Thanks to all the users who have flagged up problems with WordBay (and other eBay RSS-based software) caused by changes to the eBay RSS feed. I am sorry I did not get to grips with this sooner - it seemed to me initially that the problems were not affecting WordBay, but on further investigation there are some strange behaviours. Most notably, links to eBay items do not take you to the item itself but to an eBay login screen, which obviously significantly reduces the chances of sales (though strangely, my sales don’t seem affected that much). This appears to be due to some kind of change in the structure of the so-called “rover” links produced by the RSS feed, in turn maybe caused by a change to the RSS call itself.

I need to look into this ASAP and get it fixed - I hope it is not a complicated issue that would mean major changes, and that it will be a matter of a day or two. So please hold tight and sorry once again that this was not fixed sooner.

Problem with countries.txt file?

One user has also complained of a problem with WordBay not finding the countries.txt file (giving a 403 error) and preventing the countries being displayed in the drop-down lists in the admin panel (the countries.txt file simply contains a delimited list of all the country names). This is an issue entirely unrelated to that of the RSS feed, I am pretty sure of that, but if anyone else has had this problem please let me know. It sounds to me like the file might either be missing or have the wrong file permissions (no read permissions), but it could be some other issue relating to server configuration which I cannot really fix on an individual basis. This might only relate to v1.16 which I put out a few days ago specifically to deal with how WordBay figures out the path to the countries.txt file. The plugin now gets the path “properly”, i.e. the valid Wordpress way, so there should not be a problem as Wordpress uses these same functions internally too.As a “first resort” please do check that the countries.txt is actually present in your wp-content/plugins/wordbay directory. And make sure the plugin is installed to that directory too!

Back with a new version very soon I hope!

Posted on February 18th, 2010 by markowe
Filed under: Uncategorized | 5 Comments »

Your response to this will no doubt depend largely on how your earnings have been affected by EPN’s new QCP system, which officially came on line at the beginning of October. I assume you are up-to-date with these changes - if not, then wander over to the official EPN forum to see what all the fuss is about.

I have kept quiet on the subject so far, but am now interested to hear how WordBay users are faring under the new system. Since our model is very similar I wonder whether there are any similarities in our performance (that’s assuming almost everyone has not suffered under QCP, as the cynical amongst us might be inclined to think) or whether the niche has more influence over our performance under QCP than the business model.

My experience so far

Here is a summary of what I have seen with QPC since the preview earnings appeared on August 18th. I am just showing figures for US eBay as my earnings with other programs are far less. Bear in mind I am paid in pounds sterling, so multiply my EPC by about 1.6 to get USD (at time of writing and rock-bottom pound!)

August seemed positive as it turned out my real EPC was pretty appalling - 0.01 GBP - and QCP was set to almost double my earnings, at least for that part of August, as it was showing an overall QCP EPC of 0.02, though I now realise this was pretty dire still. I have never been a big EPC-watcher (hey, who cares, as long as I am earning a few shekels every month, right?) but I now see that my EPC has been averaging about 0.03 GBP since EPN began.

Weird things noticed: strange patterns in QCP EPC over August, with almost all campaigns showing the same EPC for any given day, but varying over the month, so that EPC was a high as 0.04 some days, but ended August on 0.01 again for all campaigns. Now how can that be possible? Surely QCP is calculated on a per campaign basis! So how can all my campaigns show the same EPC on a given day, including my “generosity” campaign, which lumps together earnings from all “generous” users of WordBay who share revenue with me and operate an extremely diverse range of niche sites which cannot all be performing the same!

One theory was that this is the “pooling” effect, whereby small publishers are lumped together and given a default EPC which could display this behaviour, but I soon found out that a small publisher is defined as having an average of 20 clicks or less per day and I certainly have a BIT more than that (though not exponentially more)!

Still, I could handle my earnings doubling under QCP, was my overall assessment!

