InfluenceFinder – Review

Earlier this week, we told you about the launch of InfluenceFinder, which debuted at SMX London 2010 on Monday. Use invite code [seochicks] for pre-approval on a 7 day trial.

Having now used the tool for a couple of weeks, I wanted to follow up with a review from my perspective. If you missed the story of the launch and case study then you might want to check that out first.

Overview

It’s really important to stress that the InfluenceFinder product (and I think the name gives it away really) is designed to help it users find influential link targets. Although the product does draw and re-index from the Majestic webmap, and is therefore extremely expansive – the main objective is to improve the working efficiency of the linkbuilding process. In their own words InfluenceFinder seeks to facilitate “more conversations in less time, with better outcomes”. Where I’m going with this, is that it may feel at times as if there is a lot of layering and filtering by default. If you’re a bit of an old hand used to making your own target decisions, then this may require an element of trust in the product.

Ease of Use

InfluenceFinder has a very clean, spacious look and feel (until you get to the results set, though more on that later) that takes you on a staged walk-through. You can tell the product is aimed at medium to large agencies with multiple teams of linkbuilders or large websites with sizable in-house teams, as even from the outset the user is taken by the hand and the potential objectives explained…

Screenshot of InfluenceFinder Start Project Stage

Once you have your project type in mind, stage two contains lots of helpful prompts and signposts at the various stages. In addition, if you omit to fill in any of the fields, or if you only enter one data point in a field that more often than not would contain multiple entries, then there are handy user prompts that check in with you, and make you aware of the potential implications – which is nice.

User Prompt from InfluenceFinder

Of course the most important screen is the view projects screen. In this view you are presented with your list of domains in a split screen view, with the list of domains on the left, and the list of pages on the right. It’s quite useful to have this split-screen view to detail instances of multiple links from the same domain, as of course each type of link may have different levels of influence and signals that can be applied as filters. On the whole the view projects screen looks and feels like your Outlook inbox, and is therefore fairly intuitive.

On the down side, even though you can drag and drop the split, I’m not sure the vertical tiling is the best method of presentation here; though I am on a laptop with a 15.6 inch monitor. It is a little fiddly to scroll and reveal the various useful columns, and the row depth is a little heavy on the eye too.

View Projects Screen

Features

InfluenceFinder is packed full of features that really do help to take some of the donkey work out of cleaning target lists. In most cases these features are pretty cool, sometimes a little like overkill.

Features that Work Well

1. When compiling your project you have the ability to add or upload keywords (for anchor text analysis) which you can classify as “brand” or “non-brand”, which can be useful if you’re interested in using type of anchor text as a filter criteria.

2. There’s a detailed range of “signals” displayed as columns, such as Influence Index, Relevance Index (both expressed as percentages), if the link source is a blog or not, if the keyword (brand or non-brand) is in the title of anchor text and to which source the domain links to – which may be multiple sources if you have submitted a list of multiple domain targets.

3. There are some pretty sensitive filters that can be applied depending on how tight you want to tune your list of targets and how large your original project list may be. In particular the Advanced Link Source filters offer some pretty cool options.

Advanced Filters

Features That May Take Some Getting Used To

Although as I’ve mentioned the user interface is a doddle and extremely intuitive, the vertical tiling and sheer amount of filters and options mean you can almost get carried away playing with the various refinements. Of course once you get used to what’s available, you can then identify more quickly which elements are going to be most useful for most projects and focus there. A couple of things I couldn’t work out…

1. When you view your project you are presented with the first 100, of the total, but only “X” are shown due to the default filters that are in place. It isn’t clear exactly why “X” are hidden.

Results Display

I think it would be helpful if this said “(552 hidden by filters).

2. I wanted to see how easy it was to reveal hidden link sources, however when I managed to run a project that in the end produced “Link sources 0 – 0 of 0 (2 hidden), I couldn’t work out what filter settings where preventing the display of these 2 hidden sources. It could just be my newbie incompetence, but I think it would be useful to have some way to reveal all hidden sources, or remove all filters in a single tick.

Summary

I’d say that InfluenceFinder is a product that does exactly what it is intended for. It succeeds in taking a large data set and applies its own meaningful signals; plus allows you to fine-tune according to some additional useful criteria. At first sight it can seem like there is almost too much choice, in terms of filter criteria, but of course just because it’s there doesn’t mean that you have to use it. It’s a great tool that should be really useful for any medium to large agencies, and also have features that allow for project sharing.

Id also suggest you check out Wiep Knol’s review of the product too.

