| This 38 message thread spans 2 pages: 38 (  2 ) > > || |
|Outsourcing link development for SEO|
my experience so far
| 4:49 pm on Mar 28, 2007 (gmt 0)|
First, let me say that blog and directory spamming are fairly common in my market. We're a long time player and have always positioned well without doing this. Two of our main competitors do it and they typically outrank us. We've tried doing "real" link development (focus on real sites and real traffic) but we haven't had enough progress to beat out these two competitors. Combined with that is the fact that some of the newcommers have started blog and directory spamming and are now making some gains on us. Because of this we decided about 6 months ago that our current link development was inadequate and that we would try to outsource it. Our goal was *not* to do blog and directory spamming, but the goal was to do link development for SEO purposes, i.e. search engine rankings.
I put a project into one of the outsourcing systems requesting link development. I defined very clearly my parameters in order to avoid directory spamming. Things like "no pages named links", "no pages with more than 25 links", "on topic sites only", "no reciprocal links will be given", "no bad neighbourhoods", "PR4 or higher sites only", etc. etc. My hope was to find someone who would go out there and truly "sell" the virtues of our site to real sites in order to have them link to us.
I received a few proposals but only one that looked worth it. They told me they were a company in NYC and that they were ISO9002 certified. I checked out their site, read their reviews and talked to them. They seemed to talk the talk and understand what I was going for. I hired them.
Things started off very procedurally. I was assigned a project manager and given a long questionnaire to fill out. It was apparent that things would proceed according to their procedures and less according to my definition of the project. I decided this was ok because they were responsive when I pointed out discrepancies and I figured this is typical of an ISO company. The other thing that became apparent was that this company was in India. Eveyone I talked to had Indian names and accents and they only communicated very early in the morning.
My first deliverable was 50 links. Exclusively directory spam. Violations of my original request everywhere. I got on the phone with them and discussed it. It was clear that they had their way of doing things and everything else received a "does not compute". I was able to get them to replace the worst offenders and decided to continue to completion. First, they were all on-topic sites, all had some PR and I was able to reject any with more than 100 links. I was also able to give them some variations on the anchor text. Second, I decided it was no worse than what my competition was doing and it was working for them, so why not allow a little for me. Finally, I figured the proof would be in the affect to rankings. If things went up, then I could call the project a success.
The rest of the link deliverables were pretty much the same. I let the links cook for a little while and then did some evaluations. The issue that I saw was that they'd pointed them all to the homepage with only a few anchor text variations. However, I was ranking #1 for one of the less common variations and overall (it's always hard to tell) I think my rankings had improved. I decided to change my tactics a bit.
First, I asked them to do another batch of links, but this time I sent them my own titles, descriptions and links. I made them all deep links with many variations on title and description.
Second, I put out another request for a blog and forum writer. I realized that I needed to manage the project on a very granular level; i.e. "I need these kinds of links from these kinds of sites on these kinds of pages with these kinds of surrounding text and you solicit with these kinds of communications". Because of this involvement I decided the most I could handle was to find a writer to participate in forums and blogs and drop strategic links. I didn't mention link development or SEO because I realized this would invite my outsourcers to use their own tactics.
I received several proposals to my second request. I hired a few and they recently got started. So far I'm getting much closer to what I expected than the first project. It requires explaining things like how I don't want links in forums that don't actually link the url or show the link or that they can't leave the www. off the front of the url or that they should use a deep link instead of a homepage link when possible. It's too early to tell what the affect of this is.
Yesterday, I received a bit of a shock about my first project (the one I'd extended with the company from India). One of my customer service reps forwarded me an email she'd received. It was a site owner complaining about the spam email she kept receiving requested that her site reciprocate a link. It turns out that the company doing the link development is basically impersonating my company in link exchanges. They set up a site like www.someweirdname.com/mycompanyname which is a "directory" of links. They find sites with reciprocal link pages, then they post a link on this impersonating site. They then send emails to the site owners that look like they're comming from my company. The from address is email@example.com (kind of thing). The signature was "webmaster mycompanyname"! Wow, did I fly off the handle.
