Using social data to predict Startup Weekend attendees

This post is the first one from the category “show me your network and I’ll tell you a story”. My goal here is to show how one can grow a business by leveraging publicly available data, mostly from the social networks.

Let’s start with the case study of the event organisers. This weekend in Cracow, Poland and 6 other cities around the world a bunch of creative people will create some crazy projects, perhaps even businesses. Events organized under the Startup Weekend’s umbrella help wantrepreneurs find like minded co-founders and convert them into doers during about 54 hours of intensive work. If you know who is organising an event, where it takes place, who are the mentors, judges or guests invited by the organiser, you pretty much know enough to effectively use social media to promote the event to the perfect audience.

Find you future clients or event attendees

In this example we will use Twitter which is a huge interest graph - its users express their interest in any topic by following brands, influential people and friends active in the niche they care about. On the website of Startup Weekend Krakow there is a list of 22 mentors, judges and organisers. Let’s take some of them and find their profiles on Twitter. Here’s the list of 12 twitter names I started with:
image

Next, we need to create a network environment we will analyze. We’ll do it by gathering all the followers and profiles followed by our 12 initial accounts involved with SW Krakow. This way we get 6130 Twitter profiles but what need to understand this community are the connections between the members. So we used the data provided by Twitter and found more than 120 000 links which look like this big, hairy ball. A dot represents one user and lines are the connections between them (arrows showing who follows whom).
image


Once we have our interest graph we can run some computations and understand:

  • who are our potential clients/attendees
  • who can help us spread the word about the event and reach non-random audience
  • who are the most influential users in our target market in general and who are the stars from the local scene



Using social network for leads generation

There are 2 factors in the equation we have to take into account: proximity and awareness. In other words, we want to identify Twitter users from our community who really pay attention to what is happening on the startup stage in general and who live close enough to consider attending our event.

As organisers of the local event in Poland we know that there is small probability that our audience will come from the distant cities or from abroad. Of course, there are some passionate people in startup communities in Warsaw, Wroclaw or Poznan who will attend Startup Weekends no matter where they take place, but let’s concentrate just on the core audience.

People who indicated strong interest in our niche are those who follow many of accounts from our base of 6130 profiles. For example on the top of this list appeared @JonahLupton who watches 542 users from our small world. Let’s call this number “Awareness score” - the higher it is for a given profile the more people from our environment he or she aware of and expressed interest in what they say by following them. But Jonah Lupton won’t match our criteria for 2 reasons: 1. he is a typical outlier - he follows 465k (!) Twitter profiles which implies there’s no way he knows who all those people really are 2. he lives in USA.

To find the high probability leads - let’s sort our user base by the Awarness score, take the top 10 per cent with higest values and exclude everyone not located in our city. These are rather restrictive criteria but as a result we filtered out a lot of noise. 

image

So here is the list of my 35 candidates for Startup Weekend Krakow attendees (sorted by the Awerness metric):

  • innovationnest
  • przystas
  • nelse
  • Palgi
  • piotrlipski
  • jimiasty
  • swkrakow
  • kaniama
  • SPINcorner
  • xaoointeractive
  • BitSpiration
  • ZSK_pl
  • MarcinSzelag
  • njet
  • agatkamazur
  • sinnot
  • proidea_news
  • ExclusivePoland
  • bzzzyku
  • k_wasowski
  • MarcinKorbiel
  • janlewan
  • popydo
  • theanxy
  • WojtekSmajda
  • webmuses
  • KamilPoplawski1
  • piotrsynowiec
  • Michal_Kraus
  • gregoryiwacz
  • mariuszwozniak
  • JustynaKwiecien
  • startupstage
  • minowak
  • Skrzypu

If your or your organisation’s profile happen to be on the list please leave a comment and tell us if you took part in SW Krakow, considered it or maybe haven’t even heard about the event.

I already know that folks from the top of this list planned some hacking at SW Krakow but it would be interesting to verify our predictions.

Finally here’s the chart showing our highly targeted leads with number of their followers and our Influence score.

image

How women organize social networks different from men : Scientific Reports : Nature Publishing Group

See on Scoop.it - Social Network Analysis #sna

Superpositions of social networks, such as communication, friendship, or trade networks, are called multiplex networks, forming the structural backbone of human societies. Novel datasets now allow quantification and exploration of multiplex networks.

ukituki’s insight:

On the individual level females perform better economically and are less risk-taking than males. Males reciprocate friendship requests from females faster than vice versa and hesitate to reciprocate hostile actions of females.


On the network level females have more communication partners, who are less connected than partners of males. We find a strong homophily effect for females and higher clustering coefficients of females in trade and attack networks. Cooperative links between males are under-represented, reflecting competition for resources among males.


These results confirm quantitatively that females and males manage their social networks in substantially different ways.


See on nature.com

Complexity Science For Lean & Agile Teams

See on Scoop.it - CustDev: Customer Development, Startups, Metrics, Business Models

Complexity ScienceFor Lean & Agile Teams

See on slideshare.net

Hooked: psychology of how products engage us by @NirEyal

(Source: youtu.be)

Kurt Vonnegut on the Shapes of Stories

"

People ask what the next web will be like, but there won’t be a next web.

The space-based web we currently have will gradually be replaced by a time-based worldstream. It’s already happening… this lifestream — a heterogeneous, content-searchable, real-time messaging stream — arrived in the form of blog posts and RSS feeds, Twitter and other chatstreams, and Facebook walls and timelines. …. All the information on the internet will soon be a time-based structure. In the world of bits, space-based structures are static. Time-based structures are dynamic, always flowing — like time itself.

The web will be history.

"

The End of the Web, Search, and Computer as We Know It | Wired Opinion | Wired.com (via interestingsnippets)

(via interestingsnippets)