I think your approach seems useful and I intend to try it out.

thanks

1. Your latest post you mention that Visual Sciences is out there by itself. I’m wondering whether you’d class SAS eBIS tool in the same light.

I have not had enough experience with SAS eBIS, the last time I saw it was a couple of years ago and I am sure it is not the same anymore. But based on the presentation from SAS at the Emetrics summit panel it seems that SAS eBIS might be a compelling alternative (and I was surprised that SAS was honest about the true cost and kinds of clients who would use the tool, that honesty is refreshing).

Most analytics tools that need heavy $$$ investment are used by big companies (say more than 7 - 10k employees). These same companies are full of “business or functional” units and loads of teams, layers, process, bureaucracy. Expensive tools (be they Visual Sciences, SAS or SPSS or…) needs superheros to pull of any insights because of the sheer intelligence and complexity (or the data, tool, business). Usually there aren’t a lot of superheros around.

So I have this simple “Expensive Analytics Tool Cultural Test“:

1. In your big company is the cultural environment such that a few (one, two or three) people can do all the analysis and find the insights and that is sufficient to drive major changes to your webiste / business / organization?

2. Or your big company cultural environment is such that loads of people / analysts need to have access to data and rather than massive global decisions made by a few, change happens by numerous local decisions made by the many?

The choice above dictates what tool you need (sort of in the vein of what Rene’s comment above).

I am not making a judgment call on which is better. I am encouraging a strong and self critical review of the organization and then finding a fit.

Analytics tools rarely fail. We let them down because we don’t make the right choice for our companies.

2. Our mutual friend (and your fellow director of Web Analytics Association (congratulations by the way)), Neil Mason, did a fantastic slide at eMetrics

I could not locate the slide you mention in Neil’s deck but I’ll ask him when I see him next. Neil is a leading authority in our space so it is unsurprising that he is full of fantastic slides! :)

3. With regards to the statement about “taking it for a spin”. I’m very aware of the exorbitant costs of running a proof-of-concept (POC) and the pressures that sales people put firms

This I completely agree with, especially for traditionally huge companies. More than the work to be done it is the process and people convincing hurdles that have to be overcome.

But at the same time I don\’t know what the options are. Most people select the wrong tool because they do the selection process wrong. My thought is that if you want to know what you need then you have to put in the effort. But having been in a few huge companies I know how really really hard this is.

Thanks again for the comment Miles, it enriches the discussion and is very welcome.

-Avinash.

Great practical post for your readers, thanks.

I like your segmentation and I think that you are right in advising to test one of each. Testing seems to be the ’silver bullet’ lately ;-)

Eric, if you allow me, as manager of a few of those ‘expensive and potentially biased consultants’ ;-):

I would like to say that a good consultant will also take into consideration the needs of the client (and if he’s a good one, not only from a technical point of view). At OX2 when helping a client (and I say help, not taking the decision on their behalf) we always start by discussing the needs and this sometimes leads to a workshop or a bunch of interviews. IMHO with the complexity of Web Analytics, before engaging hundreds of thousands of dollars/euros, it is wise to be helped by someone who has seen lots of different projects and clients in order to avoid the most common pitfalls and try to engage the Journey the right path. And in the end it’s a question of testing before buying. Nevertheless I think that the report will be very valuable for users already familiar with WA, but I don’t think that someone that doesn’t know a thing about WA should take the decision only based on a written report. Disclaimer: Eric, I love you too ;-)

Avinash, if I may, I would like to add a couple of variables in the selection process you describe:

First is to see if there’s not yet an existing product somewhere in the company. For example, IT guys sometimes have a WA product somewhere in the basement. This can sometimes avoid expensive licence fees (I’ve seen a few examples). And you might even will find out that your IT guys can provide you with technical support because one of them attended a training! This should be taken into consideration when deciding.

The second variable is about your favourite topic: Brains ;-) If you have in your team, let’s say, someone with 3 years experience in the product X, you might consider to acquire product X as it will allow to speed up your Web Analytics Journey. Taking into consideration that WA takes time, this is not something to overlook.

That was my 0,2 euro cents from shinny Brussels ;-)

Cheers and see you next week in SF,

René

PS. For those of you attending Emetrics San Francisco, a little word from our Godfather.

1. Your latest post you mention that Visual Sciences is out there by itself. I’m wondering whether you’d class SAS eBIS tool in the same light. Given the fact that both are enterprise solutions, integrate seamlessly with campaigns and additional data-sources and customisable/shared dashboards for users/content groups. I would say that Visual Sciences does have a massive appeal with functionality, the unusual, but effective database structure and the sheer quickness of the tool - and SAS whilst benefited from a world class back end does have a lot of ground to make up for their reporting interface and dashboards.

2. Our mutual friend (and your fellow director of Web Analytics Association (congratulations by the way)), Neil Mason, did a fantastic slide at eMetrics showing what he perceived as the place of each tool (believe the axis were cost and functionality). Thought this may agree with your blog and perhaps would be a good addition.