InfluenceFinder to Launch at SMX London

UPDATE: May 18th 2010 – The team at Linkdex have provided an invite code for SEO Chicks readers. Please use the invite code [seochicks] for pre-approval for a free 7 day trial at InfluenceFinder.com

Lead by serial entrepreneur and founder of Netrank Ltd, John Straw founded Linkdex in 2008. After two years in the making Lindex is about to soft-launch their first product at SMX today. InfluenceFinder is that product, and I’ve been having a little look at the beta version, plus a rather interesting case study using Econsultancy as the test subject.

InfluenceFinder is positioned as a tool for search marketers to improve their working efficiency by applying additional scientific methods to an extensive webmap; such methods include complex decision trees designed to add more value to returned data. Rather than returning the most expansive list of targets InfluenceFinder seeks to return more targetted data with additional ‘signals’ for relevance and authority.

I’ll talk you through the Econsultancy case study, as that identifies some such interesting signals…

1. Objective

I met Linkdex V.P of Product – Matt Roberts last week, and Matt explained that in the Econsultancy case-study the InfluenceFinder team wanted to see how the product stacked up against other backlink resources, not just in terms of volume, but also in terms of value.

2. Methodology

Working with Econsultancy, the InfluenceFinder team were given access to Google Webmaster Tools data for the site www.econsultancy.com and could therefore compare this to the open API data from SEOMoz Linkscape and MajesticSEO; against Influence Finder.

They looked first at commonality.

Pie Chart Showing Backlink Commonality

SEOMoz and InfluenceFinder Commonality

What is interesting to note here is that 58% of the links reported by Google in Webmaster Tools, are not covered by any of the aforementioned backlink tools. In addition, though this is well known, it is clear that Google Webmaster Tools doesn’t present all of the links it knows about.

InfluenceFinder claim to re-index the backlink data so that it is a much fresher sub-set of MajecticSEO, with which it appears to have most commonality.

3. Additional Value Signals

Expanding on the value of the returned data, the InfluenceFinder – Econsultancy case-study then goes on to establish its’ case for adding value, by focusing on two of the relevancy signals presented in the project results. Firstly, the “blog” signal…

InfluenceFinder uses comlex decision trees to identify (and they say within a level of accuracy to 95%) if a potential link target is a blog or not?

I can see why this would be useful, as blogs can often be really efficient link targets. Additionally it is always easiest to start a list for blogger outreach using the blogs that already link to you.

Bringing this back to the Econsultancy data, applying this filter to identify the “Is a Blog?” signal, whittled the target list to 2723 blogs.

Going further, the next signal applied to this data was “heartbeat”, which is how they have described if a site is active, however there is a bit more intelligence applied to the heartbeat algorithm so that e.g. feed pushers are stripped out. Essentially then the “heartbeat” signal is more a measure of natural regular activity, than just activity.

4. Conslusion

In the Econsultancy case-study, the InfluenceFinder product stripped an expansive list of backlink data into an efficient list of blog targets which they state is effective in that the product succeeds in its’ aim to “turn large lists of link prospects in to smaller more accurate and actionable lists. Better lists result in more conversations with the sites that matter.”

So that’s the InfluenceFinder Econsultancy Case-Study. Linkdex readily admit that a case study of one isn’t the largest sample size, so I’m happy to give you my perspective and I’ll share some data too! Later this week I will follow up with my own thoughts on the product in terms of user-interface, speed, performance, ease of use, amount of data, usefulness of data and more.

See you at SMX.

The Best Job In The World

It’s Wednesday night, and I’m sitting trying desperately to think of something to blog about. I considered doing something about the election & social media, but quickly realised that I am bored to tears by the election, so I’m assuming everyone else is too. That still leaves me with the question of what to blog about. Then it dawned on me, I’m spending my Wednesday night working, I’ve spent every night this week working in fact, and the weekend. This should, really, be a slightly depressing thought, I’m sure for most people it would be, but the fact is, it’s not. The reason it’s not, is that I love what I do, and (I’m sure I’m not alone in this) it’s almost a hobby as much as a job.

I look forward to getting time to catch up on my favourite SEO blogs, I love the challenge of getting stuck into a problem site with no obvious answer, and sometimes I even love writing a sites worth of title tags. I realised that maybe as SEO’s we have the best job in the world, and here are my top 5 reasons why

The debates

There is nothing like an early morning SEO debate to get the blood flowing, whether it’s on the merits of page validation, or the usefulness of the canonical tag, get more than 2 SEO’s in a conversation about, well, anything really, and you have the recipe for an engaging & exciting debate. I admit, very little is ever actually resolved, but it rarely gets (too) nasty.

The changes

It’s an oldie but a goodie, SEO never stands still. While the fundamentals change slowly, there is always something new and shiny to play with. Working in SEO guarantees you a new toy to play with nearly every week, and none of them ever come with instruction manuals, just adding to the endless hours of fun while we all work out just how piece A fits with joint D, oh wait no, that’s the shelving unit I need to put up.

