In this post I’d like to share what lead up to our prelaunch and where we are headed next.

Release early, release often

On day one I said I’m not going to let myself go for many months at a time without anything to show for it. I think it’s key for any business, and a startup in particular, to be regularly delivering value to its customers. This is essential for getting the feedback you need to validate you’re building the right product. It’s also motivating because you are consistently producing results. It keeps you focused on concrete results so you don’t wander, over-engineer, or work on the wrong things.

I decided to do a prelaunch campaign so I could get the message out about Pastime now. Why wait? Let’s start sharing the message and building a base around it. Let’s use that as a way to gain some momentum and refine the message by our launch date.

I found taking this on really challenged me. I went from thinking about implementation details to thinking about the overall customer experience I’m trying to create and how I’m going to sell it to people. It forced me to answer questions like:

In one sentence, what does Pastime do for you?

Who is the target audience and what motivates them to buy?

What key problems does Pastime solve?

How should Pastime make its customers feel?

From there I started getting some results. With the help of my designer, I brainstormed product tag lines, dimensions, logos, and colors. Before I knew it, the elements of a brand started coming together. My goal became to put it all together into a single page brochure that sells the vision. I wanted it to be digitally interactive, too. If you love the concept you should be able to subscribe to get on our list and follow our progress. If you know others who would love it too, it should be easy for you to share Pastime with them.

You can see the final version of our prelaunch brochure below:

View it live at http://pastimebrevard.com. Be sure to subscribe to see the complete experience we put together!

It took three iterations over three weeks to get this done. Check out how the design evolved from my first version:

It’s okay to laugh, although I have to say the Brevard County road sign is a nice touch.

Key Challenges

Some of the most challenging parts of the Pastime prelaunch project were:

Creating a design that appeals to both adults and youth.

We want a youthful look but not one that looks like it’s just for kids. It took us awhile to get something that feels like it’s for both adults and youth.

Creating a design that motivates people to take the next step and subscribe.

To this end, we strive to make the message as clear and concise as possible, and we use active, bold, and inviting words. We aim to connect emotionally with our audience, then invite them to join our list.

Motivating subscribers to share Pastime with their friends.

We send each subscriber a personal email thanking them and calling them to action, we integrate with Facebook for sharing, and we feature a referral incentive that has real value.

Featuring Brevard County, Florida, where we are based.

We are launching here and we will build out the first version of our product here. I believe in supporting and representing the community where you live, and the Seal of Brevard County we incorporated into our design reflects that. We will expand to other areas when we are ready.

How successful will our prelaunch campaign be? Time will tell. Our goal is to launch to over 1,000 Brevard County, Florida subscribers this spring offering a mix of team sports such as flag football, softball, and volleyball. I’m in the process of recruiting the perfect league partners now. We are off to a great start and I’m confident we will knock it out of the park!

I also cited Linode and Amazon Web Services (AWS) as the leading VPS providers. Linode has 40,000 customers, an active and vocal on-line community, and hosts several well-known brands such as TheOnion.com. AWS is used heavily by big Internet companies such as Netflix, Foursquare, Quora, Reddit, and Hootsuite, and is recognized as the leading “public cloud” or “on-demand” computing platform.

Sweet Spots

Both Linode and AWS allow you to lease VPS instances, but their sweet spots are different. Linode employs a more traditional model where you lease one or more instances on a monthly basis at a certain price, and for that price you get a certain amount of CPU power, memory (RAM), network bandwidth, and storage. AWS, on the other hand, takes “pay as you go” to the extreme. Instead of leasing capacity on a monthly basis, Amazon is grounded in an “on-demand” usage model that lets you to lease capacity by the hour. Furthermore, Amazon won’t charge you if you turn your instance off. You are only charged for the hours your instance is actually running.

I said Amazon is a leader and I mean it. They also offer “reserved instances” that allow you to reserve capacity for a one year or three year term to save money when you know the instance will be running most or all of the time. AWS also provides an innovative option called “spot instances” that allows you bid on unused capacity and secure that capacity by the hour at up to a 65% discount. In general, Amazon has the most features of any VPS provider and is the one really pushing the envelope.

