Coremetrics Doesn’t Suck

I was chatting with some colleagues about the recent IBM acquisition of Tealeaf and how IBM has really become a player in the Analytics market over the past two years, buying Unica, Coremetrics and now Tealeaf (and about a dozen other smaller analytics related companies in between). In the midst of the conversation, there was a comment made that stunned me: “Too bad Coremetrics sucks”. I responded, “Really? Why do you say that, what don’t you like about it?” This person responded back with, “Oh, I’ve never used it, I just heard it sucks.” I then spent ten minutes talking through some of the pros and cons of Coremetrics vs. Omniture and GA. Since I’ve used Omniture, Tealeaf and GA for the past year and used Coremetrics for five years prior to that.

I then had several conversations at a recent Seattle Web Analytics Wednesday, which further bolstered my opinion and at the same time helped give me further information about the tools that I have less experience with. (My conversations were with several Omniture power users and I trust their opinion on the tool)

I’ll state this before I go forward: I’m not paid by ANY analytics company so my opinion is only based on usage. None have asked me for my opinion.

My overall simplified opinion is this: If your goal is a large data set with multiple metrics spanning multiple categories (on the same level), Coremetrics core reporting suite is perfect for this. If you are looking for easily segmented advanced or company specific data, properly implemented Omniture code will give you easily accessible reports. If you are looking for an out of the box tool that is easy to navigate and you have a smallish data volume, and aren’t looking for a ton of advanced or company specific segmentation, then GA is a great tool for you. Tealeaf is not a great analytics tool, however it is a great monitoring tool, specifically for things like page latency and one off events such as error messages.

Now I’ll dive a little deeper into each tool:

Google Analytics is a great out of the box tool. Most small and medium sized business (by Internet traffic volume) can use this tool and not need any additional analytics tools.  Google’s UI is its best feature, it’s very fast while being very rich in visualization. You have the ability to quickly segment reports and can toggle between default reports and created dashboards very efficiently. It’s biggest flaw may be a lack of depth to the reporting. For example, you can only select three segments to trend in a report and you can only set ten unique metrics in the dashboard view. While I also like the quick left nav interface between report sections, however, I feel that the tabbed reports found in Coremetrics would be a great addition.

Tealeaf really doesn’t even belong in this conversation since I don’t really even view it as a reporting tool. However, Tealeaf’s greatest asset is its highly customized metric capabilities. However, this is also Tealeaf’s greatest issue. While there are some default metrics, most of Tealeaf’s unique value is based upon the events that the event admins create. Unfortunately, most of these events are simply count events and it’s very difficult to extract averages, etc without using Excel. Thus, Tealeaf for me, is a great monitoring tool but not a great analytics tool.

Coremetrics’ tabbed reporting is its best feature, though the Explore reports are a close second (the limited number of reports makes this number two and something I hope they change if they have not already). Again, for full data sets, Coremetrics is set up by default to provide this. However, where Coremetrics really falls short is a lack of segmentation options in their main reporting suite (and as of my last use, the Explore report segmentation did not include a full data set). In my experience, Coremetrics report load response times were very poor, but I’ve heard from other clients that this was not a big issue. Finally, my biggest complaint with Coremetrics was that marketing data is largely siloed from the customer interaction data, so it’s hard to tie the two together.

Omniture’s ability to quickly create new reports from pages in the current report is its best feature in addition to my previous statement about its segmentation capability. The ability to quickly add to a dashboard is also a great feature. As with Coremetrics, there’s a limitation around tying the marketing data in with the interaction data, but some of this can be solved by your variable segmentation if you choose to implement this way. As with GA, I think Omniture could benefit from having the multi tab feature that Coremetrics has so that you can more easily toggle between reports without closing a previous report, however I do think that Omnitures navigation is a bit superior to Google.

My final thoughts are that people should not throw out blind opinions. Especially in the data world. If you have no facts you cannot be an analyst. So, before you make a statement like, “XYZ Sucks” do some research, and learn for yourself before you come to that conclusion. I honestly don’t think Coremetrics is an overall inferior product to Omniture. I think that it has some limitations where Omniture beats it, but it also has some functionality and features that make it better that Omni in certain ways. I definitely think that GA should be part of the conversation as a solid product and I wish that I could get the chance to use Webtrends because I’d love to compare that product to these others.

