Brendan Cooper, your Friendly Digital Social Media PR, ummm, Thingy

Got questions about the PR Friendly Index? Me too. I’ve tried answering them here. Let me know if you have any others.

Why did you create it?

I once presented some work to a client in which I’d identified interesting blogs that we could work with. The client’s immediate question was “So, which is the most influential?” I had no idea, and furthermore not a clue how this could be quantified. It was quite embarrassing, so I resolved to find out as much as possible about it.

You can trace this all the way back to my first post on the subject in which I started out on this path. Since then I’ve come across other great efforts to do this, such as the Bloginfluence calculation and Todd And/Ad Age’s Power150 list, which I initially based my efforts on but have slowly developed my own approach.

Why don’t you have badges for it?

Badges like these, you mean? Just copy and paste the HTML into your own page, and it should work. You can also find the badges at the bottom of the PR Friendly Index page.

---

<a href="http://brendancooper.com/the-pr-index/" title="Click to see Brendan Cooper’s PR Friendly Index of top PR blogs!" target="_blank"><img src="http://thefriendlyghost.files.wordpress.com/2008/02/pr_friendly_badge_160×33_1.gif" alt="Click to see Brendan Cooper’s PR Friendly Index of top PR blogs!" border="0" width="160" height="33"></a>

---

<a href="http://brendancooper.com/the-pr-index/" title="Click to see Brendan Cooper’s PR Friendly Index of top PR blogs!" target="_blank"><img src="http://thefriendlyghost.files.wordpress.com/2008/02/pr_friendly_badge_160×33_2.gif" alt="Click to see Brendan Cooper’s PR Friendly Index of top PR blogs!" border="0" width="160" height="33"></a>

---

<a href="http://brendancooper.com/the-pr-index/" title="Click to see Brendan Cooper’s PR Friendly Index of top PR blogs!" target="_blank"><img src="http://thefriendlyghost.files.wordpress.com/2008/02/pr_friendly_badge_120×30_1.gif" alt="Click to see Brendan Cooper’s PR Friendly Index of top PR blogs!" border="0" width="120" height="30"></a>

---

<a href="http://brendancooper.com/the-pr-index/" title="Click to see Brendan Cooper’s PR Friendly Index of top PR blogs!" target="_blank"><img src="http://thefriendlyghost.files.wordpress.com/2008/02/pr_friendly_badge_120×30_2.gif" alt="Click to see Brendan Cooper’s PR Friendly Index of top PR blogs!" border="0" width="120" height="30"></a>

---

<a href="http://brendancooper.com/the-pr-index/" title="Click to see Brendan Cooper’s PR Friendly Index of top PR blogs!" target="_blank"><img src="http://thefriendlyghost.files.wordpress.com/2008/02/pr_friendly_badge_160×160_1.gif" alt="Click to see Brendan Cooper’s PR Friendly Index of top PR blogs!" border="0" width="160" height="160"></a>

---

<a href="http://brendancooper.com/the-pr-index/" title="Click to see Brendan Cooper’s PR Friendly Index of top PR blogs!" target="_blank"><img src="http://thefriendlyghost.files.wordpress.com/2008/02/pr_friendly_badge_160×160_2.gif" alt="Click to see Brendan Cooper’s PR Friendly Index of top PR blogs!" border="0" width="160" height="160"></a>

---

What does it show?

It shows how blogs are faring according to publicly available, and easily obtainable, metrics.

These show different aspects of performance: Technorati Authority shows how many other blogs link to a blog, while Google web search shows how many hits you get when you type that blog’s address into Google, and therefore gives an indication of popularity within the web as opposed to the blogosphere.

Does popularity equate to influence? I would argue that, if you want to take a rule of thumb, then yes, the top ten blogs in the index carry more influence than the bottom ten. But influence vs popularity is a difficult thing to measure. Entire companies are devoted to measuring this. I am not a company. I am just a man.

What does it not show?

This is a quantitative measure of blog performance, not a qualitative measure of what people are saying, or how they influence people, or how they change people’s behaviour. It doesn’t show who is linking to them. It doesn’t take into account the set-up behind the blog, that is, whether it’s just one person or a group of people. Basically, take what I’ve just said it does show above, and what it doesn’t show, is everything else.

Why did you initially call it the PowerPR index?