My Requirements

In my case, I am just starting out, and I need a VPS to host the web site for my business. I need the site to be up 24x7x365, so any “on-demand” model is less interesting. Rather than planning for what I don’t know, I also want to keep things as simple as possible, focused on what I need to run the first version.

The first version of Pastime is pretty simple. It’s a web application. The client piece is written in HTML5, CSS3, and JavaScript. The server piece is written in Java. The application stores customer data in a relational database. There are several third-party integrations, including integration with a credit card processing system. There will be a handful of offline jobs to run, but most of the system is on-line.

I am confident everything can run on a single commodity server from the outset. The need to scale will be carefully controlled, as I plan to employ a local, grassroots approach to developing customer relationships. I do not plan to mass market straight out of the gate and in general I am not a fan of that approach. When I’m ready for scale, I’ll address scalability then.

Linode Options

Linode’s options are simple and clear. Just visit linode.com and you can see them right there on the front page under “Plans and Availability”. I could likely get by with one “Linode 512″, which is their cheapest VPS priced at $19.99 per month. However, Java can consume quite a bit of memory. Given that, and the fact I’ll also be running a Database Server on this instance, the Linode 1024 plan feels like the best fit. It’s priced at $39.99 a month. I save 10% by paying for a one year term, for a total upfront cost of $432.

AWS Options

Amazon’s options are more numerous and complex. Their VPS instances are grouped by size: “Micro”, “Small”, “Large”, and “Extra Large”, and get more powerful (and expensive) with each size increase. The next dimension is utilization: “On-demand” is the best known, and then you have the reserved levels of “Light”, “Medium”, and “Heavy”. If you know your instance will run for some % of time, the reserved levels let you pay upfront to reduce your effective hourly rate over a 1 or 3 year term. “On-demand” utilization has the highest hourly rate (8.5 cents) but doesn’t require any payment upfront.

Since I need a 24x7x365 server, “Heavy” utilization is the right choice. A “Micro” instance doesn’t provide enough consistent CPU power. “Small” is the right choice.

The battle is set: the Linode 1024 vs. the AWS Small Heavy…

Linode 1024

For $39.99 a month or $432 a year, a single Linode 1024 instance gives me:

• 1 GB of RAM
• 40 GB of Persistent Storage
• 400 GB of Data Transfer

I did some digging and learned I would be sharing a Dual Quad (8 CPU) host with 20 other users. In the best case I would have full access to 4 CPUs, each with 2.27 Ghz of power, for a total of 9.08 Ghz. Worst case, if all 20 users were running at 100% capacity I would get 908 Mhz. Attached local discs are used for storage, and after doing some digging it appears each host houses 4-drive 15K RPM SAS drives in a RAID10 configuration. For network, each Linode host has dual gigabit uplinks (one internet, one local private network). With 20 users, worst case I can expect 100 megabits per second. Only outbound network traffic goes towards your Data Transfer quota, inbound and internal do not.

I’m pretty satisfied with the bank for my buck. 400 GB of is more than enough bandwidth, and so is 40GB of storage. I could always use more RAM but 1 GB should get the job done. The CPU numbers are impressive (up to 4 CPUs!), as is the fact the plan is conservatively limited to 20 users per host.

AWS Small Heavy

Now to Amazon. For $451.45 per year, a single Small Heavy instance gives me:

• 1.7 GB of RAM
• 1 EC2 Compute Unit (or virtual CPU)

This cost was calculated by adding the upfront “Heavy” fee of $276.25 together with the cost of 24x7x365 hourly utilization (which is $175.20, calculated by multiplying 8,760 hours in a year by the reduced rate of 2 cents per hour).

So a comparable EC2 instance with more RAM is slightly more than a Linode 1024, but we have yet to purchase any bandwidth or storage. Both are priced separately.

