So my bonus is along those lines and will reveal a lot more of my long term plans, which I even hinted at long ago in a comment on Jeff Walker’s blog when he first launched PLF over 3 years ago.

Best Existing Solution By Far

The other day I felt compelled to write about Quansite because it is the closest service to what I want to achieve.

If you are looking for a perfect “no tech” solution for small product launches, I honestly don’t think there is anything that can beat it.

For bigger launches you might start running into resource problems – you can upgrade to a bigger package (they offer semi-dedicated) but that is a long-term commitment just to cover peak traffic.

Off-loading media files to a CDN is only a partial solution.

Running Your Own Dedicated Servers

I have heard that on some of the biggest launches they run up to 10 dedicated servers in a cluster, though Stompernet squeak through on just 4 beasts. From a reliable service provider you are still looking at possibly $2000+/month for the hardware, plus you have to have the server jockeys to run them.

Then you need some kind of media hosting, as even 4 beefy servers won’t be able to stream videos to 100s of visitors, let alone 1000s.

Most of the “Gurus” currently use Amazon S3 or Cloudfront

United States Edge Locations
Data Transfer

$0.170 per GB – first 10 TB / month data transfer out
$0.120 per GB – next 40 TB / month data transfer out
$0.100 per GB – next 100 TB / month data transfer out
$0.090 per GB – next 100 TB / month data transfer out
$0.080 per GB – next 250 TB / month data transfer out
$0.070 per GB – next 250 TB / month data transfer out
$0.060 per GB – next 250 TB / month data transfer out
$0.050 per GB – data transfer out / month over 1,000 TB
Requests

$0.010 per 10,000 GET requests

I am not going to knock Amazon, as before they came around a few years ago, CDN bandwidth might have set you back 5x as much, but then nowhere near as many people were publishing videos… it was costly.

Even so lets look at a typical (small) product launch

100 MB video per person
20,000 visitors
==================
2000 GB of data

Not a lot … around $400

But some larger product launches have

500 MB video per person (I am not talking browser killing Stompernet sales pages)
100,000 visitors
==================
50,000 GB of data

A little bit more … around $6500

Just imagine how much one of the really big launches costs, and then the cost of data delivery for the actual content to members who quite rightly want to consume what they purchased.

Or you could look at a sales page

10 MB Video – Very short low quality
500 visitors per day
5GB per day
150 GB/month
============
1800 GB per year

You are still looking at close to $400 for bandwidth /year

If you do insist on using Amazon S3, do at least use something like Media Stream Guard (not aff) from my friend Craig. So many gurus leave their Amazon content insecure.

Alternative CDNs

SimpleCDN – I have been monitoring them for the last year, and they have changed their offer maybe 5 times, effectively wiping out whatever investment clients have already made in setup time. If you were a very heavy user, maybe you would have still had significant cost savings, but the feature sets have been changing drastically.

Rackspace with their Cloudfiles at $0.22 are actually a contender as the playback is probably better than Amazon.

NetDNA – recently quite popular among many bloggers using WordPress and bandwidth pricing can reach very respectable levels offering a price advantage over Amazon – the negative is that most of their “Edges” are in the USA with just one in Europe.

Edgecast – competitive in volume, fast & reasonably full featured. Notable connectivity in Asia

Highwinds – various resellers strong in both US & Europe – Asia promised.

VPS.net – A reseller of Highwinds and also cloud based VPS – were previously reselling Edgecast but there were problems with flexibility.

With most of these CDNs it is possible to get bandwith cost down to below $0.06/GB – possibly a little cheaper with some negotiation and very high volume.

You would have to be running the equivalent of 20 large product launches with Amazon Cloudfront to achieve these pricing levels, and before you reach that level there is some margin for eith cost saving or profit.

Flash Players

You can always take the free option of using a GPL player such as Flowplayer, though they also have a premium option

JW Player has some viral and data benefits that would require a liitle custom programming to get working on other players, but at the same time are not in themselves perfect. I have seen slightly better implementation, but none that I would consider ideal. The licensing for commercial use isn’t too expensive.

