The problem with Omniture Sitecatalyst
I use sitecatalyst on a regular basis and what frustrates me the most is just how much of a handicapped tool it is for real world analytics. Sitecatalyst gives the impression that it will be useful but then disapoints so easily. A Omniture staff member once told me that the Omniture consultancy team do not use Sitecatalyst, they have access to and pretty much exclusivley use the very expensive Discover tool. It also appears that Omniture use Sitecatalyst as an upsell platform for their Discover tool. So I thought I would share with you some of the common problems we encounter in Sitecatalyst. Some of these points are feature requests, interface changes and bugs but they all have one thing in common the ability to infuriate the user.
1) Sitecatalyst cannot display any segment data what so ever! Take a look at Google analytics and I can define and show up to 4 segments on a standard report it takes me 10 seconds to set this up. Sitecatalyst has datawarehouse feature which is itself a chargable addition to the standard package. Now datawarehouse allows you to build a segment then wait a few hours for an excel sheet to come back with some data. Then if I want to see the same data for a different segment I can repeat the process. There’s no way for me to compare this data in Sitecatalyst this is all done externally by pull this data together by hand.
2) Excel client can’t deal with Segments so I can’t automate pulling multiple segments together in an excel sheet in point 1.
3) The boardroom ready reports are not ready for the boardroom. Try building a big dashboard and downloading a PDF version of that file. The formatting will be slightly strange but also some headers will overlap and become un-readable.
4) At least in version 14 where an evar has a none value and is associated with a product there is no way to remove the none value and have the totals in the product adjust accordingly. We can do a filter does not contain none but the numbers don’t change. It may be a bug but its been there for 4 point releases of version 14 now.
5) Publishing lists are useless, on release of version 14 a big deal was made around this functionality which should allow you to distribute the same master dashboard for different report suites with out having to re-build the dashboard. It does not work and has not been fixed in 4 point releases of version 14.
6) Why do I have to create a default metric for bounce rate and other common metrics? Why can’t we put this metric against useful items such as search keywords, referrer types etc etc
7) Why can’t I trend on useful items such as referrer types! Can do this in excel client but why should I?
Percentage change alerting is almost useful but when dealing with percentage changes across a large amount of values I get some text in a e-mail in no sort order either by the value or the percentage change. There’s no option for an csv file just this plain text e-mail which is not useful.
9) Monthly forecast has gone in version 14
10) Contextual performance against last month and year is only available on some report types and some graphs
11) We can do stacked graphs but no 100% stacked graphs (these are very useful when reporting to clients and you want to provide comparitive performance without providing numbers).
12) The only way to report on google news performance is to look at all referrer url’s and then do a filter for news.google why? referrer is an exceptionally long report to run.
13) Why does the excel client time out when pulling larger and more usful data set. E.g I can’t event pull in top 50 search terms month on month in excel with out time outs.
14) Having multiple data blocks on an excel worksheet causes all data blocks to fail.
15) Why do I need to set-up a plug-in to establish if a visitor is new or returning.
16) Just when you need it Sitecatalyst reports never have a total colum.
17) The video reporting integration may not upload a persons viewing information until they visit the site for a second time. What happens if they dont no data for that video play.
I can’t agree with you more! If I get chance, I probably can write a similar list as well.
But don’t be deceived by the marketing claims of Discover. It is not as good as it is said. You’ll have the same frustration with it if not more.
Regards,
Ed
Knowing that no analytics tool is perfect, which would be your preference if not SiteCatalyst?
@Elysa
Elysa I must admit that I have little experience of webtrends or coremetrics so I am not sure what these products have to offer particularly in response to the issues highlighted in this post. Omniture is recognised as a leading analytics package and it does have a lot going for it.
Amen to that. Imagine these problems compounded by sitting in an international subsidiary where the SC interface is achingly slow and the data warehouse reports take almost a week to run.
My pet peeve is SC’s inability to segment bounce rate by campaign or referring domain.
I am sorry to hear that your SiteCatalyst experience has been disappointing in some areas. And, while I am pleased to see your comment noting that SiteCatalyst “does have a lot going for it,” please be assured that we take the feedback you’ve given in this post very seriously.
It seems to me that a number of your concerns center on segmentation. It’s true that the most powerful segmentation capabilities are provided by Discover, Data Warehouse, and ASI. However, many users are able to implement SiteCatalyst in such a way that they can perform some segmentation within SiteCatalyst. Fully subrelated eVars are particularly useful in this regard, since they can be broken down by any other eVar, products, campaigns, and traffic sources data.
Some of the issues you’ve reported are certainly not based on expected behavior, and should likely be addressed by opening a ticket with our ClientCare team. I’ve had very different experiences with a number of features you mention specifically. This leads me to believe that you’re encountering some issues specific to your account, at least in some of these areas. Our support group will be able to help address these. I’d be happy to discuss further by e-mail. Feel free to contact me at omniturecare@omniture.com; I will very much appreciate it if you do follow up with me.
Thanks again for your feedback.
Ben Gaines
Omniture, Inc.
Update : This post seems to be getting some traction and was written some time ago, since then have been a number of point releases to sitecatalyst. So I can update some of my experiences against these points.
3) Boardroom reports. Dashboard reporting and e-mailing has improved in recent releases. The HTML formatted e-mail is particularly effective.
5) Publishing lists are now properly integrated with dashboards and workbooks so thanks for fixing that.
10) I think contextual graphing is available on all standard graphs now in the latest release.
13) The excellent client has been updated which has improved this issue.
Also in a general note to anyone reading this blog the standard user support from Omniture is generally excellent, your efforts on twitter, blog.omniture.com and live support etc. Really make a difference. I am sure that some of the above gripes are resolved with custom integrations/implementation strategies, but it can be frustrating that relatively simple solutions are not integrated by default.
How funny that an old post like this all of the sudden gets this kind of attention
But now that we have the attention perhaps Omniture can respond to one additional question?
Is it possible to get bounce rate figures for any of the following reports?
1. Search keywords.
2. Referring domains (inc. direct traffic).
3. Products.
@Elysa
Preference should be driven very much by your requirements. It’s worth checking out any of these web analytics books
Tell me this, why is it that customers have to know the right questions to ask to get full utilization out of their implementation? How about a little more proactive and a little less reactive?
Will my thoughts as well !! How the heck do you know if something is not working right or it’s just bad product design or a product flaw. Unless it is clear as to what benefits are obtained then it’s very hard to know if you are getting the most out of your product. That goes with most products btw.