Network bandwidth costs 12 cents per GB per month, with the first GB free. Lets say I do 50 GB per month by the end of the year (it could happen). That’s $6 per month. If my site takes off, the cost starts really adding up. On the other hand, Linode doesn’t charge me anything else until I significantly exceed 400 GB per month (Linode’s extra bandwidth rate is also cheaper at 10 cents per GB).

Storage costs 10 cents per GB per month plus 10 cents per 1 million I/O requests per month. Lets keep it conservative: if I don’t exceed 1 million requests, that’s $4 per month for 40 GB of storage.

Factoring in network and storage, my “likely” effective monthly cost creeps up somewhere over $39.99. The exact cost is hard to predict and will flucutate based on how many users I have over the course of the year. To do Linode 1024 max levels, I’ll pay about $85 per month.

Amazon, to my knowledge, doesn’t publish much about their underlying infrastructure, so it’s difficult for me to guage how 1 EC2 CPU will perform vs one Linode 1024. I like that Linode publishes how many users you’ll be sharing a host with, and that number goes down with each plan increase.

One of the unique features of Amazon is storage scalability. I can go from 1 GB to 40 GB to 1 TB of storage, all at 10 cents per GB and 10 cents per 1M I/O requests per month. The top Linode plan only gives you 160 GBs of storage. However, with that said, Amazon’s storage system, EBS, is an “off-instance” SAN that won’t be as fast as an attached local disc. Lots of I/O requests will not only be costly, they might not perform very well either.

My Decision

Linode has the best value for what I need today. Amazon has the “on-demand” sweet spot. They make it easy to stand up a virtual machine, do some processing, shut it down, and pay only for what you use. But for an 24x7x365 instance to get you started, Linode delivers more bang for the buck in nearly every area.

I choose Linode!

Getting this done has involved two activities; first, the development of key branding elements such as logos and web site design; second, the provisioning of the infrastructure required to host the service.

This note takes you through the general hosting options and some of the first choices I made. In a second installment, I will share my evaluation of my candidate hosting providers and reveal the winner.

General Options

First, I knew I wasn’t going to be buying hardware upfront. At this stage, I have no idea how much capacity I will require, and I have no interest in maintaining my own data center. For that reason, I began the process of evaluating and selecting a hosting provider.

Not all hosting providers are created equal, and their hosting plans are generally grouped into one of three categories: Shared, Virtual Private Server (VPS), and Dedicated.

Shared

With shared hosting, you share the capacity of a single server with other users. What’s shared includes the server’s operating system (OS), network bandwidth, and memory (RAM). The provider dictates the OS and the software available to you. “Soft limits” are imposed on server utilization, which means if one user starts using a lot of bandwidth or memory, other users could be negatively effected. If you happen to be the greedy user, there is the risk the provider will shut you down without warning.

A shared host is generally the simplest and cheapest option (as low as $5.95 a month). This may be appropriate for simple websites such as personal blogs, but not for custom applications that require more control. For example, in my case, I am developing a custom Java application, so I need control over the software required to run Java effectively.

Shared hosting is inherently less secure. The risk always exists that a user, or a hacker acting as a user, could exploit an OS vulnerability and gain access to other user accounts.

Leading shared hosting providers include BlueHost and HostGator. If you go the shared route, it’s important to ask how many other users you’ll be sharing capacity with. Be wary of providers that oversell their hosts.

Virtual Private Server (VPS)

A VPS provider allows you to lease your own “virtual server” that you can do whatever you want with, so long as it is legal. You pick the operating system and you install the software you need. You still share a physical host with other users, but each virtual instance running on that host is securely sandboxed and has hard limits around what resources it is allowed to consume. For example, your VPS might be limited to 512MB of RAM. The virtualization software running on the host infrastructure prevents your instance from exceeding this limit and negatively impacting other instances.

Another benefit of VPS is the ability to scale up or down quickly. When running in a virtualized environment, it is easy to upgrade to another tier that has more power.

Linode and Amazon Web Services (AWS) are leading VPS providers, with the latter recognized as the leading “public cloud” or “on-demand” computing platform.

