Two weeks ago, I spent 90 minutes on the phone (a lifetime for me) with Epiar's Ken Jurina and Curtis Dueck (they're in a video with WebProNews here). They walked me through their propietary and incredibly advanced, internal SEO software, explaining how they're able to provide an exceptional quality of results for their clients. I like to think that SEOmoz gives its clients excellent service, and from a strategic and consulting perspective, I believe that's true, but Ken's team at Epiar has set the bar so high, I've never seen its equal in any of the many companies I've worked with.
To be quite honest, I struggled with writing this blog entry. During Ken's presentation (which felt like the unveiling of the Ark), I took almost 50 screenshots of their software, with examples of exactly how they conduct their processes for specific clients. After thinking long and hard, though, I emailed Ken. Here's the text of what I sent him:
I've got a bunch of screen caps and had some ideas for a post, but I'm scrapping them. I think it would honestly hurt you more than it would help, and almost anything I write about what Epiar does on the backend will bring you competition and attention (the kind you don't want) more than positive press or clients.
Let me sketch out a rough idea of what they're doing so you can understand why I was so impressed:
-
Automated keyword research (but very, very deep) - it pulls thousands of related terms, synonyms, words frequently found on high ranking pages for the query, terms suggested by tons of keyword tools, etc. and refines and sorts these using an intelligence that shocked me, then pulls search frequency and competitiveness data (much like our KW Difficulty tool). What's left is a keyword list that looks like a human painstakingly edited and researched each variation and term individually, but was in fact created by machine.
-
Automated search friendliness - determining where targeting is missing on a site, where it needs to go, search spidering issues, etc.
-
Automated link discovery - finding hundreds of sites relevant to the KW research terms that have pages with forms or contact emails or ad opportunities or URL submission areas. I never thought link sourcing could be automated with this level of precision, and while it still isn't as good as a human's touch, the depth and breadth allowed for saves hundreds of man-hours per client.
-
Competitive & industry-segment-specific research - how often is a keyword phrase targeted in a specific geography, to a specific group, in a country, etc. The answers to how much competition exists and how tough are those other sites, plus loads of other search query and industry data from a myriad of sources so vast, there were even a few I hadn't previously considered.
All these systems and much, much more is available through the tools that Ken & Co. have built.
To say that I was impressed would be an understatement. I was very frankly shocked. Obviously, I feel that SEOmoz can provide a very high level of service and value to our clients, but we've always been limited by scale, as have many other firms offering organic optimization companies. Epiar has broken new ground in my opinion, by using highly customized software (and bucket loads of API data grabbing + scraping) to refine the most time-consuming of SEO tasks. While they're not competing in social media marketing or linkbaiting and have even footing in terms of strategic consulting, I think it would be foolish of me, after seeing their processes, to assume that we could bring the same level of service (or detail) to a large scale keyword research or link discovery project. We're simply out-gunned.
Perhaps this is the direction that all SEO work must eventually go down - high levels of data gathering, data refinement and intelligent processing combined with software architects with an exceptional grasp of SEO and the time to fine-tune their code to produce exceptional results on a consistent basis.
OK Rand, that's just cruel. You basically just dangled a piece of prime rib in from of a starving kids face. You know, like the ones you see in the feed the children commercials. Then you said, oops, you can't have that.
Sounds like they let you take screenshots and you had permission to publish them and then you said nah. Your killing me buddy. Don't tease me, give me the meat.
Now there's some premium section content for you :D
Heh. Yeah - knowing how cool their tool is, that's one thing, but I'm desperate for a look. Perfect paid-up member content ;)
Jeremy - I think it would have been much crueler to abuse my position and blog openly about exactly what they're doing. Ken wanted to share and I love that - it's my guiding mantra - "share everything, share as much as possible, keep it open."
That said, once I saw it, I knew that it wouldn't be fair to him or his company to put it all on display. As someone else noted - this already opens the Pandora's Box; no need to show off how the gears turn.
As much as I'd like to see those screen shots I know you're right. The little you did reveal has already opened the Pandora's Box like you say and is generating ideas at this very moment. Still I'd love to see the images and even more give the tools a spin.
Ken's keyword analysis and consultancy tool is, without a doubt, a mind blowing experience. I had the pleasure of interviewing Ken, and hanging out with him a bit at SES NY. The performance level of Epiar's SEO software is staggering as well.