So, that's where I'm at. I'm not going to make any recommendations on what might work for you and your site, I'm still not sure what will work for mine.
| 5:13 pm on Mar 28, 2007 (gmt 0)|
Although im new in link development from reading this forum and experience i found out that one quality website on topic can be more useful than a 100 links. On avarage it takes 2 weeks of search before i can find a quality website to link too. Puting a PR4 limit is a bad idea. I had a PR3 website shoot me up the ranking better than Pr6. What would be a better idea to have a company search for quality website related to your niche and you decide wether to link to or not. just an idea.
| 5:22 pm on Mar 28, 2007 (gmt 0)|
|I defined very clearly my parameters in order to avoid directory spamming. Things like "no pages named links", "no pages with more than 25 links", "on topic sites only", "no reciprocal links will be given", "no bad neighbourhoods", "PR4 or higher sites only", etc. etc. |
First off, the above are poor metrics for quality. All of your above metrics have zero, absolutely nothing to do with quality. So your project was doomed from the beginning.
I know of one so-called bad neighborhood site, in porn, that has 106 dot edu backlinks. The dot edu pages I looked at appear to be freely given, in other words, they are actual citations. Total backlinks are almost a quarter million. That site is a perfect example of how the "bad neighborhood" metric has zero to do with the quality of a link. You would be lucky to obtain a link from that porn site.
The requirement for PR 4 is a totally random metric, also with zero relevance to what constitutes a quality link. You mean to say that you won't accept a PR 3link from a .gov resource page called links that lists quality sites in your industry? Do you see how your metrics have nothing to do with metrics of quality?
Your first step is to understand what a quality link is.
| 8:10 pm on Mar 28, 2007 (gmt 0)|
"The other thing that became apparent was that this company was in India. Eveyone I talked to had Indian names and accents and they only communicated very early in the morning."
I would only recommend something like this to a competitor who had pissed me off at some point. In fact, I have a competitor who used an outfit from India. They wreaked havoc with the reputation of his site. I have nothing against directory submissions at all, but they gave him the spammiest directory links and wrote articles for submission that butchered the concepts with which he was trying to be associated.
It's funny, I was thinking of posting a thread similar to this because I have a site that needs additional link development to move it closer to page one. Fact is, I just don't have any time to spare myself due to the requirements of other sites.
It would be nice if you could find a trustworthy and competent company to hunt down links for you, but my sincere belief is that it is dangerous to trust the task to anyone else. To date, I have only approached one individual who does such work and he was unavailable due to his own work considerations.
| 8:33 pm on Mar 28, 2007 (gmt 0)|
|Do you see how your metrics have nothing to do with metrics of quality? |
No. And from reading forums here on how to measure site and link quality I'm not sure how I would easily define quality in a job request for a company who knows nothing about SEO, my market or my company. I'm not taking offence, I'm truly curious how you would define quality in such a situation. What parameters would you put on your job request that would ensure quality links? I suppose it's quite possible that this can't be done without an understanding of the market and company, but then that's a different issue. When I posted the job request I certainly realized that I wasn't going to get the kind of link development that I would do if I had the time to do it myself, but that's a trade off I felt I had to make.
| 8:33 pm on Mar 28, 2007 (gmt 0)|
What happened to you was terrible, but just like a house built on a poor foundation, it doesn't matter how well built the house is, it's going down. It's admirable that you had parameters/metrics in mind, and you tried to build the house the right way, but on a base of quicksand.
Here's an example of why well-meaning metric demands go horribly wrong
>"no pages named links"
I understand your intent with this parameter, but it's one of the most innacurate absolutes floating around the SEO world. In fact, pages
named links or even URLs such as links.html, links.asp, etc., or within directories like /links/ are often exceptionally trustworthy, and the engines can make the crap/not crap distinction with relative ease.