September

Started off with a terrible EPC of 0.01, but gradually rose until I had a couple of days with 0.08 and 0.09, i.e. about $0.14 USD - FAR more respectable - but then things slid back to 0.02 by the last day of the month, giving me an average of 0.04, as compared to a “real” EPC of 0.05 (strange improvement there in my actual EPC). So I was only going to be down 15% or so, I could still live with that. However, the strange pattern of consistent daily EPC across campaigns continued, and no explanation has been forthcoming on the EPN forums, despite a number of other publishers reporting the same effect.

October

QCP is live! What is going to happen, we wonder. It would be nice to see mostly those 0.08 days happening but… what do you know? EPC has kicked off at 0.02 and just peaked at 0.03 the first few days of October, dropped back to 0.01 and I have a sinking feeling… Whatever the explanation for those big paying days, it seems that a maximum of 0.03 EPC is going to be the order of the day for me under QCP. Downloading the transaction reports (make sure you download the All Events one), I see that this is considerably less than I would have earned, with a number of good-paying Winning Bids starting the month off. The most I can hope is that these will translate to an improved EPC later in the month…

So what now?

Well, the jury is still out on QCP for me. I can wait a few more months to see where this is really going but I am not entirely optimistic. Yes, like most EPN publishers I think the system is unfair, untransparent and ultimately less rewarding for many (most?) publishers. EPN WAS a very generous affiliate program, I must admit - easily my best earner and one of the highest-paying out there. I imagine eBay were getting robbed blind by black-hatters gaming the ACRU system and probably there has been a very common-sense business decision behind the introduction of the new system - simply put, they were probably paying out too much to affiliates and not seeing the best return, and this is their way of taking control of this, and in a way, one can’t blame them for this.

However, what will grate with many of us is the lack of transparency and the disconnect between the sales we are bringing in and what EPN is actually paying us. Ultimately many of us will NOT be encouraged to invest further in trying to figure out what on earth is being asked of us with QCP, and we will vote with our feet - frankly I can make a lot more than 0.03 EPC from good old Adsense, without standing on my head trying to figure out QCP…

What now for WordBay?

It DOES seem that the business model of sites using WordBay (which is pretty similar whatever the niche, let’s face it) does not bring in the mysterious “quality” that eBay is looking for, though maybe you will set me straight. I am surprised at this. The visitor Googles for “buy antique Persian love-god figurines”, comes to your site, clicks the exact item they are interested in, bids for this item on eBay soon after clicking through… Surely this is “quality”, as opposed to sending random potential customers through banner links? From my results it seems not…

So I have to apologise to all those who are waiting for me to add their favourite feature to WB, or to correct their favourite bug (I know, I know, there are a few emails and comments I have not responded to) - until I see what is going to happen with my EPN earnings in the next few months I am really not motivated to invest time in WordBay. At the moment Wordbay seems to work fine, with no major bugs, and no changes I introduce now are likely to affect QCP EPC greatly (well, what does affect it, we wonder…), so I hope you can live with that for now.

On the plus side, I am looking into what I could adapt Wordbay for - something similar could be written to work with Commission Junction, Amazon, or a host of other affiliate programs which offer web services. But it would take a bit of work, and I would have to find a way to monetise it, otherwise I cannot justify the work put in - especially not with a baby due very soon and the budget about to get a whole lot tighter! My wild coding days are over, I fear!

I would love to hear your experiences so far with QCP, your moans and gripes, your amazing success stories, and your ideas of how we as small-time niche-site owners can survive this and perhaps take the Wordbay idea somewhere else, where it will earn us enough to keep us in Persian love-god figurines for many years to come!

