As the team began growing and changing, people started asking me if they could take Friday off or if a working week at home was acceptable.  Eventually came a call for a vacation policy so there were clear guidelines for time away from the office.  We debated — 2 weeks, 3 weeks, working holidays, etc.  Realizing we were splitting hairs, we went back to the most important thing for Increo–that we have an energetic environment and that people make decisions to work when they are most productive and take some time off when they need a recharge.

With that in mind, we made the decision to have the Official Increo Vacation Policy be to use your best judgment.  In other words, there is no “policy” per se.  Let the team know in advance, get your stuff done, and take the time you need to ensure you can contribute daily to Increo’s sense of energy and excitement.  Turns out, given two weeks of vacation time, most people take two weeks.  Given unlimited vacation time, we’ll almost always take just the time we need.

We’re certainly not the first to have no vacation policy–companies before us have tried; some failed, some succeeded.  But for right now, it works for us, makes Increo an exciting place to work and demonstrates an invaluable trust in the team.

If you’re fortunate enough to still be a student, here are 5 reasons why you should strongly consider jumping right in:

5. Quickly discover if entrepreneurship is for you.
The easiest way to discover if startups are for you, or not, is to try it out while you still have flexibility. Put together a small team of friends on campus and start working on something. Learn how to pitch it. Talk to potential customers. Squeeze meetings in between classes and work in between homework. If you can balance the load and still have fun doing it, that’s a great sign!

4. You have credibility.
As a student, you have the opportunity to reach out to anyone and everyone to try to learn from them. Ask for a phone call, or even a meeting. You’ll be surprised how often the answer is “Yes”. Use this opportunity to understand as much as you can about your target market and the competitive products. What works for people and what doesn’t?

This credibility can also extend to talks with investors who often hope of finding the next Google or Facebook in the form of a young, passionate, entrepreneur.

3. Fantastic resources.
Your college or university can likely offer quite a bit of support when it comes to starting a company. Talk to professors who have done it before. Use the school’s library for further research into the field. Get involved with business and startup-related student groups. All of these resources will disappear once you leave campus.

2. The best people in the world are right there.
Look around you. When seeking out specific skill sets as you form a project team, reach out to friends across campus. By recruiting in a known environment, you remove a lot of the uncertainties around teammates, and it’s quite likely that you’ll be able to find the person with the right background at your school.

1. You have nothing to lose!
The most important thing to consider when choosing what to do upon leaving school is to remember that you’ll never be in this situation again. You don’t have a family to support or a daily routine to maintain, so why not take the chance and try to build something meaningful? What is the worst that could happen? If the enterprise does not work, you’ve gained substantial experience in the process that you can take with you to your next job.

In short, seriously consider jumping into the startup world as you leave campus. You might just discover The Next Big Thing!

Silicon Valley is unique in many ways and circumstances might be different depending on the climate and culture you live in. With luck, however, the core message is still relevant.

(And if you’re still on the fence, think about joining one of the many startups already out there. What better way to join a passionate team and make a difference?)

Coming from a world of strictly making software and leaving the pitch to an already-built community of tens of millions of users, it took a while to adjust to the idea of having to always explain what the product is when you start talking about it. Whether you’re at a roundtable with VCs, at a job fair, or just talking to somebody socially, it’s always useful to have that quick pitch crystallized in your head.

In addition to the elevator pitch, you also need its slightly more difficult cousin, the bar pitch. It’s similar in nature, except given the noisy environment of a bar, you have to refine it even further to include only short, clear words that will still make sense when the other party misses half of what you say. You have to phrase it to intrigue people who overhear you instead of focusing on a single listener. You have the opportunity to add wild gesticulations and other physical means of making yourself memorable when those would come across as completely inappropriate in an office setting.

For example, the following elevator pitch that I’ve used recently:

Hi, I’m Jim from Increo Solutions, a company that makes software to help teams innovate. Our first product is Backboard, a tool that lets you get feedback on your documents, slides, graphics, or just about anything else. You just upload a document, type in email addresses to share it with, and everybody gets a link where they can go and leave feedback and mark up the document. It’s really simple, and it easily solves the problem of emailing around files all day and consolidating changes and responses.

might become more like this when delivered at a bar:

So here’s the deal: say you’re working on a PowerPoint deck with a bunch of your co-workers and you want to get their feedback. What do you do? You probably, say, email the file around to like a dozen people and then each person goes and makes their own version and now you have a thread with fifty replies and fifteen new versions and it’s like whoa, what’s going on? So we make Backboard: you upload the slides, everybody gets a link to it, and then they all come and see the document right there in the browser and they can type in their feedback or mark it up with a little red pen tool. Everything’s centralized, sane, and remains under your control.