At the end of the day, if you want to move up in the Web Analytics community, you cannot limit yourself to closed minded thinking. You can’t limit yourself to just using one product. Omniture does not dominate the category, contrary to opinion. Look at the facts, and you’ll find that Coremetrics has as many quality named clients as Omniture, and the products each have great qualities that will help you make sound and actionable insights for your company or client.

If you have used more than one of the above tools (Or Webtrends for that matter), what are your thoughts and preferences?

6 comments

  1. Tim Wilson says:

    Great post! I actually think the ” sucks” statements more often come from people who *have* used the tool, which then makes them feel comfortable passing judgment. The kicker is that these are people who have deep experience in one web analytics tool and then, for whatever reason, found them using a different one. They find themselves bumbling around in the tool and trying to understand it, and it seems “less intuitive” than the tool they really know well. It’s like a native English speaker spending two weeks in France and concluding, “French sucks.”

    All tools are complicated, and they all operate with different paradigms — both in their data structure and their interface. Having been super-fluent with Webtrends in the past (but I’m rusty), and now fluent in both GA and Sitecatalyst, I *recognize* that I bumble around in Coremetrics because I simply haven’t learned the language yet.

    I can sing the praises and bash GA, Sitecatalyst, or Webtrends. But, I’ve started biting my tongue — analysts tend to treat “their” tool as something of a sacred cow, so it’s tough to have a rational discussion.

  2. jeff says:

    Great point Tim. I agree that often “XYZ Sucks” comes from a one-time user who has been using the alternative for many years. I was actually in that boat with Omniture at first, after using Coremetrics for 5 years and not having a very good training program to introduce me to Omniture. I spent 6 months in Omni learning on my own and thought that it wasn’t a very good tool, only to find out… I was using it wrong! In the last 6 months, I’ve come to appreciate Omniture and it’s functionality. And yes, I agree that analysts tend to treat “their” own tool as a sacred cow… or in the case of everyone’s “favorite guru”, Avinash, he’s become fairly blinded to any other tool due to his ties to GA, and thus I feel that he’s lost some credibility as a “thought leader” in regard to which tool to use. Oops, yeah, I went there. Thanks for your comment!

  3. Antoine says:

    Interesting article and as a French person living in Canada, I like @Tim’s example on France (Damn those French not capable of speaking english properly …)
    In Canada, we face the same reaction against all paid solutions, as companies started adding a GA code and now use only GA, then cancel their coremetrics or omniture’s license (those companies being guilty for not having local offices and sales teams to train and to follow their clients)
    et Voila !!!

  4. While not disclosing the purchase price, IBM says it has invested $3 billion in its Smarter Commerce initiative designed to address client companies’ technology needs in the digital age. Its investments include its $440 million acquisition of DemandTec , which analyzes consumer buying behavior, announced in December 2011; the 2010 deal to buy web analytics vendor Coremetrics for an undisclosed price, and the August 2010 purchase of online marketing firm Unica for $480 million.

  5. gold price says:

    My overall simplified opinion is this: If your goal is a large data set with multiple metrics spanning multiple categories (on the same level), Coremetrics core reporting suite is perfect for this. If you are looking for easily segmented advanced or company specific data, properly implemented Omniture code will give you easily accessible reports. If you are looking for an out of the box tool that is easy to navigate and you have a smallish data volume, and aren’t looking for a ton of advanced or company specific segmentation, then GA is a great tool for you. Tealeaf is not a great analytics tool, however it is a great monitoring tool, specifically for things like page latency and one off events such as error messages.

  6. ZenoArrow says:

    I’m sorry, but as someone who has used Coremetrics, I am in a good position to tell you that to understand why it sucks, you need to look deeper. Segmentation is certainly an issue. Let’s imagine you want a report showing mobile usage in a specific country. Sounds like a basic report right? If you try to create it in Analytics you need to have created a segment BEFORE you wanted to report on it, and happened to have assigned it to a suitable report for your needs (bear in mind there are a limited number of segments available). You could then use Explore, but you come up against limitations there too (good luck getting YoY data, Explore defaults to collecting 400 days only). The big problem at the root of many problems is how Coremetrics stores its data. Immutable data structures have their place, but if you want a useful reporting tool this should be balanced with dynamic views. If anyone reading this is thinking of using Core metrics, my advice? Don’t.

Leave a Reply

Your email address will not be published. Required fields are marked *