Archive for the ‘charts and graphs’ tag
Most B2B entrepreneurs have a general sense of the ROI they would get by selling to any particular client. Sell your product to the government, for instance, and be prepared for years of investment and dedication before seeing any money. But once you’re in, your company has a client that is willing to pay very well for your work. If you were to sell to smaller organizations, you could likely get away with much shorter sales cycles — but there is a significantly tinier pot of money on the other end.
So there’s a direct relationship between the amount of money a client can spend and their difficulty in spending it. But it’s certainly not a perfect relationship, and I’ve seen the imperfections trap many entrepreneurs — including me. In fact, the relationship probably looks something like this:
When selling software or services, being below the line of best fit is critically important to keep yourself sane and your margins high. Go too far above the line and it’s unlikely that you can even make the business work — the cost of selling each unit will overwhelm the available revenue.
I learned much of this the hard way. My first company, GoCrossCampus, promoted campus-wide games through student governments. Another company I was involved in sold software to technology transfer offices within universities. For both of these businesses, the amount of bureaucratic red tape involved in the purchasing process was way out of proportion to the amount of money our clients were able to spend. University bureaucracies are notoriously budget-constrained, and student governments can rarely spend more than a few hundred bucks on any particular purchase.
While there are often fewer competitors in these difficult markets, I would caution any entrepreneurs against reading too much into the competitive landscape. Often, highly frictional organizations will stick with fundamentally inferior products (e.g., mainframe systems) rather than switching to any new solution — yours or your competitor’s. If there isn’t any competition because all the other firms have entered or left, that isn’t a good thing. In other words: when you see a graveyard, don’t play there.
Haven’t gotten your pop psychology fix today? Well, OkCupid is pleased to let you know that iPhone users have more sex than users of other smartphones. Their evidence — a handy chart — is below:
OkCupid didn’t provide sample sizes, but let’s assume we’re dealing with their entire userbase — a statistically significant sample.
This is a pretty shocking graph. At age 27, iPhone users report having more than twice as many sexual partners as Android users. Quite an impact for a mobile operating system. So let’s look a bit deeper.
First, the heading — “iPhone users have more sex”. There’s nothing in that chart about the amount of sex people have. It’s a chart of the number of sexual partners. Anecdotally, if one removes the celibate — that is, people who have had zero sex and zero sexual partners — I haven’t seen a strong correlation between having a lot of sex and having a lot of sexual partners. In fact, there may be a negative correlation given the likelihood for people who don’t have long-term relationships to go through significant dry spells. That said, I haven’t seen any scientific data on this.
Second, it is important to realize that this is self-reported data on an online dating site. Data on sexual experience is notoriously bad even when subjects don’t clearly stand to gain by inflating the numbers, as they would on OkCupid. But that doesn’t explain why iPhone users claim more sexual partners than Android or Blackberry users, right?
Let’s take a step back. There are some oddities in this data. While I’m aware that the graph doesn’t provide a longitudinal sample, let’s consider it as such for the moment. The average iPhone user:
1) Has had four sexual partners by the time they graduate from high school.
2) Has less than one sexual partner during college
3) Has 2 sexual partners per year from 24 to 27
Those numbers are…odd. Especially given that the Kinsey Institute finds that about 35% of people haven’t had sex at all by age 18. I’m sure there is a long tail of kids who are having a *lot* of partners, but many of those teens would come from poorer and more urban families, not exactly a hot demographic for iPhones. And less than one (new) sexual partner during college? Who are these people?
So they’re probably lying, at least to some extent. But that still doesn’t explain the delta between iPhone users and Android or Blackberry users, which could be attributed to either or a combination of (a) an actual difference in sexual partners or (b) a difference in exaggeration.
OkCupid is particularly popular in San Francisco and New York — markets where the iPhone is more or less useless as a phone. It takes a certain type of person to own a really expensive, well-designed smartphone that doesn’t really work as a phone. Specifically, a person who perhaps:
1) Is wealthy enough that money is no (or very little) object.
2) Is extremely concerned about appearances and social status.
3) Is in the technology industry and needs it to keep up with innovation.
