where the VOIP ones” should be “that worked around…” ?


  7. Apols for double-comment. Found this quote that kinda refers to the hierarchy/network thing… It’s from Carl Rogers and H. Jerome Freiberg Freedom to Learn 3rd Edition page 372

    He told me that while the experimental plants continue to do extremely well, and he feels pride in the work he has done with them, he regards his work with the corporation as a failure. The top management, though appreciative of the increased profits and good morale of the experimental plants, has not moved to follow this model in their other plants, even though it appears evident that overall profits would be increased. “Why not?” I inquired. His answer was most thought-provoking: “When managers from other plants look closely at what we are doing, they gradually realize how much of their power they would have to give away, to share with their employees. And they are not willing to give up that power.” When I stated that it appeared that power over people was even more important than profits- which are supposed to be the all-important goal in industry- he agreed.

  8. Hey Dwight. Thanks for the double comment!

    Interesting comments and quote too on power vs people vs prestige. I’m not sure I agree it’s easier to get the corner office but it requires very different skills and performs a different role. If nothing else I guess I’m arguing for more rebalancing of informal networks with formal hierarchies – both of which are needed but the latter is much easier to see and therefore respect. I guess it’s a balance as with all things. Love your point about becoming and remaining a useful node in a network being very difficult tho.

    PS. Corrected the typo and thanks for pointing that out.

  9. Roland

    I loved your posting too – great summary. Clearly your dad did okay by you, despite not wanting to put you in with people he knew.

    Which is why I disagree with this idea of conversations first. Transactions come first. With transactions comes trust. I helped a friend set up a great network in London We wanted to increase interaction between the best world class research units across the whole of the Greater South East (roughly half of the world class researchers in the UK So we trained one lead academic per department to help map their colleagues capabilities and find them industrial clients

    Being ex-McK I was curious to survey whether we were doing any good One of the questions we asked was whether the relationship with an industrial company was NEW to the department or whether it was an existing link.

    Here is the interesting thing. Different KINDS of interactions happened for new relationships compared to old ones. So if you mapped value of interaction against age of relationship with a company you could see a nice curve.

    Companies started out with low value, low risk transactions. Student placements, little bits of consultancy work. Further down the track they might get into contract research or joint bids for funding, which soak up a lot of effort. And only old established relationships would support the really high value licensing deals and contract equipment hire which really paid for new capabilities for the academic departments.

    So, we taught the academics to get into conversations about all the attractive students they could place with the company that would revolutionise their platform and blow their minds. We taught them to exploit the brilliant funding for KTP to help pay for just that injection of ideas and eager young minds. Because that is what will feed the long term pipeline of interaction.

    Other studies agree with our findings 80%+ of licensees for University technology are introduced to the Technology Licensing Office by the ACADEMIC INVENTOR. In other words, the relationship was already there.

    So, I disagree. Transact. But Keep It Simple, Stupid 😉

  10. Hi Roland – Just catching up after being off-line for a coupla weeks. Really liked this synthesis and thanks for the call out 🙂 Johnnie Moore and I found another use for the networks diagram recently when delivering some facilitation training. It very nicely describes different approaches to facilitation from the one-to-many, small groups and innovative participation. Still lots more work to be done around this. Cheers, Viv

  11. Hi Roland, thank you for putting this together. Dunbar isn’t the only person to talk about networks at the RSA, there’s plenty more on their podcast roll – but you probably know that!

    I’ve just sent this link to another member of my extended network so thought I’d drop it here too: It’s from Prof Greg Richards of Tilburg University, NL, on ‘Leisure in the Network Society’. Worth a scan, with a very extensive bibliography to match.

  12. Hey thanks David. Really appreciate the link. Yes I’m an RSA fellow (and had some involvement in the RSA Networks project a few years ago) so know about some of the stuff going on but appreciate the heads up again. Cheers, Roland

Post a comment

You must be logged in to post a comment.