The Affiliate Numbers

• 50% Of the total sale
• $100 Payout OR credit to your account at any threshold
• 60 Day mature period

The numbers are pretty straight forward – the main gotchas are that there is a $100 payout if you want cash or at any time you can have the affiliate commissions turned into account credit. The other gotcha is a 60 day mature period which is pretty normal for affiliate programs – it helps ensure that the commissions are legitimate and don’t cancel right after signing up.

The Potential

Egg

• Cost: $5.95/month
• Commission: $2.98

Tadpole

• Cost: $7.95/month
• Commission: $3.98

Business

• Cost: $12.95/month
• Commission: $6.48

These numbers are based off monthly subscriptions (more on that shortly). They’re indeed smaller numbers so reaching the $100 payout will be a bit more difficult, however if you decide to use it for account credit you’ll most likely be able to pay for several months of hosting simply for referring a couple of people.

The real money is the fact that the majority of our customers don’t signup for the monthly term but typically choose annual or biennial terms of payment.

Egg

• Cost: $83.40/year
• Commission: $41.70

Tadpole

• Cost: $107.40/year
• Commission: $53.70

Business

• Cost: 167.40/year
• Commission: $83.70

Or if they choose a 2 year pre-payment

Egg

• Cost: $142.80/2years
• Commission: $71.40

Tadpole

• Cost: $190.80/2years
• Commission: $95.40

Business

• Cost: $310.80/2years
• Commission: $155.40

As you can see the numbers really start to shine now. You could request a payout with as little as a single referral or most likely a couple of referrals. People are always purchasing longer payment terms to get a better deal and because of this we believe our affiliate program is well rounded and has a decent payout.

If you have any questions regarding our affiliate program please don’t hesitate to contact us via our support page.

Marketing is a big key in what makes a web host very successful.  Word of mouth only gets you so far especially when you do not have much of a customer base to promote your services to begin with.  You do not typically see a snow ball effect of customers referring others until you have at least 500 users which obviously Frog Host does not have.  Even at 500 why settle with a steady stream of growth at that point?  Why not turn that 500 users into 1000 in two months instead of a year?

So our first step of the marketing push will be to find ourselves a new artist and designer so we can start getting the necessary marketing materials created.  Unfortunately our previous artist left us high and dry and ran with some money we paid them for some new work.  So I’ll be finding ourselves a new artist and possibly even just html/css guy in general as having a single person doing it all would be best.  Previously we were using different people for different aspects which can be a pain to deal with.  The banners and various other images are the first thing people see.  That could be on a web site we’re advertising on as well as affiliates who are hoping to cash in on our potentially big payouts.

So our marketing push you’ll probably see Frog Host advertised on various web hosting forums as well as being on miscellaneous web sites potentially some who are hosted by us who Hawk Host hosts would would be great candidates to advertise on.  So if you have a site you think would make sense for us to advertise on and you’re hosted by us just send me an email which would be tony then the at froghost.com portion.

The other things we’ve been working on is just the forum now has a nice pretty theme thanks to Brian searching the internet for a green one. Then just miscellaneous fixes to our CPU tracking system and various other internal aspects being worked on.  So things are slowly being worked on and improved on but it is indeed a slow process.

MySQL usage has to be tracked as integers as the MySQL build we use to do this tracks as such.  It would be great if that was not the case but it is so what we’re seeing is users not even using any CPU according to their panels.  That could potentially lead to confusion among users as even if you used 0.99 until you reach 1 you’re going to actually still have 0.00 reported.  Now we could stop tracking usage by day and reset it more often so maybe they’d eventually hit the usage we’d want.  This though increases overhead and would no longer provide daily usage to users who use enough CPU for it to matter.