If I may, I'll post my interview with him here.
rumblepup: An Interview with Ken Jurina
Entertaining. Thanks for sharing.
I would love to see that in action. We have recently been discussing the level of automation possible in PPC management (the top software in that sector is pretty impressive too). I haven't put as much thought into what would be possible SEO-wise.
We struggle (and I'm sure you do) with the man-hour investment of creating software (even with a known payoff of time further down the line) when it's weighed against immediate consulting work. This kind of thing shows that there are huge rewards out there if you have the balls to go ahead and develop that software.
We are thinking long and hard at the moment on the subject of how big our team should be and how fast we should grow. If we are going to grow, then building assets like this has to be part of the plan.
One key question coming to mind is "how long before you can license the software (or something like it)?". The thing with software is that someone else should be able to do a similar thing - I can't see this kind of tool always staying in-house.
"One key question coming to mind is "how long before you can license the software (or something like it)?"."
Right on willcritchlow and more to the point, are these guys willing to license it any time soon? ;)
Rand, thanks for sharing!
Have you actually taken the software for a spin on optimizing an actual or even a dummy site? Scale is one thing, but I'm interested in quality of the keyword reserach and does the tool do well on a "specific per page" level?
If many websites do turn more UGC, then page numbers will inflate and we'll be seeing a more rushed need to automate alot of the processes and the creation of deliverables. Strategy, of course, wont be overtaken - but research and execution maybe.
I actually got to hang out with Ken, Damian, and Curtis quite a bit at SES and I also use WebCEO pro- unleashed... after asking them about WebCEO and their 'bleeding edge technology' well... to put it into perspective... KEI is great for low hanging fruit... What the guys at Epiar have put together is like Cheating! It's soooo far advanced, when compared to any other Keyword tool it's not even funny!
Another great way to find great keywords for your sites (client sites or your personal site) check your log files to see who / what / how visitors found your site! Those log files are filled w/ gold!
.02
NeO
I think the huge benefit is having the system being able to automate so much of the typical SEO work. Tools like WEB CEO are great for automating the link research, rankings, and keywords, and using them to get information. I don't generally like to use the reports and advice from those tools, though. It seems like the tools Epiar has are more about automating, collecting, and reporting data and then its on you to analyze it and come up with your own conclusions.
That's the kind of system I'd pay for, one that focuses on the tasks that take so much time and less on offering advice - if everyone gets the same advice, you don't get much of a benefit.
I've bought WebCeo 2+ years ago, using that for their ranking features mainly, but their keyword research features sucks compared to specialized tools - not to mention "best-of-breed" tools that are built inhouse by SEOs...
I doubt anyone really making good money with sophisticated keyword research would sell it for these cheesy $450 that webceo costs
Hi Cemper,
I agree, the keyword research feature is weak when compared with other standalone tools......appologies for the affiliate link earlier....:-)
great post rand!
I think this shows the always going on need for improvement of "organic analytics" in a broader sense (what else than number crunching is keyword and link research but narrowing down huge result sets based on numbers?)
@will: I hardly doubt thatthat software wil get licensed... I wouldn't license it if I had it, just because it is a competitive advantage for Ken and Epiar... frankly, maybe I'm already using similar tools and won't even tell you? who knows? :-)
I agree - they have a nice niche going with doing that - I suspect, though, that the tools that are available to license are going to continue improving (though of course, by the time they reach this level, these guys could be even further on).
Maybe I am, and this is all to throw you off the scent ;)
Rand,
I wanted to sincerely thank you for taking the time to review Epiar’s SEO applications and for your complimentary blog post today.
You've certainly help validate to us the level of innovation we have in our own applications which we tend to forget sometimes. Though we continually spend numerous R&D hours to refine and advance our apps, you can easily get complacent in the software's abilities when you use them on a daily basis.
Not until you show them off to someone, such as yourself (who we truly feel is an authority in the industry - someone we look to as an industry leader themselves, and personally trust), do we then ourselves open our eyes up to what we really have.
It was also nice to read comments from people who we've had the good fortune of working with or had the pleasure of meeting in person at the numerous SEO/SEM conferences we've presented at.
Internally we have had endless hours of debate on possibly licensing our technology. While that still is a possibility, the reality is that it takes us 2 months to train our staff on the 3+ applications we've developed and all of their features - which still get added to on a monthly basis.
