Clicktracks Inserting Hidden Links in Customers Pages?

29 comments
Story Text:

A chance inspection of the Clicktracks hosted tracking solution, that is NOT cheap, reveals some alarming facts about one of the better known tracking packages...

Yes, I am a blithering idiot. Which must be the reason I fell for the trick I'm about to describe:

Background

My site's host provides zero log file access and probably the worst hand-rolled analytics program ever made. Anyhow, I decided it was time to look for another solution. After reading mostly positive reviews on Clicktracks, I decided to give their Professional Hosted service a go.

Yes, the program itself lived up to expectations. I was amazed at its functionality with only a day of data and couldn't wait to have more days to examine..

Then Today

Oddly enough, I was checking out the tool pointed out in the "SEO Swiss Army Knife" thread and followed a link to a link extracting tool at SEO Chat. Why in the hell is my page showing an external link to Clicktracks?!?

<noscript> Trickery

After further examination, Clicktracks is using one of the oldest tricks in the books. Actually, they've put their own twist on the old distribute free tools to get links model. They are freaking CHARGING for the software and inserting links via noscript! And we aren't talking $15 a month.

The Code

<script type="text/javascript">
document. write ('<'+'script type="text/javascript" src="'+document.location.protocol+' //stats1.clicktracks.com/cgi-bin/ctasp-server.cgi?i=CODE">< /'+'script>');
</script><noscript><a href="http://www.clicktracks.com/"><img src="https://stats1.clicktracks.com/cgi-bin/ ctasp-server.cgi?i=CODE&g=1" alt="Web Analytics" border=0></a></noscript>

Hmm.. It seems every person running the hosted version of Clicktracks is linking to Clicktracks' homepage with the alt text "Web Analytics". Once again I reiterate, I'm an idiot for not seeing the link.

The Links

The Rankings for Web Analytics

While not the most popular term by far, it has to be pretty valuable to a company such as Clicktracks. I also concur that these hidden links aren't the only factor attributing to their rankings; however, the links definitely aren't hurting.

I'm Pissed

Even with my limited knowledge of tracking, I'm positive there is not a functional reason for the link to Clicktracks.com. The image yes, but the links serve only Clicktracks' internal SEO goals. I've found no mention of these links in the TOS, which makes me feel Clicktracks is deceiving every single user of their hosted services.

From mom & pop businesses all the way to public corporations.

The Outcome

That's for you to decide. For me, I'm still trying to decide whether I can trust Clicktracks after this little stunt, and of course trust is number one when you're deciding about a hosted analytics program.

Am I blowing this out of proportion?

Comments

Paranoia :)

I am pretty sure this is not part of their SEO-strategy. It might be a stupid way to implement the script but definately not cleaver SEO. If it was they would not have linked to a different domain than what they want to rank with (stats1.clicktracks.com - that, by the way, return a 403 forbidden access if requested outside the script) and they would not have made it an image link but a text link.

I'm fairly sure...

... but feel free to stop me, because I might be wrong. This is calling something from each of your pages from the clicktracks hosted server (remember, this is the hosted soulution) that says that so and so clicked on this link, came from so and so etc. I think it's just part of their click tracking.

Hmm..

The part of the code I'm referring to is the direct link to www.clicktracks.com in the noscript area. I agree that the image and preceding document.write statement are definitely part of the click tracking, but why include a link to their homepage?

I wouldn't have batted an eye otherwise. But when my site suddenly starts showing up in the backlinks for clicktracks.com, it starts to look fishy. I truely hope to be proved wrong, since I was enjoying the reports. For the time being, I'm testing to see if the link to their homepage in fact affects the script's ability to track users.

&nbsp;

What troubles me the most about this is the ALT text, that definitely has nothing to do with click tracking.

&nbsp;

> that definitely has nothing to do with click tracking.

And, I must say, also have very little to do with effective SEO :)

Not defending them without checking

but I do concur with Mikkel that it's most probably not related to SEO. They might require that data (including their installed basis's incoming links) to evaluate their own setup's performance, similar to html editors and blogging software generating pages with their product name in the "generator" meta tag (e. g. HotDog Pro, WordPress).