I guess in the end it’s not working as well as planned at least for low usage users.  If we had higher usage users whom are the people we’re targeting the CPU usage statistics at it would show it’s real use.  The plan from the start was to show the high usage users yes you’re using a lot here’s our system saying so.  We’ve just had the luck to have people who are receiving 20 visitors a day and not 10,000 that would show up as a significant amount on our servers.  I guess it just shows you the majority of users using unlimited hosting or any shared hosting for that matter use nearly no server resources.  It’s a few users who are costly for the web host and it’s the low usage users who are paying for that usage as using 20% of a servers CPU is a lot when you think about how many users on a server.

We received one order on Saturday just 10 minutes after we had the site up and available to everyone.  I’ll be honest I was surprised but at the same time exciting to think people were waiting just to come and sign up once it was live.  Sunday during the day we added another order once again appears to be word of mouth from a user who was looking forward to trying out the Frog Host offering.  Finally near the end of Sunday we added the third customer marking the end of the day and a surprising start when no advertising was done at all.  On Monday we had our first fraudulent order as well so really does not take time to see fraud orders coming in.

Along with the orders we’ve seen users signing up for our affiliate program already which pays out 50% of the first payment.  So that could mean if a user signed up for a plan that cost $100 for a year the person would get $50 commission which is comparable to the biggest providers out there.  I wish we could offer a straight $50 per sign up but unfortunately we’re simply not big enough to have those kind of payouts when users might only be with us a month.  So for the affiliate front we currently have three thus far bringing in a total of 29 visitors which is decent considering once again no advertising so these people were waiting for the launch to sign up.

We’re pretty excited here as we’re actually going to start doing advertising which should hopefully mean a heck of a lot more sign ups.  Even at this pace though we might see 45 orders this month which for a normal company that sort of amount on a launch would be very good.  We however will not be happy until we’re seeing at least as many orders as Hawk Host receives which I cannot reveal but it’s a significant amount.

I hope this provides even more insight as to what goes on with a web hosting company right from the start.

1. Skype incoming / outgoing number
2. Personal cell
3. Work cell
4. Google Voice number (goes to all 3)
5. Peek (NEW)

Some people might think this is a bit excessive, and I tend to agree. But when I saw the Peek I thought to myself: “This may be the perfect device for pure Frog Host / Hawk Host E-mail and SMS”. So after Tony shot down the idea as a work expense because I supposedly have enough toys I dropped the $59 for the Pronto + $20 service to give it a whirl. So here’s my initial impressions after owning it for a little over 24 hours.

What Is it?

Well the Peek is essentially a small simple device designed to purely handle e-mail (and optional upgrade to unlimited text messaging). It’s aimed to the e-mail junkie as well as small businesses as an alternative to the phone expensive phone plans.

Look & Feel

When I first opened the packaging I thought it was a nice looking design. It was simple and sleek. The keyboard appeared nicely spaced and the front of it felt like it had a slight rober cover / keypads. After booting it up I had found out that the keyboard was LED backlit with a nice blue color.

Initial Setup

One thing they advertise is how easy it is to setup. When you first turn it on you’re asked to register your e-mail account – it simply asks for your name, e-mail, and password. Now they’ve configured it on their end to properly connect to common mail services such as Gmail, Hotmail, Yahoo, etc – I believe if you type in a normal POP or IMAP enabled mail server it will attempt to connect with the credentials provided and let you in – though I did it slightly differently (more on that later).

As an initial test I decided to try my Gmail address.. namely because I’m slightly paranoid and wasn’t sure how they tried to connect to the mail server.. POP? IMAP? SSL/TLS? Normal? After entering my Gmail information it started downloading my e-mails and I could send / receive e-mails from my Gmail account. Pretty neat – the UI is pretty simple and intuitive. There is an occasional sluggishness - but for an entry cost of $59 for the expensive version and $20 for the cheaper version you can’t really complain.

Now after that I dug around in their settings (which has very few options) to see if I could modify how it connects to the mail server / tweak the ports and so forth. To my dismay there was nothing of that nature available. After digging around their forums and support for awhile I bit the bullet and decided to make a post on their forums.

