Saturday, 31 May 2008

Think of what you are saying...

A customer this week sent me a rather hotly worded email telling me that if I wasn't up to the task why couldn't I tell him and why had it taken me so long to get the website back to him. My reply, in rather a few more words, was to remind him that 4 weeks ago I'd sent him the links to the live site and asked him to review it and that after my emailing him several times, he said he was too busy with his garden to review the site.

Why then am I getting the blame for the delays - when I've been waiting 4 weeks for a reply? He also said he didn't like the look of the site, which is based exactly on the slides his graphic designer sent to us. Looks like this could be a fun site to finish off over the coming weeks... A possible redesign and goodness knows what else, at least this job is not on a fixed price.

Whereas another job that has taken most of my time this week is a fixed price contract and my sales guy is letting it run away. This customer is desperate to go live the middle of next week, let I'm still waiting for the latest changes to the items page, which need to be completed before he'll sign the pages off and I can actually build the site. Even if I get the latest changes lsit on Monday and can get the changes done the same day, it's not leaving me much time to build the database interface, especially when I have other customers wanting work from me.

But that's life - and the sooner we reign in these daft customers the better - the more new sites we can go live with.

Friday, 30 May 2008

The redesign is live...

Well the Cottage Holidays website redesign has gone live overnight and bar a few teething problems, the site is up and running. I managed to sort the Google Maps problem before publishing the site so the cottages do show a location map - now I just have to see if I can include the maps into the tourist information pages...

It didn't do quite as straight forward as it looked to be. During final testing I noticed that only 500 of the holiday cottages had been loaded, out of the 1700+. Plus there were also a selection of French properties, that didn't have anywhere to go and crashed the load.

Once the full selection of 1700 went on it was a lot more difficult to load, but I did get it there and now the latest database is loaded. I'll be loading the database again over the next few days to pick out the new cottages as each region's home page and the site's home page do list new additions to the site. At the moment they are just showing random cottages loaded today - not what I wanted to show!

The only glitch that I've noticed so far is that in a couple of places the region is not decoded in the display of the cottage name. Not difficult to change, just a lot of places to change it in.

Hopefully I'll manage to get the search engines interested in the site and over the next few weeks and months will be reporting increased traffic. Can't be much worse than the current month has been! Once we hit the end of the month I'll post the results as another benchmark of where the site is at the moment and in a few months I'll be able to look back at both sites and sets of data to see if the redesign has brought in more traffic.

Thursday, 29 May 2008

Cottages rebuild almost ready...

The first phase of the holiday cottages website rebuild is all but completed. It's still short of several aspects, but most aren't needed for a go live. Yes, it's been a long time coming, but over the last week or so I've not had the time to move it along...

The main missing features at the moment, and the only ones that need sorting before I put the new version live, are Google Adsense adverts and the Google Map feature.

Google Adsense is easy to sort - I just need to swap machines and then plug them in. The maps are proving a little bit harder. In theory it's easy, in actual fact it's a tad difficult. The code is workign stand-alone, but as soon as I try to include it into the page it's full of errors. And again, this machine doesn't show them. It looks like the phase 1 release of this site will need to blank out the Google Maps - shame!

Building the site wasn't quite as much of a pain as I expected, especially the creation of all of the pages. I thought that creating around 3,500 pages might need to be split into many scripts, but it's done in just the single step.

The plan is that over the morning I'll get the Google Adsense in, probably blank out the map section, and then rerun everything to put the new site live and let Google et al get trawling the new pages.

Then I can start adding the tourist information and hotel information onto the pages - anything to really make the site have plenty of unique content. I'll keep posting as it develops.

Wednesday, 28 May 2008

Let your webdesigner do their work...

Whilst it's actually very nice when a customer arrives knowing what they want, along with drawn out screen prints to work from, it can be taken too far.

One guy I'm working on at the moment took my first sample and put it into photoshop and moved things around a little to explain how he wanted it to look. That's great, but then when the instructions start to tell me how to optimise the site and what I'm doing wrong - then the instructions are going too far! It's especially awkward when the instructions are based on old information - "put in loads of keywords in the keyword metadata", "put some text at the bottom of the page in the same colour as the background", "list all the keywords in the picture alt tags".

Fine, thanks for that help, but the ideas you are suggesting at best don't work, at worst can get your site banned.

Here's the simple truth. If you are getting a webdesigner to build and maintain a site then don't go for over the top on page optimisation. Just go for well built, well designed pages. If the page is over optimised using dirty tricks then when the search engines change their algorithms your site is going to drop, and maybe get banned. If you want to be tricking the search engines using the latest discoveries then you need to be in a position to be able to change your site at a moment's notice - not join a long queue of work to be done as maintenance.

If you are paying someone to do the job for you, then let them use their knowledge and build the site for you - don't tell them their work, no matter how much you have researched. Or will you be telling the pilot how to land because of your research next time you go on holiday?

Tuesday, 27 May 2008

I'm not really that negative...