I'm not much for pagerank, but the entire first page of results have plenty of it, are trustworthy, and yes, they are named links.html and some also have the word links in the title tags. They might not be a fit for your content, but they are for someone's, and links from those sites are to be prized. They are golden.
I'm not picking on you, I admire your thoroughness. But I have yet to find a single absolute related to link building. All tactics work, yet none work. It all depends on the content and execution. You cannot conduct a link building campaign for site A and use the same campaign over again for site B.
In this case I'm not sure that firm would have listened to you no matter what you'd asked for.
[edited by: martinibuster at 10:00 pm (utc) on Mar. 28, 2007]
[edit reason] Removed specifics. [/edit]
| 2:41 am on Mar 31, 2007 (gmt 0)|
Thanks Spikey that was a interesting read for those of us who have not tried outsourcing. I'd be interested to hear of any further developments.
| 1:06 am on Apr 2, 2007 (gmt 0)|
Actually you will most likely find some improvement from the links they got for you, but probably not the kind of changes you were looking for.
I manage alot of full campaigns as well as others here, and here are the criteria I loook for regarding any links:
1. Themed site and theme page to our website
2. Cached page in Google
3. Not a free for all, links on the page are all related to what our topic is.
4. Link limit - simply depends on how good the other factors are but obviously 60 links on a page is not a great thing.
5. Page is linked to from the home page or in a respectable way that visitors and spiders find it often.
The best links are usually from 'in content' as opposed to pages which are link containers.
I usually divide linking into categories:
Related Specialty are the links that when done right will give you a substantial boost and trust. They take much longer to secure and need to be approached on a per site / page basis. There can require a phone call to the webmaster, or ongoing conversations in emails before they are secure.
Related Standard - the standard on topic links we see much of the time from links pages, articles etc.
| 4:40 am on Apr 2, 2007 (gmt 0)|
I am so sick of saying this.
INBOUND LINKS CAN'T HURT YOU!
Seems like every post is somebody saying they are so worried about directory links or getting links from bad places. If this were true I can go spend $50 and get my competiter banned. If you can't rank for something it is not a penalty it just means you don't know how to do SEO. If all they are doing is getting directory links and spamming blogs you should be able to catch them. Those links are not worth much.
| 11:00 am on Apr 2, 2007 (gmt 0)|
|INBOUND LINKS CAN'T HURT YOU! |
Their origin sure doesn't seem to matter as in it will not raise penalties.
But outsourcing link development, by these guidelines, to people who don't know about an industry, do not care about instructions and effectively don't know the region and the targeted audience...
Repetitive, overused anchor text is a given.
And a penalty for it.
So inbound links can hurt you. Hit the threshold for spam and suddenly you're pass the top and are going downhill.
Besides, it was interesting to see these guidelines you thought up. You've mentioned you wanted this campaign to have an impact on SEO, yet all your instructions were more or less focusing on user-related parameters.
Quality link in SEO is:
Relevancy + Trust
Relevancy as in the source, the link and the target being relevant at least in a broad manner. Pagerank, number of links, color of the background, name of the webmaster, age of the author doesn't really matter as long as the pages come up in the top 1000 in Google for a related one or two word phrase. At least when you're talking pure, dry SEO. ( Which is what you aimed for, isn't it? )
If you're worried that it's only trust that keeps them from being washed away by the filters, or that no one would believe that it's a natural link, you can have a preliminary evaluation and drop some of them from your list.