My Accidental Encounter With Support

I posted on their forums asking if I could get more information on how it’s connecting to the mail servers / what is it defaulting to. To my surprise a few minutes after posting a staff member replied asking what e-mail provider I was using. After explaining I run my own mail servers and wanted to use SSL/TLS he gave me his e-mail and said he could tweak the settings so they could use what I wanted.

After sending an e-mail explaining the settings I need for my work e-mails he replied back within a few minutes double checking the MX records he was about to add on their end. There was a small issue so I clarified it with him and he then replied back a few minutes later telling me to add the accounts in question to verify they worked. I did so and to my joy they worked accordingly.

I soon noticed e-mails were coming in somewhat slowly, so I e-mailed the same suppose technician asking how it’s done on their end. He replied back saying that by default it pulls e-mails via IMAP every couple of minutes or if my mail servers suppported IMAP IDLE he could set me up for that. Since our servers do support IMAP IDLE I had him set that up and within a few minutes I had push e-mail to my Peek.

Now I admit I haven’t verified that they are indeed connecting with the settings I provided – but due to how we’re setup internally it wouldn’t have gotten this for without modifying the MX records manually – so I’m fairly confident .

I was soon disappointed to find out that it does’t currently sync with your IMAP server – this is a bit of a turn off due to the large amount of e-mails we receive and it’s quite a handy feature to have to avoid reading the same e-mails over and over.

Text Messaging

I of course was interested on how text messaging worked with the device since I ponied up the extra $40 purely for that functionality. To send a text message you simply type the 10 digit phone number and send the message. To my dismay it sends it from a phone number located in California.. it appears to use the same one each time, though I highly doubt this number is unique to you. When the person replies back it pipes it back to your Peek device. I haven’t dug into how the text messaging functions quite yet, though it appears to “get the job done”. What I’m concerned about is if I’ll have issues using it as a means of notifications for servers and alerts (which if I can’t is going to be a real killer for me).

Conclusion

If you’re running a small business and e-mail is crucial but you really don’t want to pony up for a smart phone / keep your business separate I would recommend the Peek. The price is decent, though the monthly subscription is a bit steep for what it offers. I intend on keeping it for a few months to see how much I actually utilize it – thus far it’s been a handy tool, though we’ll see once I fully integrate it into all of my things. The support was a nice touch – I honestly was horribly disappointed when I learned I couldn’t tweak the settings to use SSL/TLS but was alleviated to learn they could tweak it for you.

If you have any questions don’t hesitate to ask .

So for the server it’s up and configured and we’ve been testing it for the past 72 hours and making sure it’s configured 100% properly for the launch.  That it has all the initial launch software and we’re not going to be making changes when customers on it.  Although I say that even with Hawk Host we always forget something or something is not being replicated out and we manually add it then forget to add it to our deployment scripts for next time.

For the web site we’ve been real busy bees on January 31st we had 14 code revisions posted to it’s repository.  It feels like a lot in a 24 hours span with the amount of code changes I see in the repository.  Although the web site is just one part of it though there is a lot more to hosting than just the web site.  For the host themselves though the web site should be a major priority it’s where you gain customers not what is on the back end typically.  So we’ve had everyone running through the web site and if they find any issues post them to our project management system so that they can be resolved.

We’ve also been pushing out more and more changes with our system to track usage.  You test it all you want on a non production server but seems like when you bring in what is the final version and have people look for issues you tend to find a lot more.  So we ended up having 10 revisions on that front to resolve issues with tracking of usage, recording usage and usability problems discovered when we had more eyes checking out the system.

So things are looking great expect a live web site to come up soon and a launch announcement.