Posted on October 5th, 2009 by markowe
Filed under: Uncategorized | 15 Comments »

OK, sorry to all of you (and thanks for letting me know) who got version 1.1 of WordBay only to discover that it did not work as advertised, i.e. did not preserve your lovingly-crafted WordBay.css file. This happened due to an oversight on my part: when using the Wordpress automatic plugin update in the backend, Wordpress deletes the contents of the plugin directory before installing the new one, thus leaving no CSS file to preserve! The CSS file would only be preserved if you installed manually, but obviously a great number of users will be relying on the automatic plugin update feature so I have completely rewritten that feature now so as to store the file in the WordPress database, which means that your css file is now edited in the backend (which is starting to look very cluttered). One proviso straight away:

The first time you install this new version, you still need to back up your WordBay.css file first! Keep it safe somewhere! Unfortunately, I cannot stop Wordpress deleting it if you update automatically.

Also, I realised from a user bug report that this version of the plugin will unfortunately not work with PHP4 - sorry guys, it’s really time to upgrade to PHP5 if at all possible (may be as simple as choosing PHP5 in your hosting control panel, though make sure all your sites work after!). It’s not really practical to keep providing back-compatibility with earlier versions of PHP - I am suprised that something like this has not happened sooner. If you get a failure like “Call to undefined function: file_put_contents() in…” that’s what that is. If you can’t upgrade to PHP5 then I suggest rolling back to WordBay v1.05 and sticking with it until you can upgrade PHP…

Instructions

1) You HAVE backed up your existing WordBay.css file, right?!

2) Update the plugin either automatically or manually. If you do it manually, you can skip the next step, since your old WordBay.css will still be in the plugin directory.

3) Copy your backed-up WordBay.css file into the wordbay plugin directory if necessary.

4) NOW, when you go into the WordBay settings in your admin backend, you will see, near the bottom, an area where you can edit your CSS file. From now on, do not edit the WordBay.css file directly - use the back-end, otherwise your changes will just be overwritten the next time you change any WordBay settings.

The plugin checks all the scenarios I could think of:

- there is no existing WordBay.css file (for some reason) and you were using a version prior to v1.1 thus none was saved to memory: a new WordBay.css file will be created using the Wordbay.css.default file

- there is an existing WordBay.css file (that you either copied in from a backup or was left after a manual update): this file will be used and again, you should edit it in the back-end from now on

- (and this is the bit we all want to work properly now!) there is no existing WordBay.css file because it got overwritten by the automatic update, but you had been using v1.15 (this new version) or above, and so the WordBay.css file was stored in the Wordpress database and is lovingly restored as soon as you go into the backend. Yes, for now you WILL have to pay the backend a visit every time you update - all these checks are done by the backend not by the plugin frontend - I didn’t really want the plugin bogged down with all these checks every time the plugin cycles, though I’m sure there would be a simple way to do this.

OK, this was inevitably more complicated than I expected so please let me know how things work - I have tested it quite extensively, but I know you guys can always find some scenario I didn’t plan for! And of course, this feature only really comes into its own once I put out a NEW version of WordBay, but you can try it out by maybe uninstalling WordBay and reinstalling it if you want.

Let me know how you get on.

Posted on August 25th, 2009 by markowe
Filed under: Uncategorized | 11 Comments »

I have recently been taking one of my all-too-regular breaks from developing WordBay, and indeed from any major affiliate marketing efforts. I guess I am a dabbler at heart, plus a April plunge in EPN earnings really demotivated me too. Seems to have picked up again now, so perhaps it was an initial reaction to the economic crisis.

One of the things I have devoted a little time to is testing a new money-making tool called WAHA CMS. Just to disclose straight away - I agreed to accept a beta version for testing in return for writing a review. Otherwise, I am not affiliated in any way, do not earn anything from sales of WAHA and am not otherwise being paid for a review. Also, I do not see WAHA as competition for WordBay - WAHA is a commercial product, WordBay is not and I do not mind competition!

What is WAHA?

WAHA is described by the authors as a “…a content management system (CMS) for quick generation of quality web sites but it is not a simple CMS…” Someone else I talked to perhaps a little unkindly referred to it as “splog” software.