Transparent Player – I haven’t checked out version 3 which was recently launched yet (and it is not even mentioned on the sales page last time I looked), but with the unlimited use plan it is a very viable contender

EZS3 provide players in addition to making it easier to work with Amazon obscuring data, usable on unlimited sites. I have seen this used by Frank Kern.

This isn’t meant to be a conclusive list, I know a ton of other options, both with and without available source code for custom modifications. Boy do I want to do some modifications.

Best Current Solution

I am laying my bets on VPS.net for some serious clustered high-availability hosting and also providing primary CDN, with some alternative backups.

With players the field is open

Why No Video Streaming With Own Players?

I can imagine people running huge launches who are extremely well monetized can afford some of the more expensive solutions, because they are then in a position to negotiate, but that isn’t suitable for most.
For some reason even people supposedly running 8 figure businesses have problems keeping servers running smoothly.

Jeff Walker is using Ooyala – I haven’t quite worked out the pricing, but it will be more than just a CDN.
Brightcove despite the free trial is not cheap
Viddler – if you don’t have their ads then commercial use even as an Enterprise client will cost you $1500 for 4TB of data or 0.375/GB = 5x as much as VSP.net in much smaller volume.

There are many other options with the price range typically $0.30 to $1 per GB.

What I Really Want To Do?

I want to provide hosting for product launches for FREE, including all the bandwidth, high-capacity clustered servers, built in membership software, and many other advanced features.

Is there such an expression “Fat Pipe Sales Funnel” ?

That is only part of the equation, the service has to make money and cover quite considerable costs, but that is where “value add” comes in with all kinds of network effects that add up to more than enough to cover costs and then some.
After all WordPress.com can offer something similar and they are really under-monetized.

Leverage :)

Pipe Dreams?

One focused goal for 4 years will eventually happen but in hindsight I probably should have just concentrated on clogging up the interwebs with small niche sites, or building a huge list, but something held me back. I would almost certainly have the resources to build this project from cashflow now.
Even the paid link battles from 2007 were very much a part of the “sounding out” process on possibilities.

I have avoided masterminds – they would have persuaded me to pursue more attainable goals.

Product Launch Manager Bonus “Can Be Delivered Now”

CDN = “Content Delivery Network: or alternatively “Can be Delivered Now”

There are tons of huge bonuses being offered by other affiliates for Product Launch Manager, because it is a great product but also the affiliate payout is… generous to say the least.

The wonderful thing about the course is that it will massively boost the number of people running product launches in all walks of industry, but at the same time boost demand for higher performance hosting with significant commercial advantages, so it finally makes my long-term goals infinitely more attractive as an investment opportunity.

Just providing one core component might be enough “proof of concept” to tip the investment balance, and just the benefit of having one major “breaking point” of product launches crossed off, the technical side of preventing launches crashing to a standstill because of server meltdown is probably the most targeted bonus you will hear about today… well other than pure consulting offers from people who have run launches.

My Offer

I will provide a preconfigured CDN platform where you can just ftp files or they can be pulled using Rsync, along with origin pull to the CDN along with $1000 of bandwidth at cost – the bandwidth will be available as and when you need it and will not expire if not used on a month-month basis like many subscriptions.
Most likely it will be through VPS.net, but I will shop around and provide additional fallback

I will also provide additional remote consulting for both the CDN and other launch issues to the best of my ability. I am a “jack of all trades” and can talk both geek and marketing. I am not going to place silly limits on time – your success is my success.
As and when I scale, I will be able to provide additional human resources to fill my shoes.

When other components become available you will have the option to use them – I have my server techs sorted out now to manage the clusters and VPS.net is ideal to scale this kind of platform, in many ways more than Amazon EC2 and other cloud options because of the nature of product launches.

Until such time as I have the backend monetization and “leverage” to offer all the hosting for free, additional bandwidth and any server use will be at cost – not inflated cost… real cost. For the CDN it will be something like $0.06/GB at current exchange rates and pricing.