Dedicated Servers

A dedicated server is as close to buying your own box as you can get. You lease physical hardware and you’re the only user. You know what you’re going to get, but you typically pay more upfront and your ability to scale to meet spikes in demand is more limited.

Rackspace and Hetzner are leaders in dedicated server hosting.

Managed vs. Unmanaged

One additional dimension to hosting is “managed vs. unmanaged”. With a unmanaged server, the provider will handle any hardware or virtualization problems, but it is your responsibility if something goes wrong with the software running on your server. You are expected to deal with software installation, configuration, upgrades, monitoring, and security. With a managed server, you are effectively leasing your own sysadmin that does this stuff for you. As you might guess, managed servers are more expensive.

Rackspace, Contegix, and Network Redux are three leading managed hosting providers. As a side note, we used Contegix in the early years of SpringSource and were quite happy.

Hybrid

It’s worth noting it is possible to combine Dedicated Servers with VPS. Rackspace is a single vendor that supports such a “hybrid” approach.

Narrowing Down the Options

For my needs, Shared hosting is clearly out. A Dedicated server doesn’t make sense either, because as a startup I have no idea how much capacity I will require. VPS is the right choice due to its low cost and flexibility, and I want unmanaged because I want to learn this stuff.

Linode and Amazon Web Services (AWS) lead on VPS. So it’s between them…

TO BE CONTINUED

At the most basic level, my goal is to inspire people to spend time doing what they love. Instead of being a sideline observer, the product should inspire you to jump back in. The initial focus is on team sports such as football, softball, basketball, soccer, and hockey.

Ideas for names have been all over the map. “All Play” was one, “Real Sports” was another, “Sport Verse” a third. One day, while trying to explain my idea to my wife Keri, I caught myself saying “Instead of sitting around surfing the web or playing fantasy games, I want this thing to get you out there. I want to be out playing sports with my team mates, that’s what I enjoy… that’s my pastime.” Then it hit me: pastime. It’s simple, clear, brandable, and closely associated with leisure and recreation. It has a sweet ring to it that fits nicely in the category of social networks alongside Facebook and Twitter, which is exactly where I hope to be positioned someday. On the other hand, it is a commonly misspelled word (passtime, pasttime). However, I think its other strengths makeup for that weakness.

After selecting the name, I took several steps to secure rights to it.

Trademark

I filed a trademark with the United States Patent Office (USPTO). Before I did this, I determined what category my product falls in and searched that category for potential conflicts. Everything looked clear, so I had LegalZoom file the trademark on my behalf with “intent to use”. Recently I got good news–the mark has been accepted and published to the Official Gazette. Soon I’ll receive a “Notice of Allowance” that will give me six months to file a “Statement of Use” that demonstrates the mark being used in commerce.

Internet Domain Name

I leased rights to the pastime.com domain. Getting this done required some careful negotiation. pastime.com is not actively used, but it isn’t readily available either. I learned the name was in fact for sale, and put in an offer to buy. The owner came back with a firm price that was more than I was willing to pay. We negotiated a lease-to-own arrangement that allows me to pay a reasonable amount now, and gives me the option to purchase at a later date. I needed to carefully design the terms for my situation–it’s too early to know if the business will gain traction, so it’s too early to invest a significant amount of money in an intangible asset. In the future, I’ll know where the business is headed and I’ll be able to make an informed decision on whether to purchase then.

Doing Business As (DBA) Trade Name

I registered “Pastime” as a trade name with the state of Florida. When I incorporated, I was unable to get “Pastime, LLC” because a “Pastime, Inc.” already exists. So my company name is actually “Pastime Connect, LLC”. However, my company can do business in Florida as “Pastime” because of the trade name. Furthermore, It has rights to the name “Pastime” in its category because of the trademark.

In my last company, we had a branding problem in the early days. Our company name “Interface21″ did not map to our product name “Spring”. I remember being at the JavaOne conference in 2006. Everyone there knew Spring, but only a small percentage knew Interface21. That wasn’t good for our business, especially when other organizations started selling their own Spring products. In 2007, we changed our name to SpringSource (Spring itself was not available since there was already a technology company that had rights to that name). That made a huge difference–people immediately knew we were the company behind Spring, and we got a lot more business. The importance of clear branding was an important lesson for me and one I tried to keep in mind throughout this process.