Because initially it was very like Todd And’s list but I wanted to concentrate on more pureplay PR sites rather than including marketing or other comms.

Since then I’ve been told that some of my blogs don’t quite fit into that category, but then again I’ve moved away from using the same metrics as Todd so I don’t see this as a problem. I recently changed it to the PR Friendly Index as a result. All is explained here.

Is the PR Friendly Index updated automatically, or by hand?

It’s a combination of techniques, none of them involving coding as such.

• For the Technorati figures I use the Google Spreadsheets ImportXML feature. So, for example for authority, the call would be: =importXml("http://api.technorati.com/bloginfo?key=[key]ebb8a8&url=[url]", "//inboundblogs" ), changing [key] for your Technorati API key and [url] for the blog address. For inlinks, just change the ‘inboundblogs’ parameter to say ‘inboundlinks’ instead.
• For all other figures I use Google Spreadsheets again, in this case the ImportHTML feature: =importHTML("http://blogsearch.google.com/blogsearch?hl=en&q=[url]", "table",6), again changing [url] for the blog’s url. The table value can occasionally change but this just requires a quick search/replace to fix.

I just pour these figures in a worksheet and figure out the rankings there. This is done by rebasing all the values out of ten, except for Google Blog Search over the past month, which I give a value of out of fifteen to give more importance to recent activity.

I know these methods aren’t incredibly streamlined and they probably don’t represent best practice, but it’s the closest I’ve come to making this thing tolerably easy to compile without delving into creating my own web-based apps.

How do you decide which blogs to include?

I simply monitored blogs over a period of months to build up my own PR blogroll, then based it on that. Some people think they should be included and some think what is in there shouldn’t. But how else would you choose your sample set? No sample could ever include all the PR blogs in the blogosphere.

Will blogs be removed/added?

Yes, but I’m trying to keep the number restricted to 100 because I’m only allowed 50 calls per Google Spreadsheet, which means I have two spreadsheets per set of metrics which is just about manageable. The criteria for removal will be if I see a blog really doesn’t get updated any more, and for addition will be if I find a really great blog that I haven’t included, or if someone asks me nicely.

How is it some blogs go up when they’ve been inactive for a while, or down when they’ve been active?

I think a blog can continue to climb despite no postings for several reasons:

• It’s been around for a while. It’s an unfortunate but valid consequence of blog longevity that it will rise up the rankings, simply by dint of there being a lot of ‘vestigial’ links to it in the blogosphere.
• It’s been posting recently enough to retain momentum. Two or three months absence isn’t all that significant. Technorati Authority looks at the past six months, for example, so I would be surprised if a couple of months here and there made all that difference.
• It’s on the blogroll of more active blogs. When they post, they contain a link to the inactive blog, which can increase its count on, for example, Technorati Inlinks.
• It’s only risen relative to other blogs. It’s entirely possible that a blog’s absolute figures have indeed fallen, but by a lesser extent than the other bloggers around it in the index. They could fall more quickly by being younger or less linked.

On the flipside, active blogs can go down for all the opposite reasons: they’re quite new; they’re not heavily linked up; or their neighbourhood blogs have all gone up. My blog went down five places in the December issue so believe me, I’m totally impartial here!

Any developments for it in future?

I recently added up/down/no change indicators. I also made it graphical and removed the 1-10 scale because that masked the true differences between blogs and between metrics.

I’m looking into developing an index total based on the total figures I obtain when I create the index. This would take as its base the value as at November 2007, when I first started using the current metrics, and would look at how the index has risen/fallen since then to give an idea of whether the PR blogosphere, as I see it, is expanding or contracting. I actually have an index running on the spreadsheet but have several ‘flavours’ of it and want to see which one gives the best picture.

It would be nice to make it cover more blogs but I’m limited by the number of calls I can make in a Google Spreadsheet. It would also be nice to make it update automatically like the Power150 but that’s way too in-depth for me right now.

How long does it take to put together?

When I started out I did it by hand, so it took literally hours. Then I discovered the Google Spreadsheets approach and, after spending quite a lot of time figuring out a good workflow between that and my local Excel spreadsheet, it’s now down to less than two hours to update, compile, error-check, format, analyse and upload it. This isn’t too bad on a Sunday afternoon.