The superiority

We’ve all done it, you watch someone you know trying to start their first blog, and tell you how they are going to get everyone they know to click on it so they can be number one in Google, or someone complains about Google’s results not being relevant, and you launch into a rant about the lack of relevance in their search term. Everyone you know thinks you have some sort of magical power, or just thinks you work in IT, but either way there is a sense of smugness in knowing something most people don’t even realise is there to know

The accomplishment

There is nothing quite like the feeling of finally getting that position for a site that you’ve been working towards. Whether it’s a new project and you’ve made one instinctual change that ramps up the rankings, or a site that you have been working on for months, hitting that number one ranking, is always a high point, and seeing the traffic rolling in is the icing on the cake. In fact, you have to wonder whether the excitement is actually some sort of addictive adrenaline high, because it seems once you have worked that magic on one site, you just keep wanting more, and before you know it, you’re hooked.

SEO folks

I have never come across a bunch of people who, without ever meeting one another, can be so supportive. Even better is that when you do finally meet them, they are just as awesome in real life as they are on twitter, some even more so. I have asked many people, and I can’t find any other industries where competitors will help each other out so freely. I know for instance that I was up against another SEO chick on a recent pitch, did we stop talking, did we stop helping each other out, no, it’s a fact of SEO life that we will come up against each other time & time again, but we still have a huge sense of co-operation & camaraderie. In fact, I think the people in SEO top my list of the things I love about this industry, it wouldn’t be half as much fun without you all!



Internationalisation is Not Just Translation

Hello, my name is Judith Lewis and I am a searchoholic.  I live and breathe search and have done for pretty much over a decade.  It has become who and what I am more than any other aspect of myself.  It is easy to become a searchaholic in this industry with its own language, secret codes and instant gratification.  What we sometimes forget is that the world does not live and breathe search, a fact I was reminded of in the Pan European Search Strategies session at SAScon.

During a break between speakers, one of the attendees thanked the speakers for helping him to understand that he needed to target the Netherlands and Belgium differently.  While this seemed quite basic to me, it suddenly became clear to me that those who live in the Land of Search learn and understand localisation and targeting, do so slightly differently than those who do not.

As part of the naturalisation process when you immigrate to the Land of Search, you are taught about the global search landscape and how culture, language, platforms and other elements differ from country to country.  You are taught about Orkut and Brazil, Yandex and Russia, Baidu and China with its interesting politics and cultural norma.  You learn how a busy page works in Asia while a minimalist page works better in Europe.  You are taught that there are cultural, language and search engine preference differences globally and this becomes second nature.

Beyond this more obvious and easily measured uniqueness globally, is the understanding of how culture and regional politics factors in to everything.  Bas vandenBelt in his Pan-European search presentation mentioned how a predominantly orange coloured site in the Ukrane indicates a political affiliation while in Holland it represents a huge party celebrating the queen.  Andy Atkins Kruger pointed out how simply translating a site from English to other languages never works because of language usage and cultural differences.  These are all important points which are sometimes lost when creating a global web strategy.

Language is one of the things which makes each area of the world, each culture and each sub-group unique.  From the differences in Portuguese around the world, to Spanish, French and Chinese, each language will not only have its own nuances but culturally people will respond differently to the same sentiment expressed using slightly different words.  This is why translating a site is never enough – not only will translation leave a site unoptimised, it could also render it unreadable.

In Finland there are sometimes four or five ways to refer to the same thing.  Checking to make sure we captured the more popular forms of the words meant not only being aware of this, but understanding what the popular searches were.  What this shows is that a simple translation could end up using the wrong word, too formal a word, a word no one searches for or the right word purely by chance.  Localisation requires a local, expert hand.  Translation is not enough – understanding local language use is key.

In the Far East, not only is the use of language complex but the cultural and political issues must also be taken in to account when localising.  There are issues around what is said, how it is said and what the target audience is.  Even images which seem inoffensive or banal could be seen differently elsewhere in the world.  Understanding these cultural nuances when creating a site can mean the difference between success and expulsion of the company from the country.

Localisation and not translation is the key but also understanding the local market for each country you are going in to is important.  Barging in like a bull in a china shop will be more likely to alienate a potential market rather than win them over.

Paid Link Noi$e

In my world of link building, compensation is often offered. That is because I see nothing wrong with paying someone to put up a link for a client, and because I work with clients in exceptionally competitive fields where we’ll be the ones losing out if we try to do things the 100% white hat way. I only make those claims to avoid having this post get turned into a paid link debate…because this post isn’t about the ethics of paid links. It’s about link noise, and how that factors into a paid link campaign.