Yes, they might generate incoming links that way but let's not overestimate the importance of "classical" SEO here: companies of that caliber don't have to rely on good search engine rankings for their business. Sure, it will help but it will only go so far.

Of course, the people to ask foremost would be Clicktrack themselves so why not get them to explain it to you first and take it from there?

It has more to do with effective SEO than you might think.

Google indexes alt text from linked images in the same way they index anchor text from text links. Having thousands of pages containing an image link using the alt text of "web analytics" produces the same results as having thousands of pages linking with the term in a text link.

In this case it worked much better than any possible method of getting the same volume of text links because

A) It didn't cost them anything (buying that volume of links would be quite expensive)

and

B) It flew under the radar for a long time. I first saw it back in December, and they already ranked well for web analytics at that time. My guess is they've been using it for at least a year without anyone noticing. You certainly wouldn't be able to do that if you had used some type of PBM tool to produce the same results.

In the end, they probablly brought in hundreds of new customers who are all paying a recuring monthly fee. Regardless of what happens now, they get to keep all that recuring revenue.

At the end of the day, I'd have to say that it was indeed an effective SEO strategy.

Responsbility for your code

Isn't this code that you agree to put into your own site, or that they put in for you after you agree to it?

I haven't used the hosted solution myself (disclosure: I do use and like clicktracks), but this sounds legitimate, as I'm willing to bet they would let you take it out or change that line if you REALLY wanted to.

I think it was good forward thinking. It really did nothing more than to leverage the power of their growing user base to propel their growth. Sounds like a good strategic plan to me. I think if you're allowing someone to place code on your site you have to question it (to them), and take some responsibility if you allow them to place it.

SEO strategy? Probably. Legitimate SEO strategy? Yep.

And SEO itself might not be as important as many like to believe

I know this may sound somewhat heretical within an SEO/SEM focused environment here, but let's not forget that there are many ways to skin a cat in marketing and SEO is only one of them. Any company (even a 100% online operation) will be well advised not to rely on SEO for their revenue streams alone: it's far too unstable a venue to build a reliable business model on.

Jim Wilson used to state that he only generated about 20% of his traffic via search engines: IMV that's a fairly sound approach. Look what happens whenever search engine indices get the hiccups and all those webmasters relying on their rankings alone start whining and raising hell as if it's the end of the world (which for them, subjectively, it may well be): this is dumb market positioning by any standard. Talk of putting all your eggs in one basket, etc.

So I'm not denying that this hidden link may aid Clicktrack's SEO effort, maybe even to a very large extent. Nor that it seems a mite suspicious and merits some investigation.

All I'm saying is that we shouldn't overblow their intent without further data. Especially not without having given them at least a chance of explaining their action. If they should start hedging, going into denial mode or playing the old blame game (you know the spiel: "former employee ... management not aware of it ... very sorry ... yadda yadda") then it's a pretty clear cut affair that they've been sneaky about it all the time.

But let's hear them out first before asserting dramatic (and essentially damaging) conclusions.

I agree

Quote:
But let's hear them out first before asserting dramatic (and essentially damaging) conclusions.

Has anyone contacted them so they know this is an issue here?

one of the oldest tricks in the books.

sure is an oldie...

You can remove the "a href", and the "alt" as well, all that is required for the tracking to work is the image, the link is not needed at all - it has no function apart from being a link (read: SEO).

>> While not the most popular term by far,

I agree that it's a poor choice of keywords. It reflects the firms view of their product - not necessarily the searchers view. I can imagine quite a few more, say, "converting" phrases for their field of business.

Sure Greg? :)

There is just a few things that made me think it was not SEO'ed besides the questionable keyword selection.

1) I have never seen linkpop transfer directly from one domain to another (from stats1.clicktracks.com to www.clicktracks.com that ranks) unless that domain is 301'ed (if nothing else just for spiders). I haven't checked to see if they actually do cloaked 301s ...