Who knows?; there might be a strategic opportunity in the future to work with a company to help us in taking our apps public? Until that time our clients get the benefit of what we've developed thus far, and they certainly don't seem to be complaining. :)
See you at SES Toronto and SMX, Rand, and take care in the meantime.Ken JurinaPresident, Epiar Inc.
*reels*
Wow... that's something else!
Makes me want to retire now, move back home and get a job with you for the sheer joy of working with such fascinating toys!
Hi Rand. I just wanted to say that it was brave and very honest of you to be so open about another SEO firm.
Of course, we're all intrigued about what this software is and how we can get a piece of it, but I'm more impressed in this piece about your open approach to a potential competitior. That's one of the things I love about SEOmoz!
Yup, I was just thinking lately that all my job as an Internet marketer (not juse SEO) is to process information.
Nothing more, nothing less. Then again, machinery will always lag
behind humans ability to estimate things. Then again, some tools can
help, I guess.
P.S. Why doesn't it just break lines automatically in the editor?
I was actually just hearing about how much revenue in clients your turn away to other people. That's pretty cool.
As far as this automation, I still prefer to do it "by hand" somewhat, because I want to get more solid in my SEO skills. I hope that I can bring in more clients to justify the cost of installing that kind of automation in to my business.
After reading this entry and the responses, I am really glad. Why? Because we have a very similar tool we've been working for the last 3 years. We use it for our own projects as we don't do consulting work. We optimize our own sites in highly profitable niches.
We plan to license the software on a subscription basis.
We plan to launch the public beta next month. We are not ready to disclose the details but our tool automates: search engine friendliness, keyword discovery, niche opportunity identification, keyword organization based on visitor intention and assigning keywords to the right landing pages to improve conversions.
We also have advanced competitive intelligence, on-page metrics editor and advanced link analysis.
One key difference with Webceo and others is that all tools work as one in our software -- as a single step by step process, avoiding having to enter the same information more than once.
Rand, If you have the time I will like to send you our latest build to get your input.
Wow.
Having worked for Epiar in the past, I've had the pleasure of seeing their MarketView and AccuLink programs in action. Pure genius. I thought I knew a lot abot KW research but man did I learn a lot working there. Ken, Bob, Curtis, James, Matthew and Damian are a great group of guy's too.
This company is very worthy of a SEOmoz blog post ;)
I've been very interested in their offerings for quite some time now...If I could only find the $$ to try them out....
I second this
It's always good to know about new tendencies and technologies in the SEO/SEM world, that kind of give us ideas of where this is going to and how to conduct our own SEO strategies, thanks Rand.
I agree with Will + Emergence -- it sounds like a great tool, but there was no way to take it for a test drive on their site...which might make it seem like just another gimmick for people who hadn't read your review, Rand?
Licensing or spreading this tool through some kind of subscription service might be the way to go for them?
David - this is an internal tool, not something they plan to open to the public. It's to help them with client work, so showing it off, as I mentioned above, seems like a competitive disadvantage. That said, realizing the extent to which they've automated, and the thousand+ hours they've poured into development of the software makes me not just impressed but a wee bit jealous :)
Hey Rand, are we allowed to publish WebCeo affiliate links in here?
Seriously. If that's not comment spam, I don't know what is....
Rand,
It doesn't matter if they open it to the public or not. Once the technology exists to invent a light bulb, and the idea exists that a light bulb can be invented, and the need for light bulbs exists, if Bell doesn't invent it someone else will. The only question is will this new technology be most effectively exploited/monetized by keeping it private or taking it public. My guess is that the first company to effectively take a similar system public will become the next big thing in the SEO toolbox. But, then again what do I know?
I tend to agree. They have spent a great deal of time and money making something that the SEO community is interested in, and which could earn them a substantial amount of money if they licenced it.
As said above, proprietary software is great, until the next guy develops something better. they're ahead of the game, they should take advantage of it. Let them use it for their clients, to their clients benefit, and let them licence it to other SEOs who can afford it and are willing to purchase it. They can set the bar high to appeal only to actual SEO firms, or inhouse SEOs of large corporations etc.
Its a great business opportunity they have, and licencing it with some decent PR should not only generate money, but also massively impact on their market share and increase awareness of their brand significantly.
They've done the work, they should now make it work for them. my two-bits...
I see, thanks Rand. I was thinking of it more along the lines of an SEOmoz "Premium Tool" kind of thing. It definitely would be a competitive disadvantage to show it off if it's proprietary!! It certainly does seem to be very impressive based on what you describe. The automation for KW research and link discovery in particular makes me drool a little bit...