I don't want to appear to be too negative when I'm talking about checking reciprocal links - just realistic. There are a good number of sites out there that are genuine, but then there are also a good number that aren't. It's less than 10 days since I ran the links check on CompareMortgageRates and when I started the process today there were exactly 400 sites in the directory. Of these, 17 no longer link back to me - yet they were when I looked less than 10 days ago.

Extrapolate over 31 days and that could easily have been around 45 sites missing link backs - over 10 percent! I had a look through some of the sites with missing links and those that I looked at were all 3-way links - initiated by the other party. It's what I was saying in that article - the person being paid to search for links suddenly has no financial incentive to keep my link there so pulls it without warning.

I see no harm in 3 way linking, if done fairly. But in my experience it's never fair. It's usually done so that someone being paid to promote a site can do so more easily. If they took the time to install link software on the site that is paying them to do the work then there would be a long term investment - the links directory would be there for as long as their customer wants it. I suppose that's the difference between my way of working and others. I would give the customer full control, they want all the links to be able to slowly crumble once the customer stops paying.

Give me a fair link and I'm happy!

Sunday, 25 May 2008

Does my aggressive link building work?

OK, so over the last month I've become very agressive / determined in deleting links that don't benefit me. Does it do any good?

Well, as I was reviewing my list of link check dates, I made an observation. In the past I've recorded the date that I've checked that links are still active. On the whole, I've deleted links that are reciprocated any more and noted that date. It was only when I neglected to do this that as an experiment when I remembered, for a couple of customer sites I tried deleting more links (they were so full of links after not checking for a few months they took ages!).

My sheet shows that from January to July 2007, every month I carefully ran my link check every month. Before that it was pretty adhoc.

There was also, at some point either before or just at the start of that period, a time when my CompareMortgageRates directory became corrupted and wouldn't add links for a few weeks - but people kept trying. Eventually I rebuilt it - with a lot of links missing. It was shortly after that it hit PR5.

What about normally? Well last year, the site's performance improved until October, then tumbled. It picked up again in January and February this year, March wasn't so good, April better and May not so good.

And whay about the links check? Well as I said above, until July I was checking monthly, then forgot it until October, then again until January. So when I was checking and for about 3 months afterwards, the site traffic was good. And when I did a full check in October, about 3 months later traffic picked up. Same again after my January check.

So it looks like careful admin of the links directory is important - let it slip and traffic falls. By checking it every month the traffic is at its best. When I let the checking slips, a couple of months later so does the traffic.

It's not instant - and that is obvious. It takes the search engines a few weeks to pick up all of the links pages and the links pages pointing in and then for Google etc to fully update their directories. Whether being more aggressive will help, I don't know. But with what happened when the directory crashed, it could be the case.

Friday, 23 May 2008

Am I right to delete so many links?

Well, for a start, let's make it clear. They are my links directories - so I am within my rights to do what I want! Ethically though, should I have mass deletions? Does it affect my page ranks?

Well I first tried a mass deletion back in early December - over 6 months and 3 page rank updates ago. I tried it on a PR5 and a PR4 website - and both still hold their page ranks. More than could be said for other sites. What's more, the page rank of the directory pages on these sites is quite good.

But why do I feel I should do it? I've noticed that many of these links that come back as grey barred pages are on pr5 / 6 or low pr sites. In the first case, I'm suspecting foul play is getting the other site a good ranking whilst hiding the link page my link back is on. So why should I help them when they don't help me?

In the second case, especially the pr0 / grey barred sites that I've been linking to for 18 months, something is wrong with the other site, so it is probably best to get out of that relationship.

I do state on the better PR links directories that I will only exchange with PR pages. This leads to people putting links onto home pages, submitting their site, getting them approved then either removing or moving the link to another page, hoping that my software will pick up the new location and I won't notice the PR drop. Very friendly when the recipient is getting a link on a PR3 / PR4 page.

Also, a lot of these are casino, mobile phone and adult orientated and absolutely nothing to do with the link page category. This is hindering the effectiveness of the page for people link swapping the way I want to be working.

Lastly, by filling up my directory with these useless swaps it is making it far harder for me to run my monthly checks. This means I'm not maintaining it as well as I should do. This is affecting the PR of the page and probably the site.

How would I feel if it was done back? Well, I've had enough link swaps rejected because my link page hasn't got enough PR and I estimate that 30% of link swaps are no longer reciprocated - the person requesting them have removed their link for some reason, and not told me.

So it seems I'm not the only one. Some of my link pages may also be grey barred - but the person requesting the link exchange can see that when they make the request. And if I never do anything to better the PR of these pages, then they will always be the same. Something needs to be increased!

Thursday, 22 May 2008

Am I too aggressive?

Some people may say that I'm too aggressive when it comes to deleting links. But then I do fairly well accept most link requests and then just delete those that at a later date I decide are not benefiting me.

Should I do this? Well, it's my link directory - I'll do as I like! If the other site's owner is upset that I've deleted their link, then maybe they should have given me a better link in the first place.