If I had to describe WAHA, I would say that the term “CMS” is perhaps a bit ambitious. Really, it is a richly-featured mashup system - the ultimate automated website generation system that you always wished you could write! I say this because I have often thought how handy it would be to have a CMS which took feeds or API data from all the sources that are out there and bundled them into one affiliate site, based on just a few keywords you type in - the lazy man’s software!

Well WAHA certainly does that. Based on a set of keywords that can be partially automatically generated, you can integrate text from news, articles, Yahoo Answers or any other RSS feed, photos from Flickr, videos from YouTube AND, the most important bit for affiliate marketers, product info from Amazon, eBay, CJ, LinkShare and a number of other networks.

The aim of course is to generate a constantly refreshing site that attracts organic search visitors with its relevant content and sells them what they are looking for via the affiliate networks.

What does the end result look like?

Have a look at my test site, Used Musical Instruments, a site you may remember as originally being set up to demonstrate WordBay, and which I never did anything with. I think it would be fair to say I spent about an hour or so on the WAHA version, and you can see the way the different types of data have been integrated with product data to produce something that looks like a website!

Installation

Briefly, installation is fairly simple, although because the source code is protected using IonCube, you do need to make sure your server supports this module. This also means a few hoops to jump through with registration of WAHA, which will also lock you in to using WAHA only with the domain you registered for. So it’s quite a restrictive license, but that’s their right of course.

Setup

The bit you are really interested in is actually setting up the site. The backend admin section is nicely laid out and full of mostly self-explanatory and very functional settings. I never bothered reading any instructions before diving in. However, I should point out that the instructions that do exist are still not great, especially as they are written by a non-native speaker. They might like to think about investing in proofreading of the text (in fact, my company can do that - maybe I’ll offer them a deal :)). However, I asked the guys about some queries I had, and they were quick to reply.

The only slightly onerous bit is setting up some of the affiliate networks. It goes without saying that you need to be a member of EPN, Amazon, and any other networks you want to use on the site (and for some of the content feeds too), and to insert your various affiliate IDs, campaign codes etc. But for some of them you also need to obtain an API key. This is actually quite simple in most cases and makes sense - if the WAHA people inserted their own API key in there they could quickly run into trouble with the number of calls being made to the API. eBay only allows 5000 calls per day, for example, in the basic package. This is the reason I have not used the eBay API for WordBay up till now, but this is an OK solution and I may think about this…

Without going into all the settings, the fun part is of course setting up your niche. This is done through the Category/Keyword section, where you name the title of your category (for example Used trombones, on my example site) and then choose keywords to populate that category. This is made very easy through an automated system which presumably polls eBay for related keywords and pops up a whole list of them. For instance, if I type in “French horn” I get a list comprised of: “Double French Horn, Holton French Horn, Yamaha French Horn, Single French Horn, French Horn”, and I can choose which to include in this category.

It’s as simple as that! The category will then be automatically populated with data from ALL the sources you have specified. It will take a few seconds to generate the page the first time you click (not surprising with 10 different data sources to query!), but after that the page is cached and displays much more quickly.

So after the initial settings, setting up your site pretty much boils down to choosing keywords!

Other settings

Used Musical Instruments is completely generic at the moment, except I changed around the order of some of the data sources, which you can do in the back end, as well as specify how many items to display etc. (e.g. 3 Yahoo Answers). You can also choose from a number of colour schemes, or if you are a bit more adventurous, make your own template - WAHA uses the Smarty template engine. It would have been good to see some more choices of template by default, but then again, all built-in templates would quickly grow old, however many there were.

There is a built in “synonymizer“, which processes articles and swaps in substitute words using some kind of thesaurus, presumably with the aim of simulating original content. You can also manually edit articles. Personally I would not use this - as a native English speaker it is blindingly obvious to me that an article has been “synonymized” and it makes the site look very dodgy.

The autogrowth feature is a very nice one - with it, you can set the speed at which new keywords are “published” to your site. So you can create 100 categories off the bat, but have them slowly appear on the site over a period which you specify, creating the impression of a gradually growing site, which the search engines like! Sneaky!

