At SES London next week, Chris Sherman wants the linking panel to talk about what works in 2005 that didn't work, say in 2003. The general feeling being that people have pretty much figured out that linking is important, but also some of the rules have changed. I thought I would try to put various linking strategies into a grid, to give people an idea of the best linking strategies being in the top right, with the poorest being in the bottom left. The horizontal axis represents the quality of traffic that will result from the linking strategy and the vertical access represents how difficult it is for a competitor to duplicate the link strategy.

For those looking to learn from the forums, I hope this gives some ideas, but for the pros around here, there could be huge disagreements about where I put things and I would be very interested in any feedback on what else could/should be added and what is in the wrong boxes as well as whether there any ideas in here that you haven't thought of yet.

What could be improved on this table for tactics for developing inbound links?

http://www.myap.co.uk/images/linking.gif

Key (For the best ones)
Strategic partnerships: Here I mean real vertical business relationships - Microsoft and Dell; Ford and Michellin; Haliburton and the Iraqi government oil field development agency...
Have this feature in your site: Here I mean neat gimmicks that add real value to the linking site - examples are: have this foreign currency converter on your site, have this link popularity analyser on your site, etc, where the link is in the code donated to the linking site.
RSS Feeds: Pretty obvious and more or less exploded over the last 12 months, but is this in the right box?
Private/white labelling: More than simply an affilliate program - white labelling is where your entire technology is wrapped/branded using another company's brand

I am pretty sure that for different websites and industry sectors, different rules will apply, and of course a good strategy can be badly implemented (like a link from Microsoft.com to Dell.com with an anchor text of "click here"- but in general what is right and what can be improved on this table?

What could be improved on this table for tactics for developing inbound links?

Offhand I think there could be 2 grids - traffic vs SE benefit (possibly a 3rd for conversion rate? Or a 4th for brand / image?).

In that respect, PPC for example would have low SE benefit (arguable), medium traffic benefit and high conversion rate benefit.

Another example could be PDFs - maybe lower SE benefit, but potentially higher traffic benefit (ie, if it's a newsletter going to a large audience). A bit vague though! :)

Also, I would think about including (or differentiating) links from authority sites, particularly academic and government.

I think quality vs difficulty to copy is a great measurement - it certainly brings in elements of marketing and business (rather than just pure tech as can often be the case with SEO type stuff), which is why I thought traffic / conversion / brand may be good alternative measurements as well.

MG

Offhand I think there could be 2 grids - traffic vs SE benefit
Many thanks MarketinGuy. I have to agree that there are definitely two elements to linking strategies - you've hit the nail on the head identifying them - rankings vs human traffic - and tackling the two issues are very different, making two grids quite logical. But as you pointed out, this logic quickly moves to three grids... fours grids... By forcing myself into one grid, I am trying to avoid the blackhat/whitehat discussion - concentrating on the business case for linking before the rankings case. This is why I have put search engine links themselves in the grid (and not at the top right either). Perhaps it is naive to try to put two very different philosophies on the one page, but in the SEO world we tend to ignore the overall marketing case for linking and in the marketing world we tend to ignore the SEO case for linking. Given the two different audience that will be in the room, I am hoping this amalgamation forces both groupings to think slightly outside the box.

Maybe colour coding might be a good compromise Grren for links affecting SE rankings and Blue for links only effecting users and red for "For goodnees sakes - what are you doing to your buiness trying a stunt like that?) - that might help to identify whether links for SEO motivated reasons compare favourably to links in offline PDF files for business reasons.

Hmm perhaps that may complicare the system though? How about just just refining the remit of the grid, eg:

Linking strategies that work for SEO in 2005
Linking strategies that work for building traffic in 2005
Linking strategies that work for increasing conversions in 2005
And so on?

MG

You mean three (or more) columns, rather than a table. Yes - that's a thought - but then I run into more problems, since leverage of two or more tactics affects the quality of the metric in all three columns... a good on-theme landing page affects conversions (but is this in otself a "link" strategy or a "content" strategy? - good anchor text affects the rankings and visitor traffic, but combine the anchor text and the landing page content and you have a winner. It starts to feel very granular. (Maybe I already made it is granular and the "creative guys" will switch off anyway?)

I'll have a play with other layouts and representations (not my forte!). Many thanks MG.

Anyone got thoughts on the substance? Have I got the mix more or less right for 2005?

No I meant simply keeping the grid purely "What works for SEO" or "What works for building traffic" rather than to accommodate all aspects - just makes life easier! ;)

If you want to have multiple grids, then one could be SEO vs difficulty to copy. Or another could be Building traffic vs difficulty to copy. And so on.

My thinking is that although "difficulty to copy" would remain the same, the benefit to traffic levels can be different to the benefit to SEO, etc. Therefore different grids respresenting different aspects.

MG