What’s Next

Our next step with branding is a logo and theme for pastime.com. That will include an icon (think iPhone app), text logo (comparable in function to the logo you see on the top left of facebook.com), and a site color scheme. For a little fun, we’re also working on an idea for a mascot inspired by the Github Octocat. I hope to have the first version of a branded pastime.com up with a Coming Soon page early next year!

I had four choices for the structure of my business: sole proprietorship, LLC, S Corporation, or C Corporation. I want personal liability protection, so that immediately ruled out sole proprietorship. I want to minimize tax liability, and as you will see, what structure is best for that can vary.

Taxes

LLCs have the advantage of not being taxed, as business income “flows through” to the owners who then pay self-employment tax (~15%) and personal income tax (up to 35% depending on bracket). Similarly, a S Corporation is not taxed. With a S Corp, there is also no self-employment tax, instead you pay a payroll tax that is approximately 15% of your salary. A C Corporation, on the other hand, is required to pay corporate income tax. This leads to the so called “double tax”, as the company pays tax on its income and each owner pays tax on his salary and any dividends.

So this is where it gets a little complicated. The structure that is best from a tax perspective can shift depending upon how profitable your company is. In the early stages, when income is low, a LLC treated as a “disregarded entity” tends to be the most attractive. You simply pay the 15% self-employment tax and a personal income tax that depends on your tax bracket. If your business has losses, you can use them to offset other sources of personal income.

S Corp Tax Treatment

As income grows, S Corp tax treatment can become more attractive as it allows you to balance income between salary and dividends to better manage the amount owed in employment taxes. This is illustrated by the following example:

Suppose my 2012 income is $50,000. As a disregarded LLC, I will pay $7,500 in self-employment taxes. As a S Corp employee with a fair salary of $50,000, I will pay about the same in payroll taxes.

Now lets say my 2013 income jumps to $100,000 while my fair salary remains the same. As a disregarded LLC, I will pay $15,000 in self-employment taxes. As a S Corp employee, I will continue to pay $7,500 in payroll taxes. S Corp tax treatment saves me around $7,500.

C Corp Tax Treatment

As your income continues to increase, personal income tax can become an issue for two reasons. First, you get bumped up into a higher tax bracket–the highest being 35% with the potential to increase to 39% if and when the Bush tax cuts expire. Second, neither a LLC or a S Corp can retain income for re-investing–all income must be reported and is subject to tax. As a result, in some cases a C Corp structure offers the most attractive tax treatment. This is because the effective corporate tax rate can be lower than your personal tax rate and because income can be retained by the corporation to reinvest.

To illustrate this, lets run some example numbers again:

Suppose my 2012 income is $50,000 and my fair salary is $50,000. As a S Corp, I’ll be looking at a tax of $28,000 on $100,000 of personal income (28%).

As a C Corp, I’ll pay $7,500 on $50,000 of corporate income (15%). If I retain those earnings, I only pay $12,500 in personal income tax (25% of my $50,000 salary). That saves me $8,000.

Uncle Sam can get you on retained earnings, though. Retain more than $250,000 and you risk being slapped with a 39% retained earning tax.

Two other notes about C Corps: First, a C Corp may deduct 100% of the health insurance it pays for its employees. Second, owners of a C Corp that qualifies as a small business can claim a capital gains tax exemption upon disposition if their stock has been held for five or more years.

Taxation Summary

So what structure wins for taxes? It depends. LLCs offer the most flexible structure. By default, they are treated as “disregarded” and subject to self-employment tax. However, you can elect to have a LLC treated as either a S Corp or a C Corp for tax purposes, depending on what is most appropriate for your situation. In general, this is one of the advantages of LLC’s–they’re flexible.