Let’s put (3) aside for a moment — while tech innovators are a big chunk of my own community, it’s probably a small percentage of the sample here.
(1) is probably correlated with having more sexual partners, although I can’t find any data on this. Once again, however, it is probably a small percentage of OkCupid’s users and could hardly explain a >2x difference in reported sexual partners by age 27.
(2) is more interesting. While someone who is more fashionable probably has more sexual partners than someone who isn’t, I’m more interested in the meta level — the need to appear fashionable and trendy. Awareness and desire to maintain appearances should be strongly correlated with inflating hard-to-verify information on an online dating profile. Like number of sexual partners, for instance.
In reality, the delta is probably driven by a number of factors. But a big piece of it is just about keeping up appearances — in profiles and phones.
Both Matt Mireles and Mark Davis write about startups. Both write well, and both are must-reads for entrepreneurs. But their blogs are really, really different. Matt is controversial — he’s gone after a number of high-profile targets over the past couple months, including David Rose, just about every venture capitalist and New York itself. Matt’s writings are about working outside of the system, struggling against larger institutional forces in an attempt to hack together a company. Mark’s (or Larry Lenihan, or Phin Barnes or any number of VCs)’s writings are insightful, but they focus on learning the ropes and working within the existing system to achieve success.
For the visual learners among us, I’ve thrown a few startup and VC bloggers I like to read on a chart. I guess this is my form of a blogroll. Apologies in advance to anyone not included, I love you too but I budgeted 10 minutes to make this damn thing:
The placement of the dots is a gross estimate, but you get the point. While those at the VC-oriented side of the spectrum tend toward more educational, less controversial fare, entrepreneurs tend to be a bit more inflammatory.
Does this mean that VC bloggers are more restricted by the environment of the VC firm and the potential downside of annoying entrepreneurs or (worse) LPs or other partners? Maybe. But just as controversial writings could be enabled by freedom, controversy is also a tool for gaining pageviews and notoriety. Just look at compete.com’s chart of The Metamorphosis, Matt Mireles’ blog.
Plus, you have to consider supply and demand. There are fewer VCs than entrepreneurs, and content written by VCs simply tends to be in greater demand since they are the ones dishing out the money to young entrepreneurs. So you could say that VC bloggers are don’t need to write controversial content in order to get noticed. It’s certainly true with guys on here like Fred Wilson and Mark Suster — they write controversial stuff when they want to, but they certainly don’t need to start crafting linkbait.
I’d argue that it’s not only smart, it’s necessary for anyone in the startup world to pay close attention to the entire spectrum. Focus too much on the VC writings and you’ll lose sight of the bigger picture and the way founders really think about things. Focus too much on the entrepreneur crowd and you’re just delusional — and missing out on a lot of good posts.
The more time I spend working with startups, the more I find useful lessons for growing companies in random places. Take my gaming (industry) meetup, for instance. I’ve been running it for over a year, but only recently has it begun to “hockey stick”, in industry parlance.
The Opportunity: After running a gaming company in New York City for six months, I realized that there wasn’t a good place for people in the gaming industry to meet others in the gaming industry in an open, cross-pollinated environment. The International Game Developers’ Association’s New York chapter held regular events, but they were primarily focused on software developers, not the entire game creation ecosystem.
The Tactic: Create the New York Gaming Meetup, a (monthly) event where game developers can freely interact with others in the gaming industry as well as those outside the industry. Events would be regularly attended by investors, marketers, designers and others with a big role in making successful games. Meetups would be oriented around a series of demos of games built in the NYC area with networking before and after the demos.
(1) The NY gaming industry is highly fragmented with a focus on small (1-3 person) indie development shops. This isn’t Seattle or LA; there are only a handful of mid-sized gaming studios in NYC. It was critical to recognize that New York is a very different place and build a program that caters to those differences.
(2) There are few potential sponsors of such a meetup in NYC. This event would have to take root with minimal budget.
(3) Space in New York is hard to come by. The event would have to be structured and timed to let us take advantage of free space in bars and restaurants.
