Archive | Technology

09 December 2012 ~ 2 Comments

Scandal! Drama! Intrigue! Results?

The Fact.

About a week or so ago Jeff Parker did what no one had expected – he changed the timer on his traffic exchange to 0 seconds, effectively having no timer.

Like with every tiny piece of potentially controversial action that happens in TE Land, the names took to social media and made sure the world knew what they thought about it.

There were two fairly clear sides of the debate, one saying that it will increase results or not effect them then the other side saying that it will destroy results everyone will cheat and it will destroy the ad. medium.

In the case of cheaters, there are people who cheat a traffic exchange regardless of what the timer is, so will people exploit the 0 timer? Absolutely, will people not exploit it? Absolutely.


The Opinion.


My opinion on the matter is essentially clear – People who experience good results with the type of advertising currently – still will.  If your page is great, pulls attention, is fast, and sells – it won’t matter if you stare at it for 5 minutes or 5 seconds or .5 seconds.  If your advertising sucks, your results will go down.  A long timer, at the very best, only forces your viewer to stare bored at your lame sales page for an extra period of time, most will switch tabs and do something else, with a little luck they’ll see a glimmer of something hopeful on your ad if they stare at it long enough and might click.

The fact of the matter is this, for people who claim to be “internet MARKETERS” what I saw, on all sides of the argument, was a total lack of actual understanding of marketing.  What I heard a lot of is “When I surf…”   “When I see a page…”  – and then what followed was horse manure.  Without going in to too much details, if you are one of these people – do some research.  The way advertising effects your brain is nothing to what your conscious reaction to it is, and this is a wide-spread problem, I see it everywhere – people mistaking feedback for results.  I don’t care if you like my ads, I care if they work.  You may hate me, hate my sites, hate my ads, but one thing I can guarantee to you is that every time you see the colors or the logo, you know what the ad is without even having to read it, if you instantly think to yourself “ah crap, this guy” and roll your eyes and move on without reading the ad – then my ad was clearly successful.  If you don’t see how that is, then call off the search party because we found the cause.

When I hear people say that the results of Traffic Exchanges are declining, and I look at my business and my results and the results of my affiliates, I always kind of chuckle.  The world around us is evolving, very rapidly, and if you are a person who doesn’t understand the way advertising works, you’re right – your results are falling. You’re advertising based on what you see other people do, on what you are told works, on what is supposed to work — not on what does.  You are mistaking feedback for results.

I know all this might sound a bit malevolent, but seriously.. look at it objectively – you say that results decline, but then you stick to the same old thing for year after year after year… if something isn’t working right, then change it!  This should sound like common sense, but someone changes their surf timer on THEIR OWN TE and it’s like the world is ending – without even spending a second to test it, to see if it works.  If you are advertising the same way, the same style now as you were 5 years ago, 2 years ago, 6 months ago – then yes, you’re right – you’re results are declining.

So why No Timer?  What made Jeff come to the theory to use no timer?  I have no idea, what his motives were.  To me it seems obvious that the more organic the traffic is, the better your results from it will be.  If you force me to build a fence and I can’t stand building fences – then I probably won’t do a very good job building it, or put much care into building it.  If I love to build fences – I’m going to seek out fences to build, and I will do a great job at it.  The same goes for advertising, the best conversion rates I get are traffic from Search Queries obviously, because people are looking for my product.  So the idea of the zero second TE is that people will see your site if they want to, and if they don’t, they won’t.

The execution is the only place I question it – from where I stand, changing the timer isn’t going to effect the quality of traffic.  People are still being forced to do it.  Despite now being able to choose how long they spend building the fence (weeding out the real fence builders from those who hate it), the people who don’t like building are still required to do it.  What’s forcing them to build the fence?  Not the time they spend doing it, but the money they get for it.  In my opinion – if you want to remove a timer then you need to remove the incentivization also, credits, cash, etc.. the motivation for the visit.

If you’re just scrolling through the paragraphs – the TLDR version:  My Opinion – A timer won’t decrease results, but in my opinion (OPINION, NOT PROVEN TEST) a better result would be to remove the credits instead of the timer.


The Results.