For all social situations other than riding in an elevator with that general partner, you may be surprised to find that the bar pitch works even better at getting your point across. It’s certainly been a more effective message with us for Backboard.

It all goes to show you in the end that you must consider your audience for every message and tailor it accordingly. Too often, Silicon Valley types forget this and get stuck in Elevator Pitch Mode, unable to speak to anybody without launching in to a diatribe about how they’re going to change the world. Learn your bar pitches to keep your message fresh and approachable.

The hackathon was “distributed” in that all the participating teams stayed at their respective company locations and communication was done via an IRC chatroom and dedicated conference calls. For goodwill, SocialText organized it as a competition in which the winning team would win a gift certificate toward kiva.org.

After the opening conference call shenanigans, we dove right into coding. Now, I love what I’m currently working on for Backboard, but getting to work on a miniature side-project was very refreshing. I got to play with technologies I hadn’t yet explored and learned how to make and deploy widgets. One of the tricky things about creating a widget is that it doesn’t sit on your site, so we had to do some refining of our own API in order to format and grab the data we wanted to display. After that, the rest was fun and easy: it turns out that making a widget is as simple as wrapping a little HTML and JavaScript in XML file.

At 4:30 pm (the “deadline” to turn in our projects) we dialed back into the conference call to join the show and tell session and presented our three completed Widgets. “My Backboards” is simply a listing of your backboards with recent activity. “Get Feedback” allows you to upload a document, set permissions, and create a backboard all within the widget. “LOLPirates,” Jim’s masterpiece, cycles through incredibly cute pictures of cats dressed up as pirates. Like Fluffy-beard here.

For young ambitious companies, it’s often hard to be the treated as the underdog, but at Increo we draw confidence from the supportive startup community and believe in giving back in any way we can. Our participation in Widget Wednesday was just one example of our philosophy — we fostered relationships with other startups and simultaneously strengthened Backboard and SocialText Dashboard.

Great.  That is awesome, it really is.  But it’s not enough, and even people who should know better often forget.

Determining whether a company is investment-worthy (for you and your time, or for someone else and their money) is certainly a complex process, a balanced combination of team, idea, market, and technology.  But you can make it easier–for both you and your investors–even in this time of economic uncertainty.

Before you go out to raise money, find concrete, indisputable evidence that your company will succeed, and succeed ahead of the curve.

Bad ideas with evidence of success will probably be able to raise a round or two, but the “next big thing” without evidence of success will look just like everybody else’s “next big thing.”  With some concrete evidence, your concept stands out from the crowd as an opportunity rather than an annoyance.

It’s against our nature as entrepreneurs.  Our job is to be stubbornly optimistic about our concepts and ideas and to push toward their adoption with everything we have.  Gathering evidence of something we know will succeed seems like an inefficient use of time, or it seems impossible until we have everything built.

What is evidence?  I think of evidence as real users, real customers, or signed deals.  With a solid (and growing) pool of users, your consumer startup looks like a compelling investment opportunity over someone else’s idea.  A couple of deals are the best proof of concept that an enterprise play will work.

Of course, this isn’t always possible.  In a pinch, evidence can be anything that can reasonably stand in for users, customers or deals.  Beta customers are a good example, as are customers on something similar you built (maybe attract people to a very small part of your product that you can launch early).  Live software helps, and so does a proven advisor or team member.  The further away you are from real customers, the harder the sell, but reasonable evidence helps you stand out from the noise.

Combine this evidence with a technology, a market, and an idea that hasn’t been over-pitched (ask a lot of people) and is consistent with someone’s view of the future (again, ask a lot of people), and you’ll position yourself most optimally.

I started at Increo last Monday the 3^rd. I’ve taken over the resident newbie position from Henry and have had many questions over the past week. The thing that has struck me the most in this short time is the ability to make decisions and move quickly.

Prior to joining Increo I was working at a large company with an established product. One of the things about having an established product is that you also have established processes and everyone has their established ways of doing things. People would say that you were ‘ahead of the curve’ if you could get to the point where you could check out the code and build it before the end of the first week. These processes and tools come in handy when you want to maintain the quality of a large codebase or facilitate interactions with other teams, however, they also can slow the development process with their overhead.

Increo, being a small company, avoids most of the overhead that one experiences in a large company. It also has the important side effect of allowing new people to come up to speed quickly since there are not a countless number of ‘standard’ tools to learn. I was able to check in more code for Backboard in my first week here at Increo than I was in the first month at my previous employer.

Please don’t mistake this as a blanket bashing of having established processes and tools since it is important to standardize the way design decisions and code changes get made to keep large projects running smoothly. I’m just saying that upon joining Increo, I was struck by the ability to move quickly by not using a strict process. Being small provides the ability to be flexible, adapt and get stuff done.