Ignore the icon that was my test icon it was a MySQL one.  The misc CPU on this was is N/A since I never actually tracked anything.  The days are also spread out since I was not running the system that collects CPU all the time.  So I only tracked two of the days and strictly their MySQL usage at the time.  The system provides monthly totals of Misc CPU which includes Perl, Ruby, PHP and anything else that ran as the user.  The MySQL CPU is the cpu time used by all mysql users controlled by that user.  It will include not only the primary username for example username but also the sub mysql users like say username_wordpress, username_forum etc. etc.  You will see the current months totals on the page as well as click-able links to see archived totals from other months.  So this is what our initial layout is going to look like on launch but I imagine we’re going to be changing it quite a bit as we go as we receive feedback.  You’ll notice the MySQL numbers are actually links right now so if you click on them you’ll be brought to a break down of the month or day.  If you click on the monthly total you’ll be brought to a day by day overview.  On that page you can then break down the usage by MySQL user which will look like this:

I only included one user in this case as I only had one with CPU usage.  It shows the CPU time used by each user during the month of January 2010.  So this can be useful in figuring out what is using up all that precious CPU.  We can also break down this on the previous pages by clicking a specific day:

So now you can get a daily break down.  If you were using a lot of MySQL CPU on a certain day you can see where it’s coming from.  Just like the monthly one the test user I took the screen shots with only had one mysql user with usage.  It will however break down all the users.

Well this is a quick sample of what our initial launch CPU usage module will display.  We’re not going to be setting quotas on the usage or anything of that nature.  It’s simply there to show users yes you’re using CPU and here’s where and how much.  So it will not come as a big surprise if we’re informing them about excessive CPU usage.  The initial version will not include comparisons against everyone else using the service.  Later on when we have more users we will most likely provide you with a comparison against the rest of the user base.  So you’d probably see the average for the server as well as the average for our entire user base during that same day, month etc.

So you visit the page http://www.cpanel.net/developer/overview.html which has a form to fill out for creating a plugin.  So in our case lets say we’re making a plugin called CPU Usage.  We’re going to make the plugin name cpuusage and then display name CPU Usage and the rest is pretty self explanatory.  Once you have your plugin then it’s time to get it to show up on the cPanel page.  You do this by doing:

cd /usr/local/cpanel/bin/
/usr/local/cpanel/bin/register_cpanelplugin cpuusage.cpanelplugin

So in our case since it’s the cpuusage plugin that’s what we’re doing but you can put whatever you’re calling your plugin.  Now in our case we’re using PHP which for the plugin portion isn’t as well documented as doing them in Perl.  So I’ll quickly run through this and how to make use of it.  The first step is putting the code into our folder the plugin is suppose to be in so lets say we’re doing it in the theme x3 we’d have our file at /usr/local/cpanel/base/frontend/x3/cpuusage/index.php .  Basically just the folder cpuusage then you can put whatever you want in it whether it’s a single file index.php or multiple files totally up to you.

PHP plugins in cPanel you can include cPanel code just like they do in the html files.  So in our case we have the default file set to index.php we could do this:

<cpanel setvar="headerimg=../images/topwebftpstats.gif">
<cpanel setvar="dprefix=../">
<cpanel Branding="include(stdheader.html)">
<div class="body-content">
<style type="text/css">
	<?cp Branding::spritelist(.spriteicon_img_mini {float:left;margin-right:5px;background: url\{leftparenthesis}%\{rightparenthesis};} %,spriteurl,images::#icon-${img}_mini {background-position\{colon}0 -${cssposition}px;width\{colon}${width}px;height\{colon}${height}px; }:) imgtype=icon,subtype=img,method=scale_60percent,format=png,img=cpuusage ?>
</style>
<div id="h1Title" class="h1Title"><div class="spriteicon_img_mini" id="icon-cpuusage_mini"></div><cpanel langprint="CPU Usage"></div>
<?php echo "This is my test page";
</div>
<cpanel Branding="include(stdfooter.html)">

You’ll notice the cPanel code in there to do the header and footer.  Then I just use PHP to echo just some random text.  So yep that’s right if you’d rather not work in perl you can do the plugin in PHP and lose nothing in terms of features that the html versions are doing when it comes to retrieving data via perl scripts in /usr/local/cpanel/Cpanel . In my case I don’t have the branding languages in at this point and I also stole the image topwebftpstats.gif you’ll notice. You can fill in the blanks by looking at the other cPanel pages as all their code will work within your php page as well. Hopefully this helps some people wishing to write plugins but to scared to write them in Perl.