... of the story about the quality factor. But there are many others to consider. ( PageRank gives the WEIGHT to the links, thus is important as a multiplier. And there's also the factor of how much traffic the link brings, which sometimes has nothing to do with either of the above, think popular forums with no cache, famous, legit blogs using nofollow etc... )
[edited by: Miamacs at 11:16 am (utc) on April 2, 2007]
| 12:18 pm on Apr 2, 2007 (gmt 0)|
*INBOUND LINKS CAN'T HURT YOU!*
I'm seeing "intent" feature quite a bit, these guys have set up a three way system whose intent is all too clear and in linkage terms curtails the "room to maneuver" the site would otherwise have.
| 1:23 pm on Apr 2, 2007 (gmt 0)|
I'm not convinced that no inbound links can hurt you. If you have shown a history of trying to manipulate google's algos, how sure are you that they will never penalize you for it? Maybe not currently. But maybe tomorrow.
| 2:06 pm on Apr 2, 2007 (gmt 0)|
|I'm not convinced that no inbound links can hurt you. |
I'm pretty convinced inbound links can potentially hurt you. I suspect nasty competitors can do it all too easily. :(
|Red Mud Rookie|
| 2:24 pm on Apr 2, 2007 (gmt 0)|
Outsourcing link development is risky business whichever way you look at it.
I know of recognised SEM agencies in the UK who offer this service to their SEO clients, but who, upon hearing you wish to terminate your SEO contract, claim they will have to remove all the inbound links they've been maintaining.
Now this may be calling your bluff to make you sign another 1 year extension, but the fact of the matter is that if these guys are doing it and doing it well, they probably own the network from whence your inbound links come.
Do you want to be held to ransom like this? I don't.
We need to be honest with ourselves here folks... sooner or later the engines will identify these "highly effective, yet 100% ethical" bought link networks. If you're paying someone to link their network of sites to you, you are effectively manipulating your listings by artificial means.... in other words the natural/orgaiic listing ain't so natural after-all.
Short term gains may be there for the taking, but long term we need to stop chasing algorithms and get on with developing great user content and websites that make other sites WANT to link to us.
| 2:50 pm on Apr 2, 2007 (gmt 0)|
Spikey, may I ask how much the first company charged you? I recently got quotes that allowed me to draw a clear line between spammers and potential quality firms - with the latter ones so expensive that for my own humble endeavours, writing the e-mails myself is much more efficient.
| 5:32 pm on Apr 2, 2007 (gmt 0)|
|Short term gains may be there for the taking, but long term we need to stop chasing algorithms and get on with developing great user content and websites that make other sites WANT to link to us. |
Good call, Rookie.
In my experience, the only way to truly move up in the SERPS is to increase IBL's by building the best site you can. You do that by adding quality content, marketing, and increasing organic traffic.
Everything else (IBL's, revenue, popularity, authority, etc.) will come... as long as you keep up the quality.
Unfortunately for those who are impatient, all of the above takes time. Trying to buy your way onto page 1 of the SERPS will often cause problems.
| 6:42 pm on Apr 2, 2007 (gmt 0)|
before you start and outsource work, you should try and do the job by yourself.
Try and define what you think is a quality link. Try to get some. If you can't do it nobody else will. Why? It's your baby, you're enthusiastic about it.
Once you had your first success stories and failures you will be able to refine your goals and have somebody else do it for you.
| 7:37 pm on Apr 2, 2007 (gmt 0)|
Spikey, I think your tale is more generally representative of the attitude of many outsourcers to detailed specifications: some just ignore them.
I recently outsourced a small HTML project to a coder that was new to me but had a fairly extensive history of excellent project completion and near-perfect ratings - exactly the kind of coder I recommend when someone says, "how do I choose an outsource company?" The first iteration of the project looked good, but ignored almost every detailed spec for table structure, CSS use, etc. After multiple cycles of us pointing out more places where the submitted code didn't meet the spec, the coder opted not to continue. I made a partial payment for the graphic design and finished the project internally. A huge waste of time for both parties.
I couldn't believe that this high-rated coder didn't read 90% of the spec. To test the degree to which bidders were reading the spec, in a subsequent bid, I put a clause in bold specifying some information to include with the bid, and noted that any bids lacking it would be discarded. Although this was at the beginning of the bid request in bold type, at least 2/3 of the bids failed to include the requested information.