So what happened with the zero-timer anyway?   I took a single splash page that has had overall about a quarter of a million hits.  I made a brand new tracker just for this TE so I wouldn’t confuse it with any pre-timer results.

For the Disclaimer, I only purchased 6,000 credits at Traffic Fugitive for the test, which to me is much to small of an overall sample size but does give a decent idea of direction.  Of course the  next 10 hits could have all generated signups, or maybe the next 15,000 wouldn’t generate any at all.  That  being said, the results… The same.

The same splash page on the 0 second timer generated just barely slightly less (less than a tenth of a percent) then the average in both clicks AND conversions.

SURPRISE!  It didn’t really effect much at all.


The Final Word.


So now that the fire is out, the lackluster increase or decrease in results should sort of muffle most of the obnoxious outliers, I have a couple of other things for consideration of the zero-timer TE.

First of all – many TE owners talk about TE’s as having a bad reputation for advertising, the opponents say that much of the traffic is bad or non-targeted or automated.
When you approach someone who may have these oppinions of TE’s, how will you convince them that it is good traffic, while telling them that their ads have 0 second minimum view?

When people surf, they may tend to stare  at the timer to watch it countdown, increasing exposure to the things around it – including text and banner ads.
What do you think the results will be to Text and Banner ads in  a 0 timer environment?  Has anyone tested the results?


Continue Reading

15 September 2012 ~ 2 Comments

Liquid Web Hosting Reviewed – September 2012

If you’ve stumbled across this blog post from Google or another source and are not familiar with me, I’ll share some background with you so you can get an idea of my situation and hosting needs.  I am John Bell, and I operate a company called Insidmal Design LLC, I was  hosting 11 domains on Liquid Web dedicated servers, about 5 of which receive fairly heavy and steady traffic, and are heavily database driven sites.  I previously was hosting with GoDaddy but their lack of support and configuration options lead me to outgrow their hosting options, after much research I went with Liquid Web Hosting.

For the first while Liquid Web was an excellent company, I had superior support, fast service, everything was excellent, I reccomended them as a hosting company over and over again, as they were absolutely perfect for my needs, and their support was very helpful at discovering and resolving issues that were beyond my level of knowledge, but then everything went down hill.

Eventually I reached a place where one server was not enough to handle the capacity of my websites, and it was time to add a second server into the picture.  The idea was one server would handle the apache requests and another would handle the mysql requests.  The sales staff was very helpful at getting the perfect configuration for my needs, and that is when the positive experience ends.

I purchased the server and received an email from the tech support person who was to deploy it, after about 24 hours I received an email saying there was a problem with one of the packages and that the installation was corrupt and he would need more time to set it up, another day passed and I was told again that there were issues, again, and that it would have to be essentially started over from scratch.  It was setup and I was informed I would  need to change all my database connections to point to the other server, so I updated it will all the information provided, and then later was informed that because of the system technology I actually didn’t need to change it at all, as connections would be forwarded.  After this I proceeded to change them all back to how they were before on my 11 domains.  Later I received yet another email saying that due to some issue I actually would need to use different mysql login information for the domains, this being different then the initial data, so I yet again changed all the code in the scripts.  This seemed to do the trick on testing, so we decided to go live.  The server was live and not even 15 minutes after it was all set up, it crashed.  I waited 17 hours for a reply to my support ticket about the server being down, this means my sites were all inaccessible for an entire day, under their “Heroic Support” Sonar Monitoring blah blah blah bullshit.

I wasn’t aware as much about how servers worked as I am now (because of a lack of support) but when I finally called to complain, they finally decided to reply to my 20 hour old support ticket (30 minute response guaranteed?) .  They quickly discovered that apache had been grossly misconfigured so all the requests were going into Swap memory which was causing very slow response time and essentially the server overloading itself and crashing.  This issue was finally resolved, however then it was discovered that the initial tech who had to set up my server three times before he got it right, did not install the backup hard drive!  So I had to face even further downtime while they installed the hard disk.  Finally about 10 days later, things are running as they should.