(4) As I’ve previously written, the New York tech landscape is very siloed, with little cross-pollination between verticals. In Silicon Valley, anyone working on a tech-enabled solution considers themselves part of the tech industry. In New York, we frame ourselves in terms of the particular vertical we are tackling — the “advertising industry”, the “gaming industry” or the “fashion industry”, for instance. This makes it difficult for events to reach across the social graph, and to this day I rarely see any Gaming Meetup regulars at other big tech events like the NY Tech Meetup or the Y+30.
Execution: For its first year, the event took place at Gallery Bar in the Lower East Side on Tuesday nights. On a scale of 1 to 10, I would give the location a 2, the venue a 7 and the cost a 10 — it was a free (but good) space with AV equipment in a out-of-the-way Manhattan neighborhood. Don’t get me wrong, I love the LES, but it’s a suboptimal place to host an after-work event.
Initial Results: The Gaming Meetup got a decent but not overwhelming response. We had a fairly predictable number of attendees — 55 to 75 per meetup — over our first ten months. The event wasn’t really gaining traction, but it was establishing a good core of game developers and people who loved what we created. The content (demos from local game developers and entrepreneurs) was hit or miss. There weren’t enough games being developed in New York City for us to be truly selective, and for every awesomely cool and instructive game that took the stage we had one guy just trying to sell something to the audience.
Iteration: A few months after starting the meetup, I started iterating on the model. Here are some things we tried and the results we got. Since metrics are important, changes were evaluated on (a) the number of attendees we got, (b) how long those attendees stayed and (c) how people reviewed the event.
Moving it later: Most people would show up at 7:30 anyway, so our 6:30 start time didn’t make any sense — especially since attendees had to travel to the Lower East Side. Good change, kept it.
Focusing on networking rather than demos: The demos started to get stale after a while, so I created one networking-only meetup to see how people would react. Bad idea; many people will only travel for content.
Fewer Demos: This was partially out of necessity, but ultimately it proved to be a good call. Six demos is simply too many. Four is much better.
Themed Meetups: We ran our first themed meetup (on Mobile Gaming) in March, and it was a tipping point of sorts. As it turns out, there is a certain “optimal specificity” in this kind of stuff — make it too general (“Game Demos”) and people aren’t sure what they’ll get. Make it too specific (“Android Development Best Practices”) and most people won’t care. Something in the middle (“Social Games”, “Mobile Gaming”, “Innovation in Consoles”) is ideal.
Higher-profile speakers: Last month, Kenny Rosenblatt (CEO, Arkadium) came and spoke on the topic of social games, and our meetup got 2x the number of people we’ve ever gotten. I’m a bit surprised that I hadn’t gone the high-profile-keynote-speaker route before. I’m certainly capable of sourcing them, and they give me far fewer logistical headaches than half a dozen demoers (one of which will inevitably bring a mac without the right VGA adapter).
The Hockey Stick: As you can see from my chart of RSVPs, I’ve started to figure this thing out. Popularity, of course, is self-reinforcing — now that we’re getting real traction, we’ve landed a great venue at AOL Ventures in the Union Square neighborhood. And our May meetup already has 90+ RSVPs, which is well beyond what any previous NY Gaming Meetup has gotten by this point. Most excitingly, we’re lining up partnerships with other Meetup groups for this summer — for example, we’re getting together with the Y+30 to host a panel on the future of gaming.
If you’re living in San Francisco and working in a hot Valley startup or tech company, you probably have a bit of a commute. If you’re living in NYC and working in a hot startup in the area, not so much. Unlike those in the Valley, most tech companies in New York are in the city itself.
For the entertainment and edification of my northeastern readers, I’ve created a map that superimposes Valley companies on New York and Connecticut, assuming identical driving times (from midtown Manhattan and downtown San Francisco, respectively) and distance from commuter rail (Caltrain and Metro North’s New Haven line, respectively).