If you've ever seen the Far Side cartoon where the human is giving a detailed lecture to a dog and the dog hears only its name ("blah blah Ginger blah blah blah blah..."), you know what dealing with some (but not all) outsourcing firms is like. I think in your case they just heard, "blah blah 50 links blah blah blah blah." :)
As a general rule, I'd select only the highest-rated contractors with lengthy histories, and when you DO find one that listens to what you say and executes accordingly, hang onto them! If you post a project and get just one or two bidders, you may end up dealing with a less than ideal contractor.
| 7:39 pm on Apr 2, 2007 (gmt 0)|
*INBOUND LINKS CAN'T HURT YOU!*
PLEASE STOP. Improperly linking to your competitor will kill the link-to page, the thing is if you get cought your (.) is in big trouble.
| 11:19 pm on Apr 2, 2007 (gmt 0)|
|Everything else (IBL's, revenue, popularity, authority, etc.) will come... as long as you keep up the quality. |
Hmm - This is completely affected by the industry or type of website. Some with quality content will attract links, while others wont to any great degree.
| 10:18 am on Apr 3, 2007 (gmt 0)|
Link development work can be compared these days to cold calling. Although cold calling is a marketing initiative and helps directly in sales, link building helps indirectly (helps in increasing search traffic).
|First off, the above are poor metrics for quality. All of your above metrics have zero, absolutely nothing to do with quality. So your project was doomed from the beginning. |
This is one of the occasions where I have to agree with MB. One of the biggest mistake outsourcers make while defining Quality website is put up a minimum qualifying PR for linking to other website.
Lets look at some numbers (these are just based on our internal experience) :-
1) An average link researcher can research and find 40 links in a day. The work includes finding links, uploading them on your website, sending link request to the other website, do a few followups on previous links and delete old links which did not respond. This is assuming this site has not done extensive link campaign before. If it has, then the number of links found in a day can be less (because then sites have to be compared with previous database of sites already approached and over a period of time, number of related sites left for link exchange decreases).
2) Out of these 40 links, if the site is not PR 0 then there is a conversion rate of 10 to 15 %. This ratio goes down is the link exchange is three way.
3) Link exchange has become a menace these days and hence lot of people complaint about spam. This is one of the reasons your link exchange company used your name while doing link exchange. If they used their own name for each of their clients, their web host would most certainly take their website down.
4) Using above metrics you get roughly 880 link requests in a month (22 working days approximately 175 working man hours). The company will charge you based on what their man hour rates are.
So you are looking at about 110 to 130 reciprocal links in a month. Now you add your criteria for minimum PR and your conversion rate decreases. Any additional criteria will bring down conversion rate.
However, there are many vendors in market who have already developed a network and are ready to provide you quotes on per link basis. Since they have contacts in place, they dont need to spend time in research or reminder link requests. I once heard about a company in south america which charged $2.5 per reciprocal link. The advantage is cost. The disadvantage is that they are most probably selling it to hundreds if not thousands of people. It might or might not work long term.
|I would only recommend something like this to a competitor who had pissed me off at some point. In fact, I have a competitor who used an outfit from India. |
I am not sure if I understand you correctly. But if you are implying that companies from India either cannot understand SEO/ SEM/ Link building then I think you are mistaken. Companies can be good/ bad /worse irrespective of their geography.
| 4:05 pm on Apr 3, 2007 (gmt 0)|
I think this discussion is getting a bit too academic - from a rather practical viewpoint, I wonder if somebody here has made good experiences with outsourcing link building and if so, what the briefing/checking of the firm involved.
Personally, I have made no good experiences with outsourcing anything at all so far.
| 11:34 pm on Apr 3, 2007 (gmt 0)|
I haven't had very good experiences either. I don't have time to do it myself but I know I should find it. I have always been concerned about people not knowing what they were doing or having to train someone, which takes a long time.
In the last couple years, I even hired a couple name conference circuit/forum experts figuring I could maybe get the best out there and have a level of trust. But all I got were big cashed checks and future conference stories.