If you want to set up your own VPS servers or clusters I can also offer nodes at competitive rates as I will end up with 100s of nodes, and point you in the direction of people who could manage them for you.

I hope you will consider my offer – the PLM offer isn’t cheap, but will be attractive to existing PLF owners, and maybe a few others.

Alternatively go with the big $$$ packages. You know the game – all those leads you send to the big launches that didn’t convert (but which you presold) will now be offered the exact same package (or very similar) as a bonus, with no commission for the original affiliate despite doing a significant chunk of the selling.

Making crashed servers on product launches a thing of the past

I have a cure for respecting the value provided by smaller affiliates in the sales process as well…

Warning: This article contains lots of really useful information that will probably help people but also contains a few affiliate links – not every link is an affiliate link, and not every link should be considered an endorsement

p.s. I know my server crashes occasionally – I personally suck as configuring linux servers, which is why last week I hired a great team to do this stuff for me.

When the sales letter goes live, <<< USE THIS LINK >>> – hopefully the previous links will still lead to some of the more descriptive videos.

This guide or checklist is for product owners to think about. Very few of the points raised are a deal-breaker.

Deal Breakers

• Junk products
• Don’t fit my primary target audience
• Poor treatment of customers e.g.
  • Support tickets unanswered
  • Major product issues not addressed in a timely manner
  • Heavily moderated forum – only “happy bunny” threads
• Poor treatment of affiliates e.g.
  • Lack of or late payment
  • Sneaky changes in payment options bypassing affiliates
  • Change of affiliate system without notification
  • Radical change in deal structure
• Generally dishonest sales pages
• Skimming Sales

Major Issues

• Tracking problems
• Disappearing accounting data with no explanation – if I made a sale, and then it is refunded, I need a record of it all. I need the numbers to add up
• I really need SubID tracking to be able to use PPC
• I love the ability to deep link to content
• Upsell Hell
• Viral Hell – (see previous post)
• Major Sales Page leaks
• Excessive promotion of 3rd party affiliate products in launch materials – if your free ebook is crammed full of affiliate links that I don’t receive a benefit from, it is effectively the same as if I was sending my traffic to a PPC review landing page using someone else’s affiliate links.
• Promoting 3rd party offers on a mailing list built with my traffic rather than focusing on the intended launch product during the launch period.
• Excessive promotion of 3rd party offers in general
• Teaching some methods I don’t agree with
• Promoting products I wouldn’t promote due to major issues/deal-breakers
• Integration marketing – sounds nice in theory, in practice if your added value is a load of free offers from your cronies, that I don’t get a piece of the action on, there is a problem.
• No way to export SubID data
• No way to export, or be pushed sales data
• Elements of Proof – proof used on the sales page that is possibly factually accurate, but misrepresented
  • AWStats for traffic measurement?
  • Your Clickbank / Paypal sales didn’t mention PPC costs, affiliate payouts, established list just from previous product launches in the niche etc.
  • The testimonials are from direct business partners
  • When you tested your sales page for conversion, was it with a very warm house list of buyers? Thought so…

Significant Issues

• Payment period – if I am paying for traffic, I want to be paid by you before I have to pay my traffic costs.
• Checks (Cheques) – especially if low value – I can still sorely remember receiving my first cheque from Linkshare, for something like $2.73, that my bank wanted me to pay a $20 fee to cash, and I would receive the money in 6 weeks. I have a better bank now, but offer alternatives.
• Paypal – whilst I hate the transaction charges, especially currency, in many ways it is easier for accounting
• No IBAN support – Having to qualify by receiving a number of checks is just adding friction. If an affiliate knows about the benefit of receiving funds directly, they are going to make this worth while. (I am not sure if there are international banking regulations forcing a few checks to establish identity)
• Private tracking systems – I appreciate the advantages of using Infusionsoft for a publisher with a refined sales funnel, but why do I have to have a separate login and accounting panel for every single program?
• Customer/2nd Tier/Opt-in For Life That Changes – often caused by changes in affiliate tracking system. Suddenly the residual you were expecting dries up, and you had been ramping up on PPC
• Pushed Data or API – I want to be able to offer incentives automatically with all affiliate offers that allow it (where there is a transaction), and I don’t want customers to have to wait for it. I know there are solutions by offering the bonus directly on the thank you page, but in some ways that is a little clumsy.