The key here is that when you’re buying links, you have a lot more control. You can set your desired anchor text, tell a webmaster which page to point to, where to put the link, that you’ll hunt him down and kill him if he slaps a no follow on it 3 weeks from now, etc. Thus, the concept of intentionally going after a bit of link noise may seem somewhat strange. You see link noise in most sites’ backlinks of course but you don’t always think about the role that it serves in a profile.

Before I go any further, I will define link noise as being the type of link that you’d exclude if you were picking out your best backlinks. The anchor text won’t be optimal, the sites may not be relevant, the links will just somehow not be optimal in any way. Now, the idea of approaching a webmaster and requesting a noisy backlink and offering decent money for it still makes me a bit queasy, because I’d prefer to pay for something…good.

You know how all link builders talk about a natural link profile? Link noise is part of that. My number one point with this post is that it is a critical part of paid link building campaigns, because you don’t typically think about offering cash to get a shit link. However, it’s those shit links that can sometimes save you.

So what IS a noisy link? It’s probably a link that, as a professional link builder spending money, you’d never ever want…and considering my feelings on outing people and/or pointing out actual stupidity on someone’s site, I’m not giving any examples to go along with this. Well, there IS the beaver down below…hahahaha.

beaver

The Acceptable Noise Links
These are the links that most link builders can agree make a profile look more natural. They’re the ones least likely to annoy and frustrate people who think that buying a link means you’re shaking the hand of Satan. These are the links with anchor text such as “Click Here”, “visit”, your URL, your site name, your company name, “more”, “here”, etc. Not that it can be argued that some of these are not, in fact, link noise. However, remember that I’m talking about the active, paid pursuit of links.

Incorrect links
These are links that are misspelled, go to the wrong page for the anchor text, have incorrect URL but go to the right place (like “seochicks.com” which does go to our site properly but, um, we have a dash…a dash of sass! ) have a space where one should not be, etc. These are links that you might expect a newbie link builder to set up with an inexperienced webmaster.

Missing Anchor Text
These are the links with no anchor text. Need I say any more?

404 Links
These links go to a non-existent page on your site. Again…not much explanation is needed here. However, the very IDEA of getting someone to put up a link to a non-existent page on your site is just so counterintuitive that it makes my head spin. The other way, of course, but it still spins.

it ain't thar no more

The Pesky Second Link
These are the links that aren’t supposed to count, provided that the landing page for both links is the exact same. Some people try to be cool and get in two anchor text combinations, thinking both will count. Some webmasters think that they’re doing us a favor (thanks assholes) by giving us a free second link. In any case, these are excellent for link noise.

Just Plain Stupid Links
These are links that contain a paragraph in their anchor text. They are the links that you’d expect someone who’s a bit of a moron to give you, not knowing any better. I’d put hidden links in this category too, actually, as those really are just plain stupid. However, I do draw the line at specifically requesting hidden links.

idiocy, pure and simple, with stars, stripes, and a gun

Now, with this in mind I spent a bit of time analyzing the backlinks of this lovely site. I know, without any doubt whatsoever, that the SEO Chicks backlink profile is completely 100% organic. We’ve seriously done, well, fuck all to build links. To do this, I used Raven SEO’s backlink tool, which is my current favorite, and I exported, sorted, and did everything but produce a pretty graph that makes it look like I’ve done some work when I haven’t. I didn’t look for the second links or 404 links, mainly because I just wanted a quick and dirty analysis and didn’t feel like digging. Cheese dip awaited.

Here’s what I found:

64% of our backlinks are what I would consider link noise. Out of this 64%, 1.4% have missing anchor text, .6% are incorrect, but only 1 link (0.2%) was just plain stupid. Yes guys, I found the beaver link, and thanks for that. The most interesting thing is that a full 61% of our links are of the acceptable link noise variety. A good bit of these are site names and URL mentions, which, as I have said, could arguably NOT be link noise but hey, we’ve never tried to optimize for our own name so I feel ok putting them into the link noise camp. Before any of you mathletes gets testy about the math, note that I’ve rounded. Sometimes up, sometimes down. It just depends on my mood, really.

Therefore, 46% of our backlinks are something for which, as a link builder, I’d be trying to optimize, rank, and get traffic. Now, if I were paying for links, I might not think that spending money on noise was a good thing of course, but as you can see…noise is 64% of this organic profile. I am certainly not proposing that you invest in 64% link noise, of course, but I don’t think that there’s any way your link profile will appear natural if you ignore those kinds of links.

Interview With Tami Dalley: SES London 2010

In the final of our SES London 2010 interviews, Tami Dalley of Acronym Media gives us some great in-depth advice about multi-layered approaches to data analysis; particularly useful for difficult, non-comparable and small data-sets.

« Older Entries
Newer Entries »