capitalBLOG

Mayfield Fund & Foundation Capital - Management Interlock Relationships

December 8, 2006 on 12:39 pm | In Investor Analysis, Network Training, Management Team |

All of the links to LinkSViewer maps in this post are viewable by anyone. However, if you want to manipulate the maps, you must be a registered LinkSViewer user. Registration is free and easy. Just go to one of the maps, try to click around, and you will be prompted to login (or register). Or just click here to sign up now.

I was playing around with the network for Kevin Fong, one of the most connected board members in our database when I digressed and began to play with the combined network for two fairly large investors - Mayfield Fund and Foundation Capital. If you click on Kevin’s name (above) you will see a map that shows how Mayfield and Foundation are very centrally located because of the amount of common companies they share with him. Mayfield should be no surprise because he is a general partner there, but it got me to start wondering about what else Mayfield has in common with Foundation…I started a new map by searching for Mayfield and Foundation, and then mapped them together. Now I set the ‘MinLink’ to 1 (a feature in the taskbar, the default setting is 2) and the result was an explosion of information…

Interesting stuff, but unreadable. I decided that I really wanted to look into the management team interlock between their networks of investments. This is the perfect opportunity to make use of the updated tabs in our LinkSViewer tool. If you haven’t used the tool in a while you may not have noticed, but the tabs have been changed to serve like radio presets, creating views that we think serve the interests of the users, and minimizing the time spent applying filters to get to the interesting information.

By clicking on the “Management” tab (5th from the left), LinkSViewer will automatically filter down and display only the management interlock relationships in the network. This change helped a little, but there are still so many management team members involved that the map is cumbersome and unreadable (see this map)… but I’m not interested in all of the management team members, just the ones who connect to more than one of the companies.

Well the ‘MinLink’ is still set at 1, so let’s change that back to 2 by clicking on the “+” icon next to the ‘MinLink’ box. This will filter out any of those management team members on the periphery who don’t have at least 2 links to companies in the map (we’re setting the minimum number of necessary links at 2). Here is the map (click on it to go to the live version):

This is great. I’m getting down to what I really want to see now, but there are still all of those pesky companies that don’t have connections to any of the management team members with at least 2 links in the map. Now I’m going to perform an advanced procedure. The MinLink feature has a way to filter these nodes out as well.

Most of the time the MinLink box has only one number in it. However, if you want to manually insert a comma and add a second number, it can manipulate the map to a greater extent. When just one number is in the MinLink box, it sets the minimum number of necessary links that a node must have to be in the map, but it applies only to the most peripheral set of nodes (the furthest outside - which, in this particular map, can be thought of as the management team members who are on the outside connected to the companies). For the map above, the MinLink is set at “2″. If you enter a comma and a second number, for example setting the MinLink at “2,2″ then we can eliminate those companies with no management team connections. The second number (after the comma) applies to the first degree nodes from the target nodes, or in other words, instead of the most peripheral nodes, it applies to the companies that are 1 degree away from our target nodes of Foundation and Mayfield. So by setting the second number at 2, we are dictating that in this particular map the companies must have at least two links, and just being connected to the target node is not enough. Now the map will be filtered down only to the nodes involved in the management team interlock relationships… check it out:

Now we can see who the management team members hare that connect these two VCs to each other. I think that the large number of common members between the two companies at the top - Packet Design and Precision I/O.

I hope this served to show you some ways to whittle down a big map to a specific point of interest.