I can assume that you have a lot of new ideas on the SEOMoz tools? Obvioulsy there is a limit to what you can add (even for premium menbers) but, did you have any take aways that you plan or would like to implement into the SEOMoz Tools?
Thanks for the post,
It looks like Epiar are using KEI to determine the best keyword to target, as Shahid pointed out and I have confirmed in Epiar’s website.
I think there are some problems you need to take into consideration when you are using the KEI (keyword effectiveness index).
An indicator like KEI is only as good as its source. KEI uses the number of pages in the search engines index and the number of searchers (Unless Epiar using a different kind of KEI).
There are several problems with the accuracy of the number of pages. I could write several pages, but here are the most important:
Google and Yahoo don’t give an accurate number of pages. MSN might give an accurate number of pages in their index.
There is probably less competition for the phrase with one keyword than the keywords phrases that consisted of 3 keywords with the same number of web pages, because more words makes it more unlikely to find the keywords next to each other (if you use the number of pages in quote).
The number of searchers is just not reliable.
Conclusion: KEI consists of two variables that aren’t reliable. If you are optimistic you hope the elements of uncertainty neutralize each other, if you are realistic you interpreting the result taking the keyword phrases into account.
Is anyone using KEI to choose keyword phrases?
It is a mystery to me how you can have an advance keyword selection system based on KEI. Maybe someone can help explain it to me?
I have only seen two efficient methods used by keyword tools to help sort keywords:
Filtering the keyword list in different ways.
Example: Take a search database and filter all keywords out that include one specific keyword. Like “loan”. You then get a list of keyword phrases that include loan. The problem is that you don’t get the keyword phrases that include loan, but doesn’t use the word loan, as “borrow money in bank”.
The other method is to choose a database that has low number of irrelevant keywords.
Example: Like collecting meta tags from relevant sites. The problem here is that you are collecting your data from the supply side. It is therefore known keywords, there is a supply for the keywords and you don’t know if there is a demand for the same keywords.
I like automation of keyword research, but there is always a cost. So in the end it is a cost benefit decision.
NB: I plan to analyze every variable for its pro and cons, based on several big keyword reports in my blog.
Howdy Folks,
I thought I'd chime in as I've seen a couple of people mentioning KEI in their posts. I thought I should clear up that Epiar really doesn't use KEI to any extent nor have we ever. It's one of those things we had in there many years ago and never really took out.
Since we use our applications only internally we didn't much care if it was there or not, but now that others are looking at what we've got I can see how it could raise an eyebrow.
The version of our MarketView application that Rand saw was a bit dated - we do updates monthly. We have been working on our own algorithm for some time to calculate keyword difficulty as well as to uncover opportunities amongst hundreds of thousands of phrases with little to no human intervention. It's just another way to analyze the data.
We call it "competition index" and it is now officially used in our apps and replaced the old spot where the old KEI use to be. Unfortunately it wasn't quite ready for Rand to see when we gave him the tour a couple of weeks back.
Lastly, we data mine our keyword research from multiple sources, versus only one, to factor out any abnormalities.
I have just writhen an article to explaine more about Keyword Effectiveness Index (KEI).
I hope it will clear things up a bit.
Hi Rand,
After watching the video, it seems they have used the KEI index as a factor for establishing new keyword phrases and niches. Don't get me wrong, I'm sure they have alot more going on also, but surely when carrying out keyword research, KEI should be a strong identifier to establish how much competition and possible searches there will be for a particular keyword or phrase.
Maybe i'm abit jealous too....:-)
Heh heh, when Ken walked me through the tool, I asked him about that, too. Apparently, while it is in there, they're getting smarter about the competitiveness levels and using some more advanced searches to get that data. However, whenever you use Wordtracker's API, you're at least going to have that data piece :)
Hi Rand,
I'd love to see the software in action, sounds like so many individual SEO tools we use on the web, rolled into 1. Obviously there are quite a few tools available, some better than others. I like to use WebCEO, a very powerful SEO tool. To put it in comparison, I paid approx $479 for the WebCEO licence and WebCEO Professional Exam.
One thing to remember for all you SEO newbees, anyone can use an SEO tool/software, but if you dont have the underlying SEO knowledge, it can be difficult to interpret what to do with the data presented back from these tools.
Thanks
Shahid.SEO
Nice plug on WebCEO. Tracking URL and all ;-)