Then again, we’re also not the types to spend fourteen hours straight sitting down; there’s enough action in the office that we’ll be moving around enough to stay healthy.

In fact, while having the world’s most comfortable chair would certainly be enjoyable, we took a different tack: what was the best, most ergonomic setup we could create for six employees for less than the cost of one Embody chair?

Step 1: Desk placement

The main building at the Increo Campus has a floor plan something like this:

where the large rectangle is a single room and each of the smaller rectangles is a desk. The three gray rectangles are the doors to the room. Not only does this surprisingly skewed design maximize traffic flow, it has several distinct ergonomic and functional advantages.

• Nobody is looking at a wall.

  Looking at walls prevents you from relaxing your eyes by periodically refocusing on a distant object. It also inhibits verbal communication; while this may or may not be a good thing, having to physically move to talk to someone has significant implications for productivity.

• No desk is parallel to a window.

  The strange angles give us a great compromise between putting everybody together and reducing glare from the windows next to the door on the bottom wall.

• All four walls are still accessible.

  Whiteboards. Silicon Valley loves them, because they’re critical to forming ideas. While the construction of our space doesn’t allow for whiteboards on all four walls, we have them on two and may put them on three eventually. Being able to use them without having to run in to desks is incredibly fortunate for us.

Step 2: Lighting

Like most offices in America, Increo’s office has harsh overhead fluorescent lighting, designed in the 1960s to meet the needs of people pushing paper all day long. This was great in the 1960s, when the primary mode of work of America’s office workers was pushing paper. Today, though, the average office worker spends most of his or her time using a computer and its light-emitting screen, making the overhead light level entirely too much.

Pro tip: turn off your overhead office lights!

Having no light, though, makes it too dark; now you have a huge light level contrast between the monitor and the rest of the office, which creates more eye strain than it solves.

Using our walls for whiteboards prevents us from lighting the perimeter of the office, so we decided instead to put floor lamps in the corners and a task lamp on each desk. IKEA has a store just down the street, so we ended up with this “ANTIFONI” number:

It has a halogen bulb (low energy use and good light spectrum at the same time!), it adjusts to really any angle, and the construction feels so much more solid than the standard flimsy Anglepoise rip-off.

Between lighting and clever layout, the Increo office is pretty people-friendly. As we grow, we may have more opportunities to implement some classic suggestions about work environment, but for now, we have a highly functional one that allows us to concentrate without wearing ourselves out every day.

For Backboard, we have our server set up to allocate just 16 MB of RAM to PHP for the execution of each request. This allows us maximize the number of Apache processes we can run on a single server and thus maximize the number of users we support. To do this, though, means that the site has to be pretty light on its memory requirements.

Backboard, as a CakePHP application, has a theoretical minimum requirement for RAM: namely, the amount needed to execute a base installation of Cake. In our testing, this falls between 4 and 6 MB. The goal, of course, is to minimize anything above and beyond this.

Here are some concrete steps you can take to limit the RAM needs of your site:

1. Install a PHP opcode cache, such as eAccelerator, APC, or Xcache.
This prevents the server from having to recompile your PHP scripts on every request, a process that is both RAM- and CPU-intensive. The three options I’ve listed are relatively interchangeable and relatively comparable in terms of performance gains. We went with the first one. eAccelerator took just a few minutes to install and required no code changes, so it is time well spent!

2. Use CakePHP Components wisely.
In Cake, Components are relatively heavy - they add a fair bit of processing overhead and memory use to your application. They definitely have their place, but take care to include only the ones you really need, since they are loaded on every request.

3. Load CakePHP Vendor files lazily.
Cake has great support for vendor code - that is, 3rd party libraries that you call from your application to avoid re-implementing the wheel. Be wary, however, of calling App::import(’Vendor’, …); at the top of any of your controllers or components. What this does is load the vendor libraries for every page on your site, even if that code is not used! Instead, place your App::import() statement immediately before you use the library. That way, it is only loaded if you actually make the call!

4. Controllers are the answer!
The best way to minimize the amount of RAM a given request requires is to separate your actions into multiple controllers, based on what components, models, and helpers they require. This is always a good idea for clarity of code, but with Cake it is beneficial to take it a step further if speed and memory are a concern.

To provide a concrete example, let’s say you want the home page of your site to be as fast as possible. (Sounds reasonable, right?) If the page is not entirely static content, it has to go through a controller, so the optimal technique is to place it in its own controller and only include the components that it needs. By so doing, you eliminate all unnecessary code from the request. Now, it would be ridiculous to place every action in its own controller, but it should be pretty easy to group your actions into a few smaller clusters.

Hopefully these tips provide a launching point for your optimization efforts. What other tricks have you discovered?