2) Even though engines do read the ALT-text I have consistantly got better results with text links in all cases where we switched from one to the other. Off course, my testing is as limited as any other individual, so I may not have the "full" picture :)

Off course I could be wrong and they did in fact try to SEO the best they know. I would probably have done a few things differently - not the least rotate the text (alt or link format).

&nbsp;

Quote:
1) I have never seen linkpop transfer directly from one domain to another (from stats1.clicktracks.com to www.clicktracks.com that ranks) unless that domain is 301'ed (if nothing else just for spiders). I haven't checked to see if they actually do cloaked 301s

I'm not sure what the stats1.clicktracks.com has to do with it. It is simply hosting the image. The link is pointing to the root domain, and that link contains an alt attribute with a keyword phrase. My experience has been that the alt text is always credited toward the page being linked to, not the domain that is hosting the image.

But to be certain, I'll run it through some tools. It could be that there is enough traditional anchor text out on the web to explain their rankings.

However, I do know for a fact that it was done as an attempt to rank for that term.

Getting blind

Sorry, I must be getting blind - i was so focused on the stats1 domain in the image that I overlooked that the link was not the same. My mistake :)

just alt seems to work well at the new MSN too

I don't see it helping clicktracks much right now, but I just recently viewed results of a test on the new MSN that clearly showed alt text and only alt text resulting in rankings. This was not a competitive area test though which may be why clicktracks is only on page 2 in this case.

Disclosure/SEO

It seems pretty obvious it was done for SEO reasons. If I was a hosted customer, it would leave a bad taste in my mouth if I found this out myself. (I'm a pro customer and not affected)

Clicktracks should disclose to their hosted customers in the terms that this link is there. If potential customers don't like it, then they can use a different service. But at least Clicktracks won't appear to be underhanded.

Clicktracks should also have their alt text randomly rotated to get more SEO benefit.

Hopefully, John will comment on this thread.

have some respect

First off, John Marshall is one of the nicest guys you'll ever meet in this industry. To use words like trickery in the same sentence with John Marshall's name is just plain idiotic. Stop by the ClickTracks booth at any search related conference and meet the man and tell me if I'm wrong.

I just got off the phone with John after having a short discussion about this issue. I'll admit he's a bit worked up about this - not because he got caught out - but because of the sensalistic kneejerk spin that has become the hallmark stamp of Threadwatch breaking a story. Where's the journalism here? Did anyone email or call John to ask about it before firing up the presses?

Was it used for seo? Yes it partly was but it's just an alt text - it's not even a real link which would have been far more effective. As it stands the sum total gain of this has been about 3 spots on Google. ClickTracks did very well for Web Analytics prior to employing this technique.

John sent me a paper on why the are doing this. I'll not post the whole paper here but there's one paragraph that sums it up:
Do I have to use the stuff?

No, not if you prefer not. There is alt text in there because we felt it was OK to get some slight keyword ranking from it but there is no obligation to include this if you prefer not. As long as we have enough sites using our stuff our data will be statistically good and our analysis of this will improve the accuracy of our tools.

The rest of the document presents a well thought out argument concerning text only browsers and surfers with javascript turned off.

Yes they could have done it without the alt text but would you? I know I'd have done it. I'd have probably even made it a real text link and rotated several phrases through it.

This practice is common and nobody seems to have problem when it's right on the page and free ala the MarketLeap tool and pretty much any free counter service. Drop it in a no script and everyone gets all fired up.

You talk about trusting ClickTracks...what's not to trust? Did John take your money and not deliver what he promised? Is the data inaccurate? Does the service not perform as advertised? Those are valid reasons to not trust ClickTracks. A bit of alt text in a noscript is certainly not a reason to distrust anyone.

Are you blowing this out of proportion? yes

&nbsp;

Quote:
I just got off the phone with John after having a short discussion about this issue. I'll admit he's a bit worked up about this - not because he got caught out - but because of the sensalistic kneejerk spin that has become the hallmark stamp of Threadwatch breaking a story. Where's the journalism here? Did anyone email or call John to ask about it before firing up the presses?

It's nice to know you're friends with him, and that you think he's a great guy - and on that recommendation im quite sure he is.