OK, my link pages aren't whiter than white and many are not page ranked. But it's got to start somewhere and there's a good chance that those pages that aren't page ranked, but are obviously very popular with people requesting links, could be being hindered by all the spam requests. And if that's the case, then the good sites I'm linking to aren't getting a fair exchange because of the other sites.

So I like to delete links that aren't beneficial or relevant. At least I'm that open to requests! And as I've already mentioned, I only expect pages to be page ranked after they have been through 2 full page rank cycles.

My problem is that some idiots set up links that are absolutely irrelevant. Why try to fill my 'activity holidays' page with links to fake rolex and casino sites? OK, take over that page and they then become relevant to the theme of the site and give me a link back from an uncached page. Is this what the owners of the properly categorised links want, who are giving me links from PR2 pages? I'd much rather have 1 link exchange with a similarly themed site that gives me a link from a PR2 page, then 50 casinos linking to my travel website from grey barred (and usually uncached, or sometimes using rel=nofollow...) links pages.

Yes, it's harsh just deleting the links. I did used to send the owners an email first asking them to put the links back, but very few replied and with many the email just bounced. So now I just delete the link and have done to it. Some of them email me months later asking where their link is and telling me how important it is to have a link from their site. If I'm generous I look at their links page and usually find the reason I've deleted the link. Not one has so far inspired me to put the link back in place.

Wednesday, 21 May 2008

What to check on existing links

Every month, at least, you should check ALL of the links you have exchanged. If any aren't playing ball, then contact the site owner, or be like me (a lack of patience) and just delete them.

Why, well why link to a site that's not playing fair and linking to you? It's of no benefit to you. Removing the link teaches them a lesson and means that the number of out links goes down and the space on your links pages goes up - more room for good links.

So what are you watching for?

First and foremost, is the site you are linking to still live and is the page that your link was put on also still live? If the site has gone, certainly delete the link. If the page has gone, then either ask the owner why or delete it. If your link isn't on the page you were expecting to, it's possible that you have moved up or down a page, so check that.

Next I refer to this blog and lock back to see when I've last recorder page rank updates. I give sites the benefit of the doubt, I'm not looking for the most recent update, but the date of the update before that. That would mean that the check in progress looks back not to the 2nd May update, but the 29th February update. Allow a week before that (it takes Google a week to prepare page ranks) and (bear with me) I then expect sites that I've exchanged links with prior ro 22nd February to at least have a page rank 0. My thinking being that they have been included in 2 page rank updates, so should have been noticed by now.

If the page linking to me is grey barred, then I consider deleting my link. It's likely that either there's something wrong with their page structure or site, or they are blocking the link.

Also, and my link building software does this for me, check that the site owner hasn't changed their robots.txt file and hasn't slipped rel=nofollow into your link. If they have, they are up to sly tricks. You could contact them and ask them why, but if they are up to tricks, they will probably do something else.

My moto when it comes to accepting links is simple - "if in doubt, delete it!"

Tey, that's pretty harsh. Why, well where's the point in having 200 links in from non-cached pages? I've just deleted around 250 links from my comparemortgagerates.co.uk website, leaving around 400 active. Why, well most weren't linking back to be for whatever reason, but the rest just weren't giving me anything in return.

Remember that Google's guidelines don't have anything against structured links directories of selected sites - it's free for all links they don't want. As long as there is an element of links being rejected / deleted then it is accepted that you aren't free for all and you shouldn't upset Google and so on.

So I've just made room on my links pages for plenty more new sites. No doubt 30% will quickly remove my link, but I'll remove theirs in return.

Tuesday, 20 May 2008

What you should be looking for when you accept a link?

I've already mentioned that there's a grey area about accepting / rejecting links because the offered links are on pages that aren't cached. The page might be cached because of bad practises, or because it is simply too new.

But what else should you be looking for?

Well, if the page is cached or has a page rank (even 0), then that's a good start. It means Google has found it and has access to it.

What it it's not cached / page ranked? Go to the site's home page and look for their resources / links page link. Click it and then probably you will find yourself on the link's directory home page with a list of categories. From here try to click through to the page your link is on.

If you can't find the links page by click from the home page (sometimes you need to go via the sitemap) then search engines aren't going to find it either. And that's why it's not cached. If you can't navigate from home page to links page then I would alsways say no.

As you click through the pages watch the page rank. If suddenly it disappears then go back to the previous page (with page rank) and look at the link. Has it been hidden in any way, e.g. through rel=nofollow in the link.

Also take a look at the site's robots.txt file and make sure that's not blocking the links directory. There are many different tricks here. Basically, have a quick look and if there is any blocking then reject it. Otherwise, give the site the benefit of the doubt and check again in a few weeks, or after the next page rank update.

Other things to look for include rel=nofollow on your link. With a clever bit of coding, actually not that difficult, a person will see an OK link whereas a search engine will see a link blocked by te rel=nofollow, or hidden totally.

How do you detect this? Quite simply, look at the cached version of the page and then view the code. If there is javascript hiding or altering the link or it is any way blocked then you will see it in the code. And if the site is presenting a different version of the page to search engines, now you will spot it.