So it came back to doing it yourself or taking risks and/or blowing money.
| 7:54 am on Apr 4, 2007 (gmt 0)|
So with adamxcl, spikey and me it's three now that havn't made any good experiences outlinking.
Anyone out there that knows not only how to specify a job, but who also did specify it and made good experiences?
I would be interested in an example rather than the theory of what constitutes a good link.
| 10:58 am on Apr 4, 2007 (gmt 0)|
Add me to the list of people with bad experiences outsourcing linkbuilding. From the descriptions above, I believe I've spoken to some of the same companies.
I was willing to throw a few dollars at these sort of projects as an experiment. Like Spikey, I set guidelines (realising it was far from ideal).
In return, I got several dozen spammy reciprocal links c/o Linkmachine auto-gen pages. The contractor had found a way to game the reciprocal links.
If anyone can recommend good link-builders (beyond the usual A-listers with pricetags to match), I'd be interested in hearing more via PM.
| 11:44 am on Apr 4, 2007 (gmt 0)|
>>I would be interested in an example rather than the theory of what constitutes a good link.
Good luck with your site(s) then.
I would also be interested in examples of how to drink lots of beer without getting a hangover or a beergut. But people insist on telling me not to drink so much beer. Can you believe it, eh?
| 1:02 pm on Apr 4, 2007 (gmt 0)|
What I meant was that those people sharing theory don't say that they have ever outsourced themselves - EVERYBODY with experience in outsourcing that I have talked to so far has made only bad experiences.
I think many people on WW know already what a good link is. What I wonder about is whether I could get a "don't drink too much" advice rather than a lecture on the effects of ethanol on your brain (to stick with your example).
After all, WW is also about practical advice.
| 1:55 pm on Apr 4, 2007 (gmt 0)|
Well, my practical advice for you (in particular) is to put "outsourcing link development for SEO" (i.e. sourcing low-quality reciprocal links by employing people who don't give a hoot about your site or the other one) down your list of options and instead to think first of all about the "linkerati" who control the high-quality - sometimes one-way - links in your area.
That would be my very practical advice.
| 2:02 pm on Apr 4, 2007 (gmt 0)|
Over the years, some WebmasterWorld members have reported good experiences with inexpensive in-house link monkeys - high school students, stay-at-home parents, etc. Obviously, some training would be needed, but these individuals are far more likely to respond to your direction than a firm that wants to get your links as quickly as possible and move on to the next project.
I haven't done any offshore link dev outsourcing, but based on my experience with other kinds of projects, I'd suggest looking for the following characteristics in a bidder:
1) They tell YOU how they build high quality links, not just say, "no problem, we do that!" when they read your specs.
2) Excellent rating history on similar projects, preferably over an extended period of time.
3) Realistic pricing - if a price seems too good to be true, it probably means the firm is taking shortcuts. (Needless to say, though, a high price doesn't by itself indicate quality.)
Regardless, be very specific with your criteria and communicate well and frequently with your contractor. Ask them to start with a smaller number (say, ten) for your review.
Not all poor outsourcing jobs happen because the contractors are unethical or untruthful - in some cases, the problem may be either an incomplete understanding of your specs or the feeling that these specs are "suggestions" instead of firm requirements that must be met in order to get paid. Most contractors communicate in English, but that doesn't mean they find it easy to understand your concepts. Communicate & clarify.
Don't pay up front - use an escrow system with impartial arbitration. Most contractors will try hard to meet your specs both to get paid with no hassle and to preserve their quality rating.
| 2:17 pm on Apr 4, 2007 (gmt 0)|
I'm going to run the risk of drift off topic here a bit.
I haven't had to go link hunting, but if I had to, I think I'd rather outsource content development and do the link hunting myself.
You Can always edit the outsourced content before it gets posted online. That might be harder to do with links you didn't personally negotiate for, on sites you don't control.
| This 38 message thread spans 2 pages: 38 (  2 ) > > |