LiteSpeed

I’m sure there are a few hosting providers offering the unlimited packages who run Litespeed but not exactly well run companies are even remotely reputable.  Most are out of business in a short time and are extremely small that the common user would not have heard of them anyways.  For the big providers though not a single one of them as far as I know use this on their web servers so this is indeed something unique to us.  Litespeed is gaining steam among users as they associate it with higher performance and being able to handle more traffic.  So it’s ideal as it increases performance for us as well as providing a marketing advantage.

R1Soft

Another feature pretty unique to us would be the fact we offer R1Soft as a free option to users.  Like Hawk Host it’ll be 7 days worth of backups at all time and be accessible to the users.  A lot of users associate unlimited hosting with poor backup policies which will not be the case with us.  We’ll give users the ability to see them at all time which is a big advantage to show the users yes we do backups.  It is also good for us as doing backups with R1Soft is much better than using say Rsync or even worth cPanel’s built in backup utilities.  R1Soft is also a demand from users in all hosting aspects as a feature they want.  So once again another marketing advantage for us having it available which is relatively unique to the hosting market Frog Host is targeting.

System Resource Usage

One big complaint from a lot of hosting customers is where is that imaginary line with regards to system resources.  They have no way of knowing where that line is or when they compare to other users on the same server or even the entire service overall.  We will be making it possible for users to see these sorts of statistics which should help with the problems associated with this for a lot of providers.  As I said in another post we’ll give the user the ability to see CPU (PHP, Perl, Ruby, Python Ect.) MySQL and inodes at the start.  So while not entirely unique it’s still a competitive advantage we have over the majority of our competitors.  You can view more information about this portion more in depth at the blog post Resource Management System

So that’s what makes Frog Host unique in the market place.  Though with any host you need to evaluate what makes you unique and a reason for customers to sign up with you.  If you’re not unique you’re going to be a treadmill host with no serious gains after a while or even worse you do not even get off the ground before you’ve decided to close your doors.  So remember the hosting market is saturated so make yourself unique otherwise it’s not going to go well.

Server R&D

This encompasses several things: software, server specifications, architecture, etc. A lot of this may seem trivial initially but as you add more servers to your fleet you’ll be glad you have setup a baseline server setup which will allow you to have a more “stable” environment as well as one you’re more familiar with (as opposed to running different server specifications on each server, different software, different versions, etc).

Policies

Now when I mean policies I’m actually referring to those internal variety. This includes things such as maintenance policies, deployment policies & guidelines, backup policies, and so forth. I have a sneaking suspicion a lot of new start up companies never fully outline these policies before going into business. I think this is a fatal flaw that will cause some to fall flat on their face (loss of backups because no backup policy, constant server crashes due to different kernels on different specifications, etc).

Deployment

This is partially lumped into policies and R&D but I decided to elaborate a bit more as it ties everything together. Typically this consists of setting a system up for production – but what does that really mean? An example deployment at Frog Host may look something like:

• Order server based off what is the “best” specification for our environment
• Setup the server to use CFEngine to prep the server (software, files, config files, etc)
• Setup & verify backups
• Setup server monitoring / health (Munin, etc) / watchdogs
• Setup user environment (cPanel, Postgresql, common libraries, etc)
• Test run the server (test everything that it would need to do in production)
• Put it into production

*NOTE*: CFEngine is a tool that allows you to easily manage numerous machines based off certain configurations / templates. This allows you to maintain a steady baseline of configuration files, applications, and similar things. Ultimately our CFEngine instance will copy over the configs / install some things such as Munin, Watchdogs, etc. Our current CFEngine instance is currently only doing lightweight management – configurations, basic application installs, and so forth. We intend to have CFEngine instance to eventually be able to almost fully put a server in production automatically.

Now this post only encompasses the “theory” behind prepping / deploying for a launch of a web host – in future posts I’ll go into more depth on the actual process and some quick tutorials on using the solutions we use (CFEngine, example policies, etc).