It becomes a pain when I change things, for example when I moved to charts rather than figures, and when ‘odd’ results pop up that I have to verify.

Isn’t it just link love?

I can’t deny that my readership shot up but this was a total surprise to me. Since then I would say, if you want to pump your subscription rate, do a list! But this was honestly not the reason I did it. I was too naive in the ways of blogging to have predicted that it would have that effect.

You’re using two Technorati metrics. Isn’t there redundancy?

Technorati Authority is the number of blogs - not links - that haved linked to a site over the past six month. Technorati Inlinks is the number of links in total that come into a website.

I guess I could subtract the one from the other and this would show you all other links, then I could also maybe ‘weight’ Technorati Authority more highly. But I’m not sure how often Technorati Inlinks is recalculated. So I decided to leave well alone in this respect.

This is actually where I started out on this journey several months ago: having compiled a load of metrics I started looking for relationships so I could better understand them. And you know what? I couldn’t find any.Therefore, I don’t think the relationship between these three metrics is quite as simple as a nice linear equation. This is why I separated them. And you can see in my latest index that this does show that some blogs have, proportionally to the population, different scores for each metric.

I need to dig more deeply when I can find time, to get to grips with this non-linearity. I’ve considered using regression analysis but it made my head hurt.

Does Yahoo measure inbound links differently from Technorati?

It must do. The Yahoo figures are much, much bigger than Technorati. Quite possibly they are closer to Technorati Reactions than inboundlinks, but there’s no API call for Technorati Reactions so I can’t verify this without manually compiling them all. Again, I cannot find a decent correlation between Technorati inlinks and Yahoo inlinks.

Isn’t there a lot of overlap between your metrics?

Probably, but it’s nearly impossible to discover exactly where the overlaps are. There is quite possibly a bit of Technorati Authority in the inlinks figure, and maybe some of the Technorati inlinks overlap with what Yahoo is picking up. So, given the near impossibility of working out exactly what is what, and of having absolutely discrete metrics that definitely show only one or other aspect of a blog, I decided to include them all.

Why did you include Google?

Google I take to be a decent reflection of the number of hits in the ‘web world’ as opposed to the ‘blogosphere’. Again, I get quite different figures when measuring Google hits vs Google Blog hits.

Why didn’t you include Alexa or Bloglines?

There is plenty of circumstantial evidence to suggest that Alexa is heavily biased towards tech users. Also, I cannot get Alexa and Bloglines figures across all the blogs. I can obtain figures for all blogs for all the other metrics, so to my mind they’re a fairer measure.

Of the metrics you use, which do you think are the most valuable, credible, important?

This question was very timely and hit upon an issue I’d been wanting to explore for some time. Here’s the answer. Then I realised I got it a bit wrong, so here’s the correct answer.

What have you learned from doing this?

• Your stats go through the roof if you put together a list.
• People not on the list will hate you.
• People who are on the list but have moved down will dislike you.
• People who are on the list and have moved up will love you.
• You will spend ages maintaining it.
• You will spend ages replying to people’s comments on it - but in a good way.
• You will understand that you understand nothing. To wit…

• Influence. It’s not about how many, but who. To take an example, Jonathon Porritt’s blog is not one of the most popular environmental blogs, but he is undeniably a highly influential person in this sphere. If I had the resources, I would create my own tool that would trace the people who read a source rather than just count them. Various companies already have this. I’ve taken the approach that there are metrics freely available which enable me to establish popularity, and you could to an extent argue that popularity equates to influence. Roughly.
• Blogs. Is this one person? A group? A company? A ‘pro blog’ where people make money from it? An ‘am’ blog where people just dabble in their spare time? What about top pros who blog in their spare time? This links to the influence question. Actually, nowadays I just skirt around the issue of whether something is a blog, by classing it as whether or not it appears on Technorati. I’ve sidestepped it and handed responsibility for classification to them. They offically state that they measure ‘the live web’, of which the blogosphere is a part.
• Bloggers. ‘Am’ journos? What about when they’re paid? Who says they’re exempt from legal issues? I’ve decided that a blogger is whoever runs the blog! Generally if someone seems to me to speak sense, I read them.
• PR. If it feels like PR, it is PR. That is, it looks to creating relationships and conversation. This is why I include people like Steve Rubel. He tended to touch on issues that I felt were relevant to PR.