3. With regards to the statement about “taking it for a spin”. I’m very aware of the exorbitant costs of running a proof-of-concept (POC) and the pressures that sales people put firms. I feel that the high cost of POCs may force managers to go with the tool despite a rocky/unsuccessful implementation and trial - merely because the tool is already installed and the accountants (which run most companies) say “Justify this expenditure of implementation, removal and the tool itself”. In financial services (for the last three years) getting any dedicated IT resource for web analytics is pretty much on par with trying to get to the moon with an elastic band and a lollypop stick.

So that there is a helluva lot more cost to consider that managers perhaps don’t realise on the onset.

I feel that a level of standardisation (which I hope the WAA can help on a global basis) will enable non-technical users and non-experienced companies to understand web analytics, measurement and its importance in the 21st century. Looking at my most recent blog £72bn ($150bn) in the UK will be spent online by 2010.

When considering an analytics solution, companies should first determine the questions they need to have answered, and then look for a tool that helps them find those answers.

We prefer Indextools over even the most expensive analytics packages out there because of it can be extensively customized, it has a very useful API, and it’s more competitively priced than other enterprise level tools. Heck, you can get a basic account for $50 a month!

And no, Dennis did not pay me post this comment!

Wayne Vaughan, President Fuscient
Fuscient - The Science of Marketing

While Visual Sciences was not able to participate at the level Phil wanted, pretty much everything you’d want to know about the other major vendors is contained in this document. And while some will consider $1,175 pretty steep, given the cost of the alternatives and the amount of work that went into the document, this is a thousand bucks that could save you tens of thousands of dollars worth of headaches down the road.

Anyway, if you’re in the market for some unbiased research on the vendors, check out Phil’s work here:

cmswatch.com/Analytics/Report/

If you’re not sure about dropping a grand on the report, you can get a free sample (not sure exactly what you get) here:

cmswatch.com/Reports/Try/

Anyway, nice work Phil and nice post as usual Avinash!

Eric T. Peterson
http://www.webanalyticsdemystified.com/

P.S. Dennis, thanks, I love you too! I look forward to meeting you in San Francisco next week.

Dans cette série sur les grands enjeux d’une implantation en Web Analytique (ou plus précisément en analyse comportementale) voici enfin le quatrième volet, tel que promis, sur le choix d’un logiciel. Toujours heureux de faire le paresseux (et occupé à lancer ma pratique de consultation), le temps mis a écrire ce post a fait en sorte que je me suis fait damner le pion par Avinash Kaushik qui a reformulé dans un post très clair sa position déjà exprimée par le passé. Si vous lisez l’anglais, allez plutôt consulter ce qu’il en dit cette semaine.

Quelques points cependant:

[...]

Great post.

I loved the soundbites, they wonderfully communicate the essence of each vendor. I am sure even they could agree with what you have said.

I have been reading this blog for a long time, your post today was so good I had to comment! Keep up the good work.

Passive capture has the other huge advantage in that web data is available to experiement with with out altering web servers, etc. which is the organization’s bread and butter. This means taht you can experiment with fraud-detection, compliance, feeding other data warehouses, or whatever without the fear of harming your web site.

Avinash Kaushik hat einen netten Artikel zur Auswahl der richtigen Webanalyse-Software gepostet. Ihn erreichen oft Anfragen nach dem Motto: “Ist Webtrends besser als Omniture?”, “Empfehlen Sie WebSideStory oder CoreMetrics?”, etc.

Wie so oft ist die Antwort: “Kommt darauf an …”. In dem Post gibt Avinash jedoch eine sehr gute Grundlage und Hilfestellungen, wie man an die Auswahl des Webanalyse-Tools herangehen soll. Besonders schön finde ich persönlich die drei Eimer aus denen man seiner Meinung nach für die Evaluierung zunächst je eins auswählen sollte:

[...]

“It is a true analysis tool, that some might say is expensive (based on pricing shared by a competitor at Emetrics 2007 in London)”

That competitor was ME! :-)

And I stand by it. ANY client looking into a price difference of some $400,000 per year on a average+ sized web property – should make sure that they not think twice and hard on how to justify a additional tool cost of that size.

I truly believe that the Analyst is the one creating the questions and the one trying to figure out the answers and the one finally recommending potential opportunities for improvement based on his findings. Do not get me wrong; I love Eric Peterson and his Visual Site as much as anybody else, but if I had the opportunity between a Contender (like us) and the opportunity to hire three “Avinash’s” (or would $400K only get me 2? :-) – I would go for the contender and the 2 to 4 additional Analysts any time.

But, hey. That is my 2 cents.

Dennis R. Mortensen, COO at IndexTools
My Web Analytics Blog

I not only think you are spot on in your recommendations, but it seems like - and this is a subjective comment based on my direct client experiences over the last year - most of the clients we deal with (from a RFP, RFI or any other sales pitch point of view) do exactly that!

We end up competing with one of the four American (bucket one) companies every time. Not because ClickTracks or NetInsight does not exist – but clients tend to add in only one or two Contenders (that is what I called bucket two – sounds better …. and you can add “Eye of the Tiger” as background music while reading this.. ha ha)

But to conclude: Well spotted Avinash!!!

See you in San Francisco on Sunday.

Dennis R. Mortensen, COO at IndexTools
My Web Analytics Blog