However, your_store was concerned about what he discovered, and thought it a good story, so he posted it. I couldnt care less, but i did think it was an interesting story so i promoted it to the homepage.

If members were required to email the subject of a story before posting, we'd never publish anything.

It is a good story, and one of interest to many people i think.

And where is John? If he's so worked up about it, and Threadwatch is doing him such an amazing injustice, why does he not post here about it? It would seem a simple thing to dispel any misinformation, set the record straight and demonstrate to those of us that aren't personal friends of his that he is indeed a straight up, great guy no?

&nbsp;

First off, my intention was not to out Clicktracks for using <noscript> links. I could care less if they had bought <noscript> links from every site on the entire web and were using them to rank for Mickey Mouse. However, they didn't.

This practice is common and nobody seems to have problem when it's right on the page and free ala the MarketLeap tool and pretty much any free counter service. Drop it in a no script and everyone gets all fired up.

The key difference isn't <noscript>, it's the fact Clicktracks' is a paid service. If they didn't want to charge a monthly fee, perhaps I'd discuss hosting a ROS link in exchange for their service. However I'd much rather stick to the agreement I thought we had, which was for them to provide tracking and for me to fork over $1,000 a month. It seemed like a win/win to me.

Instead they want to charge for a product, while seeking the linking benefits of a free, viral program. I'm sorry you get one or the other but not both.

Was it used for seo? Yes it partly was but it's just an alt text - it's not even a real link which would have been far more effective.

The argument that it is a non-effective technique does not diminish the fact it is being employed. It's akin to saying Clicktracks didn't have enough to gain for this to be about rankings.

There is alt text in there because we felt it was OK to get some slight keyword ranking from it but there is no obligation to include this if you prefer not.

Fair enough regarding the alt text, but that's not really the issue. Why is the link there in the first place? The script is fully functional with just an image call. The only functional reason I can even dream up for the link is to track spiders. If that's the case, wouldn't it better to send the spider to a tracking uri rather than Clicktracks' homepage?

You talk about trusting ClickTracks...what's not to trust? Did John take your money and not deliver what he promised? Is the data inaccurate? Does the service not perform as advertised? Those are valid reasons to not trust ClickTracks. A bit of alt text in a noscript is certainly not a reason to distrust anyone.

To base my trust solely on Clicktracks' promises, would be idiotic. If your plumber promised to fix your sink, but failed to mention he'd be spray-painting his name on your front door, would that be a reason to distrust him in the future?

Again, it's the links I'm concerned about:

  • Either I was duped into inserting an ROS link, or
  • The ROS is an unintended side-effect

Either way, it hurts my confidence in Clicktracks. Especially, when I'm supposed to be trusting them with one of my businesses' most valuable assets.

Please put down the rakes and pitchforks...

Quote:
Where's the journalism here? Did anyone email or call John to ask about it before firing up the presses?

I think we've all come to expect TW to break stories first and ask questions later. Not necessarily a bad thing...unless it's your head on the chopping block;) We're like a sewing circle (c'mon admit it), and we want to hear the news first.

I had mentioned this thread to Clicktracks staff, and actually spoke with John earlier today as well.

Gotta say I'm totally in agreement with Oilman on this one. Trust hinges on being delivered what you were promised, and I don't think that was violated in any way. Users KNOW what is in the code because THEY put it their themselves. Can you really hold any fault in someone else for code that YOU placed?

From my conversation with John, I gathered a few things -

  • John is a very pleasant guy, and mentioned he had considered posting here, but didn't want to start an unnecessary frenzy, and wasn't familiar with the site or it's rules. (*one benefit is John got to learn about a great site...perhaps not in the funnest way;)
  • There is use for the tracking code. (I'm not a huge tech guy, so I don't have all the details here from a phone conversation)
  • Clients always place their own code.
  • Yes, there was some intended SEO benefit as an added bonus. Is this a bad thing?
  • Clients can certainly contact them if they have issue with this, and they will allow them to remove it...it is not a mandatory thing.

This is a paid service, but so is hosting or webdesign. If I place a VISIBLE ROS link on a web design clients site (back when I did client design), I've generally not even had them ask about it.