We’ll fast forward about 3 months here – there is an issue on that same server and it crashes, I submit a support ticket because their 24×7 monitoring team somehow hasn’t noticed that my server has been down for half an hour, and I am informed there is an issue with Apache and the tech says he has to remove it and then rebuild it.  I said ok, do whatever you have to do to get my servers back online.  He does this, and from that moment on I was dominated with continuing server issues.  Before this point I was running with Keep Alive Off as the sites are very high traffic and keep alive on was filling all the server connections.  After this point, with Keep Alive off every page load resulted in tons of broken images, missing data, a complete failure in rendering the pages correctly.  I let them know, and they tell me that I just need to Turn Keep Alive on.  I inform them that this was a non-issue before the action, and that message goes literally completely ignored.  I turn Keep Alive on, and things run ok.  The next day the server crashes again, and again I have to submit a support ticket because their 24×7 Monitoring Team again goes completely unaware that my servers are completely inaccessible.  I am told this time that I do not have enough RAM to handle the connections I need to allow to not fill them with Keep Alive on.  I tell them, again, that before I had no problem with Keep Alive and now I can’t run with it off, that comment again goes ignored and I am suggested to upgrade the RAM.

So I upgrade the RAM, and everything seems ok, again until the following day.  Each day DAILY the server would crash at least once, it went at most two days without crashing.  I also had discovered that the firewall was blocking IP’s left and right for ‘Port Scanning’ even though they were IP’s of people who regularly used the sites as they were intened to be used, again a problem that did not exist before they “fixed it” the first time.  Each time I have to submit a support ticket because their monitoring team didn’t notice.  I spent an entire month of nights with little to no sleep because I was having to monitor my servers 24×7 because their monitoring team did not even once detect downtime.  I also am at this point receiving about 10-15 support tickets daily from various members complaining of latency on the server.

I had a support ticket open with LiquidWeb for 40 days for this issue, the cause of all these issues was NEVER determined, every time I would get a support tech, tell them what is going on, tell them how since this change they would reccomend the same simple solutions as far as configuring apache to reduce load, despite the load NEVER going up, it was all the same generic basic problem responses.  One support staff shift would end and the ticket would get passed on to the next, and the process would start again.  This went on for well over a month of time, still unable to determine the cause of the problems, I literally had no choice but to leave their company.

The “24×7 Sonar Monitoring” team not even once was able to prevent or even made aware of over 60 periods of server downtime.  Their Heroic Support would take as long as 6, 10, even 20 hours to reply.  Their Heroic Support was unable to resolve issues on the server at all, when it came to basic simple stuff they were very kind and helpful, however more complex issues like I had, it was a complete failure and breakdown of their support system.  I mentioned serveral times about how when a previous technician performed whatever they did to Apache, is when every one of the laundry list of issues began, and every single time I mentioned it, it was ignored and they would say the problem was something else, depending on the tech it would be a different cause, one tech went as far to tell me that it was from DDoS connections because an IP had 15 simultaneous connections to my server.  15 connections is not a DDoS attack by any stretch of the imagination.

It is simply unfathomable for me to see customer service run this way.  From the start when the intial tech had to set up the server 4 times before he could get it right and even then he failed to install a hard drive, to over a month of continued stability with absolutely no resolve and long response time waits.  I am unable to treat my customers with such nonchalance, as my support requests skyrocketed, I had the most difficult and stressful experience of my life trying to provide support to my members when I had no clue what the problem was, and even the company I was paying over $500 a month to do for me, was failing at it also.  I could provide no answers to my members about the cause of the problem or when it will be fixed, because of the failure of Liquid Web Support.

The very beginning of my experience with Liquid Web was one of the  best hosting experiences I had in my life, and it quickly turned in to the worst.  I really enjoyed the company and wanted to stay, but their complete lack of support and inability to monitor the servers with enough haste to justify the cost.  I was left literally no option but to leave the company because their lack of apparent interest in my purchases, was literally destroying my business.

Can LiquidWeb be a great hosting company? Absolutely.  Is it? Sure, so long as you don’t need to rely on their monitoring and support.

LiquidWeb Support was there when I didn’t need them, but when I needed them the most, they left me out to dry.  I know they get great reviews, and have great pricing compared to their managed hosting competitors, but please keep this in mind before making a purchase and supporting a company that is not willing to invest enough to properly train their employees beyond the most basic server operations.



Related Posts Plugin for WordPress, Blogger...

Continue Reading