To make this more explicit, here are exactly where our Silicon Valley darlings would find themselves:
Adobe: Noroton Heights, Connecticut
Sun Microsystems: north side of Stamford, Connecticut
Apple: North Mianus, Connecticut
Google: North Greenwich, Connecticut
YCombinator: Greenwich, Connecticut
Facebook: Greenwich, Connecticut
Electronic Arts: between Mamaroneck and Eastchester, New York
Oracle: between Mamaroneck and Eastchester, New York
Wikia: Pelham Manor, New York
YouTube: Soundview, Bronx, New York
Zynga: Harlem, Manhattan, New York
Editor’s Note: This post was created in collaboration with Chris Paik, who was invaluable in helping me crunch the numbers. He’s looking for an internship in venture capital, so if you like this post, get in touch with him via his blog.
Lots of you enjoyed my post a few weeks ago on buzz and fund size among NYC venture firms. But why not take it further? Why not use all the data in Crunchbase of financings of NYC companies over the past five years?
So that’s what we did. And we got data for 814 venture financings since March 2005 worth a total of $3.1 billion. We were careful to exclude angel and strategic investors, since data around those deals are poor and would make the results harder to parse.
To start, let’s look at all venture firms that have completed over 7 financings of NYC-based companies in the past 5 years. Here, you can see how they stack up based on number of deals done:
Keep in mind that there’s a long tail here — this chart represents 300 total financing events, only 37% of all the venture financings of NYC-based companies in Crunchbase. The rest of financings were done by other firms.
But this is just parsed by the number of financings — with no thought given to the size of the deals. Thus, let’s look at the (relative) deal size by the firms listed above when investing in NYC companies:
You’ll probably notice that there aren’t any labels on the Y-axis. In brief, I don’t trust the absolute data here. It’s often impossible to distinguish the relative contributions of investors in a syndicated deal. For example, if Union Square does a $1 MM seed deal, there isn’t any ambiguity there. But if the company’s next round is a $10 MM round syndicated among two growth capital firms and Union Square, there’s no way to really know how much each firm invested. However, it is probably safe to say that the growth capital firms do bigger deals than Union Square, since they first joined the syndicate at a later (bigger) round. Thus, the relative data is accurate, but the absolute numbers are highly questionable.
Since we selected these financings based on the zip code of the funded company’s headquarters, we can drill down a bit further and draw some really interesting conclusions. Specifically, where are funded companies? The following map looks at two factors: the number of financings in the zip code (the color of the dot) and the total amount of venture money invested in the zip code (the size of the dot):
There are certainly some surprising things here, at least to me. This entire map seems to be shifted a bit further north than I expected; are there really that many well-funded startups in Murray Hill? I also expected to see a bigger presence in TriBeCa.
There’s a lot of data here, and I’m sure there will be follow-up posts — especially as we dive into the data on the types of companies that are receiving this financing.
I like data, especially when I find it in unexpected places. Our favorite urban bureaucracy the MTA has a wealth of data on subway and bus ridership just waiting to be parsed here. I suck at graphic design, but I’m just going to throw something out here in hope that it inspires someone with actual Adobe Fireworks skills.
While playing with this data, I came up with one interesting metric in particular — something I call the “Fun Index,” which you can see below. It’s simply a comparison of ridership per subway station on weekdays versus weekends. Initially, the outcome seems obvious: stations that serve offices will be heavily trafficked during weekdays, with a substantial drop-off during weekends.
But what about other stations that don’t serve Midtown East or the Financial District? It may be reasonable to expect those stations’ traffic to be reasonably smooth, as residents use the subway to go to work on weekdays and to fun places on weekends. Since most people go to work every weekday but may not go out every weekend day, you’d expect a slight decrease in traffic on weekends. And that’s exactly what you see in stops in the UES, UWS, Chelsea, Murray Hill and other heavily-residential areas.
But if not to work, where are people going on weekends? Interestingly, two subway stops in Manhattan actually show an increase in traffic on Saturday over an average weekday: Canal Street JMZNQRW6 and Prince Street NRW. If you view this data across all subway stations, you can create a “Fun Index” of sorts that compares New Yorkers’ destinations during weekdays versus weekends. On weekdays, we go to Midtown and the Financial District. On weekends, we go to TriBeCa and SoHo. Those places are simply more fun.
Would love to see what others can do with the MTA’s data.