For a rapidly growing company, C Corp tax treatment tends to win out. This might seem counter-intuitive because you always hear about how corporation’s are “double-taxed”. Yet, as we have seen, corporate income tax rates can be less than personal income tax rates, and the ability to deduct taxable income and retain earnings for growth can make a difference. A shrewd C Corp owner can work to reduce his or her tax liability by paying himself a modest salary and reinvesting profits. Furthermore, many C Corps find ways to exploit various loopholes that get them out of paying income tax all together. As a side note, I believe this to be morally wrong and not fair to the corporations that do pay their share.

While on the subject of right and wrong, I also believe it is wrong for our congressional representatives to fight cuts to payroll taxes that would help all employees (and especially those with lower incomes), while opposing closing corporate income tax loopholes or increasing income tax rates for the most profitable individuals.

Overhead

Besides liability protection and tax favorability, I also want low overhead and flexibility. LLCs win here. Corporations are subject to formalities such as holding annual meetings of shareholders and directors. Processing payroll is also quite complicated. LLCs don’t require any of this. And, as we have seen, LLCs have the flexibility to be treated as corporation for tax purposes, if desired.

Equity

One advantage of Corporations is they can issue stock and stock options. LLCs have a similiar concept in “membership interests” but it is somewhat awkward and less well understood. As an example of this, suppose, as an incentive an LLC awards a key employee some “capital interest”. Upon receipt of the interest, that employee becomes a member of the LLC. In a LLC with default tax treatment, you cannot be an employee and a member at the same time. You go from being an employee on payroll receiving a salary reported on a W-2, to being a partner deriving self-employment income from profits based on your % interest reported on a K-1. This is an awkward transition to make. With a Corporation, nothing really changes. You remain an employee, you now just also happen to have a stake in the company.

C Corporations are generally more suitable for outside investment. I know several VCs, for example, that won’t invest in LLCs.

My Decision

What did I go with? I decided a LLC structure was right for the early stages of Pastime. It provides liability protection, which is essential. It provides simplicity and tax advantages along with the flexibility to be taxed as a corporation later on if desired. The admin overhead is low. I will not be accepting VC funding, so I do not care about that. And when I have the opportunity to hire, I expect to favor compensation via fair salary and benefits to issuing stock.

If I was planning to raise outside funding or issue stock to employees, I would have formed a C Corp. I have the option to convert in the future if I need to, and it is something I will continue to evaluate.

So there we have it! I am now the owner of Pastime Connect, LLC, formed in the fine state of Florida. It is a single-member LLC. I am the member/manager and my wife Keri is also an officer. It is authorized by Brevard County, Florida to do business under the tradename “Pastime”. We have no employees yet, but we hope to make that all important first hire in 2012.

I hope this note has provided you with useful insight into my startup incorporation experience!

My new venture is called Pastime. It aims to be a service for people like you that enjoy playing team sports. This note captures my goals, values, and what I have accomplished so far. There is a lot involved in starting a company, in getting organized and focused, and I would like to document and share what I learn.

First, my goal is to build a sustainable business based here in Brevard County, Florida. I desire to build something that makes a positive impact in my local community and something I would be proud to have my kids run someday. I am not interested in chasing trends or selling out to another company. So with that said:

1. We will not accept outside funding; we will grow organically by earning our own money.
2. We will focus on the business model from day one; none of this “we’ll figure out how to make money later” stuff
3. When we have the opportunity to hire, we will hire great people by fostering a culture that gives product ownership and rewards performance; no “lets pay dirt cheap salaries and instead hand out incentives to sell the company”

In essence, I want to be Bon Jovi, not Breaking Benjamin.

For my first month, I have focused on two things: 1. getting the company setup, and 2. developing a business plan. Specifically, I have incorporated in the state of Florida, opened a bank account, filed a trademark, secured a domain name, purchased health insurance, and selected an accounting service. In terms of the business plan, it is split into three phases and I have enlisted some help from the Melbourne, Florida Chamber of Commerce in drafting it and building local support. Last but not least, my daughter Annabelle has been working on a company logo, slogan, and mascot.

I plan to go into all the details about my first month in my next note. Can you tell I am excited already?