Quote:
...but failed to mention he'd be spray-painting his name on your front door, would that be a reason to distrust him in the future?

What if he gave you the spraypaint and asked you to do it...if you did it would still distrust him, or would you just ask him why you were asked to do it, and perhaps not do it if you didn't like his name on your front door?

Not tryin' to be a jerk here by any means, but if you had issue with the code wouldn't it have been easiest to ask the vendor FIRST before accusing them in a public forum?

&nbsp;

It sounds to me like you and oil know these guys, and naturally want to defend friends - hell, i do it all the time!

At the end of the day, it's an interesting bit of contraversy - and no amount of "hey, they're really very nice!" is going to stop it from being just a little bit sneaky...

I think they should have made a point of telling people about this in their agreement (which i assume they did not), and i also think they should get in here and make a statement.

Im not knocking you or oil st, but im certain you can see that this would be different if you were not friends with these guys eh?

Put it to Bed

Simple stuff really;

Quote:
Hi all, im john from clicktracks - sorry we caused a bit of a kerfuffle over this link thing - we've added a note in our agreement telling people about it, so thanks for letting us know! If you have any questions, post here or email me...

All done and dusted...

>one last note

No worries at all Nick. Today was actually the first time I had ever spoken with John. I had just given their staff a heads up because I have been very happy with their other standalone optimizer program and their support and communication has always been top notch. John e-mailed me to ask that I call and give him my opinion on the matter.

Glad to see they're adding the statement anyhow. It's always good when in doubt to let folks know.

Sorry!

No!

Sorry, that's just a made up quote :( I meant it to mean what they should do...

Your post proves something to me though - when customers are really happy with a product, and really happy with the support, they become evangelists - i'ts a wonderful thing.

All they need to do now is get their arses in here and say hi, and all will be well i reckon.

Come on John, you know you want to...

Fancy Goo being that quick

on the uptake, eh? :-)

[Update: Hm, bummer. Better make stuff like that clear in advance, Nick - sure, rumors are great, but I'd say we have more than enough of those around anyway, heh.]

Come on Put your hands up...

Of course it's part of the SEO strategy and pretty smart if you ask me and also a little cheeky!

Lets be honest the link isn't required and serves no purpose except to link back and since its hidden it can only be for SEO purposes... surely :)

Actually, they are inexpensive

I have found clicktracks to be very reasonably priced, and they give *wonderful* metrics. Much better than many web metrics programs.

The Kerfuffle

I've been surprised by the vigorous debate here, and I'm sorry to have stirred it up.

We're doing the link thing and the img thing because of some research into robots. Without going into too much detail, detecting robots in our logfile based products is tricky because they obfuscate the useragent. We need good quality heuristic data, and we get that by comparing logfiles and javascript for known sites, mainly our own because we can examine that minutely. We need to attract the devious bots and linking to our site is one way to get them to crawl. We do other things as well, of course, and have other research into bots, gathering data so we can improve our heuristics to detect automatically.

The alt text was indeed thrown in as a minor SEO effort though I doubt it's made much difference. We need the img and we need the href for the above to work, the alt text was a step into SEO territory. This was a mistake and we'll change it.

You're welcome to email me : jmarshall at etc. We're not some faceless corporation. I value having the chance to discuss stuff. Really.

John m.

Thanks John

Welcome to TW, and thanks for the explanation.

I think that just about wraps this up now right?

alt text

>> Yes they could have done it without the alt text but would you?

Nope. Alt text is required by the W3C everytime you use an image on the web. Of course you should have an alt text.

The alt attribute must be specified for the IMG and AREA elements

Further - the alt text must make sense and be relevant:

* Do not specify irrelevant alternate text when including images intended to format a page, for instance, alt="red ball" would be inappropriate for an image that adds a red ball for decorating a heading or paragraph. In such cases, the alternate text should be the empty string (""). Authors are in any case advised to avoid using images to format pages; style sheets should be used instead.

* Do not specify meaningless alternate text (e.g., "dummy text"). Not only will this frustrate users, it will slow down user agents that must convert text to speech or braille output.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.