The bottom line

Well, I hope I have given you a fair overview of the WAHA CMS and its features. The cost of a licence (locked to one domain, remember) is $28 USD and I suppose you want to know if it is worth it!

Well, first the “don’t likes”: I guess the major criticism of WAHA would be that, well, it IS a little “sploggy”. At the end of the day, the claim that it creates a “quality web site” would be seriously contested by Google et al, since it ultimately relies primarily on unoriginal content. Eventually I fear that Google would start getting wise to WAHA-created sites. In fact, after an initial spike, Google search traffic HAS reduced to a trickle, though I am currently getting some 10 Bing search visits a day, which is enough to generate something in sales. However, this does remain a concern.

The “quality” claim could be qualified with the clause, “…for an automated site generating tool”, since the site actually generated DOES look pretty good - it would not necessarily be immediately obvious that it is all auto-generated, mostly because there is so much going on on-screen

I would also like to see some sort of built-in stats. Sure, you can plug in Analytics, but I’d like to be able to see the number of click-throughs for each page, for example. Maybe it’s a lot to ask, with all the other features, but there’s an idea anyway.

Perhaps the licence is a little restrictive too - the price, well, you can judge yourself from reading below. But maybe a licensing option for multiple sites might be worth considering, it could net the authors more sales ultimately too.

The “likes”: The fact is, if you have no objection to an auto-generated site, then they don’t get much better than WAHA. It IS essentially “white-hat” - I believe that none of the data is actually being “scraped”; it’s all taken from official feeds. I was pretty impressed by the effort that has gone into this, the huge number of options, and the inversely proportional amount of effort needed to produce a content-filled site. If you are fine with creating this kind of site, or you don’t have much technical knowledge, or you are short on time (or just plain lazy like me) you should certainly consider WAHA.

The very bottom line

OK, so how much did it net me? After an initial slow start, and having only created a few categories, a few sales started to drift in (after about 3 weeks), such that my sales since mid-March when I installed WAHA have been about $25, and that only through EPN - my sales have been zero through the other networks, (which is about usual for me!) I have recently added a lot more categories, so I hope this will improve things somewhat.

Now that doesn’t sound like much, but I can’t begin to describe how LITTLE work I did to promote the site - no link-building, no AdWords, nothing. I barely looked at it. All visits are through organic search. And the fact is that if I had paid for the licence I would have almost made the money back in a couple of months. After that it’s all profit. So I think a bit of effort in marketing would see you recoup the investment pretty quickly - and if you were ambitious, you could invest in, say, ten licences, make ten sites, and start to see some reasonable returns within a few months.

Hope that’s enough opinion for you to make a decision - well done to the WAHA guys for the effort, now we have to see if it has been worth it! But that’s affiliate marketing, right?

Posted on June 12th, 2009 by markowe
Filed under: Uncategorized | 3 Comments »

I need to point out something that I have only recently discovered. If you have noticed that the keywords you are using to produce WordBay eBay listings are giving you different results to when you type them direct  into eBay search, there is a reason, and right now, if it’s what I think it is, there is not much I can do about it.

I believe it’s because of the new eBay search that is being introduced, to replace the old style search by the end of April this year, and which you may have already noticed on the eBay site (US, at least). Among other features, it offers a more intelligent search which gives you a greater number of relevant results. The old style search only searched on the item title, unless you ticked the box Include title and description in which case you were likely to get all kinds of irrelevant results, but new search has an improved algorithm which finds more relevant items.

But guess what - the eBay RSS feed, which is where WordBay and a number of similar products get their listings from, still appears to implement the old-style search. So unfortunately we are rather limited now in comparison to the search that can be performed directly on the eBay site and there is nothing much I can do about it - believe me, I would be very glad to have the new search working with WordBay as a number of my sites are displaying inferior results to what they could be showing. I hope this new search will also be available through RSS too soon. I will keep you posted.

Posted on March 18th, 2009 by markowe
Filed under: Uncategorized | 1 Comment »