The following might be looked on by many as “par for the course” of big launches

Stealing Sales

Imagine I sent you 100 opt-ins for your product, but you didn’t get any sales. A few weeks later you offer exactly the same product as a bonus if the prospective customer buys an affiliate product.

• It could be looked on that I didn’t do a good enough job in preselling or boosting my offer with bonuses, thus it is your right to enhance your offer of another product with the original product as a bonus and take all the cash.
• Alternatively I sent you a great prospect who is now a buyer, yet made nothing from my efforts – the lack of initial sales could have been your own value proposition, sales letter, etc, things out of my control.

This is something that is really hard to prove, because there is so much split-testing and list segmentation going on – I might have seen you make the offer only because I am on your buyers list, and that the prospects were sent something else

Solutions?

• Have products that you only ever offer as upsells or bonuses
• Extremely complex cooperative tracking

Many respectable marketers do the first option, and have tons of loyal affiliates possibly because of it.
The second option I might eventually have a solution for, it has been a constant gnawing problem in my head for 4 years.

Servers and Launch Material Delivery

This I can break down into 2 phases

• Primary server stability – I don’t care how much unexpected traffic a launch caused, you should have been able to handle it if you really have the high quality support team in the background.
• Video delivery – You have no idea how bad Brightcove video delivery is from Europe, and also Ustream recorded sessions. They don’t support the ability to buffer a whole video. Amazon is a little better, certainly from the UK, but in Poland half the time I have to open a browser and go and make a coffee.

Note: Amazon also isn’t an ideal solution for premium membership content – I haven’t seen an Amazon implementation yet that didn’t reveal the download location in the Http headers, though it is possible to secure more effectively, probably requiring an EC2 instance to feed unique expiring links.

Reciprocity, Karma & Great Stuff

There are tons of counter arguments – amazing free value can win against almost any issue. There are a few more I could add, but let me know your “pet peeves” in the comments.

Watch out for another post tomorrow, I won’t be pulling any punches

The culprit is TwitThis, that seems to be using Amazon S3 to store images.

Here is the standard TwitThis code:-

<!-- Begin TwitThis (http://twitthis.com/) -->
<script type="text/javascript" src="http://s3.chuug.com/chuug.twitthis.scripts/twitthis.js"></script>
<script type="text/javascript">
<!--
document.write('<a href="javascript:;" onclick="TwitThis.pop();"><img src="http://andybeard.eu/wp-content/uploads/twitthis_grey_72x221.gif" alt="TwitThis" style="border:none;" /></a>');
//-->
</script>
<!-- /End -->

What you need to do is host the TwitThis logo on your own server to speed things up.
Here is the image used that you can save to your harddisk and then upload to WordPress or other blogging platform.

Then all you need to do is insert it into the code

In my case this is what I am using

<!-- Begin TwitThis (http://twitthis.com/) -->
<script type="text/javascript" src="http://s3.chuug.com/chuug.twitthis.scripts/twitthis.js"></script>
<script type="text/javascript">
<!--
document.write('<a href="javascript:;" onclick="TwitThis.pop();"><img src="http://ndybeard.eu/wp-content/uploads/twitthis_grey_72x22.gif" alt="TwitThis" style="border:none;" /></a>');
//-->
</script>
<!-- /End -->

Actually I have listed a broken URL above, I don’t want people just copying the code and trying to use the image from my server.

As an alternative to the TwitThis button, Jemmille has created a modified version of my hacked Sociable plugin Antisocial that includes a “TwitThis” button, and a fix to StumbleUpon.
I will probably update my distribution with Jemmille’s code in a few days, but you can get it directly for now.

Update: This wasn’t a total fix because the Javascript was still being loaded from Amazon S3 and was causing horrendous page load times. TwitThis has been removed until I can hack a version that doesn’t use javascript.