BI vs. Analytics

I know many of you already know my position on this, but I thought I'd get it out in the open and challenge all of you with a controversial discussion. In my definition – and believe it, I am fighting and defending it every day – analytics has always been, and will always be, part of BI. What many of the vendors and analysts describe as "the new age of analytics" I built at Citibank in the early '80s and then built in about 50+ enterprises in the '90s at PwC. I think the effort of trying to differentiate analytics from BI is a vendor-invented hype, since many BI vendors are running out of ways to differentiate themselves (and incorrectly so: see the next paragraph, and many other next-gen BI trends). I also disagree with the “old BI = bad”, “new analytics = good” premise that I see in many analysts' papers. You and I know that you can’t build analytics (OLAP, advanced analytics, etc.) without basic ETL, DW, MDM, etc. So nothing’s really changed as far as I am concerned: we are still fighting the same battles – silos, data quality, etc.

Besides, while this was indeed a differentiation a few years ago, today most of the top BI vendors do have OLAP and advanced analytics (see my upcoming 2010 BI Forrester Wave™ sometime in July) functionality, so it's really a commodity now. Instead, I prefer to write about REALLY new and game- changing trends like self-service, agile BI, BI SaaS, in-memory analytics, and many more. This is truly where vendors differentiate themselves and, much more importantly, what makes a true difference for the users of BI. Another potential game-changing trend that I see is that rather than fruitlessly trying to align business and IT for BI, I say: let IT handle data prep, and then let end users do their own BI and analytics.

Bottom line: there are plenty of real and actually useful BI trends and next-gen technologies and approaches out there. Let's concentrate on them.


BI versus analytics?

My first response was -- "he's correct that the debate is vendor hype" followed by "his attempt to steer the debate appears to be analyst hype".

So goes the risk when challenging others to a debate we are invested heavily in-- whether in a financial sense or personal legacy.

I think it's a foolhardy debate-- my view is that my time is better invested on designing better products. That said, your use of the word always for BI is interesting-- analytics is more appropriate to the cultures of governments, non-profit, and basic research-- and BI for business. So I guess I have to disagree to a point-- analytics by proper use of the words considers more than BI, although within the corp market it's certainly plausible-- even likely, that BI will continue to rule-- limitless budgets can often achieve whatever they please in ESW given customer apathy; although not always...

I can also see industries like health care moving more towards analytics than BI, even though quite often intending to define very similarly-- a cultural issue.

I am in agreement from my seat on what you prefer to focus on -- although predictive analytics certainly is included in our universe and those of large organizations we talk to, among others.

One macro trend I see that apparently most leading vendors and analysts don't is that most of the decision makers in large organizations could care less what the product is called--- they are primarily interested in function, interoperability, crisis prevention, more accurate decision making, total costs, risk, lock-in, etc. etc. etc.


Mark Montgomery
Founder and CEO

It's all semantics

True, I agree it's all semantics. And I, as a person who's been a hands on practitioner for over 20 years, before I became an analyst, actually couldn't care less about what we call it. And I especially agree with your last point, hence my position on business and IT misalignment

self-guided systems

It makes a good companion post to this piece. We've been working on self-guided systems for many years-- actually long before the commercialization of the web -- the promise was what originally attracted the medium to me -- initially for serving small biz who couldn't afford on site experts we teamed with in our consulting firm.

Two decades later I (and many others no doubt) am constantly appalled by the status quo in some of the largest orgs-- particularly relating to crisis prevention and sub-functionality.

Good luck with the clean line between IT and business strategy.... -- the minority of course have trended towards business leadership in the CIO, but emphasis on (super) minority is warranted. I certainly agree with the basic assumptions and need, however, and don't see anything other than the relationship between vendors and the IT ecosystem getting in the way.

One who often plays the role of contrarian -- simply due to blatant need -- and self-preservation for the species... I appreciate the effort.

I would agree with this

I would agree with this statement as well and it is rather confusing to differentiate them. Traditional BI (old BI) is a key enabler and must for advanced analytics (new BI) because how can you embed a cross-sell prediction into a transactional system without MDM, ETL or even Reporting? There are still plenty of advancements to be made in the traditional BI area.

On the topic of advanced analytics I see that a little bit different because advanced analytics capabilities are not all equal that is also the main reason why the adoption rate is limited to a few user types and business functions in many cases. The ability of self-service becomes much more challenging for example with advanced analytics, e.g. a Market Basket Analysis is easy but a Predictive method or even most statistical functions are quite challenging. So I wouldn't say that advanced analytics is a commodity necessarily, just because you support a data mining algorithm or a statistical method does not give you advanced analytics capabilities that can be used readily.

This Still Holds True

This is an older post, but still holds true. There is no BI vs. Analytics. If you don’t have a strong semantic layer, your analytics are only as good as they once were. There is no sustainability. As your data changes underneath your analytics, there is no way you would know. This is why data visualization and data discovery tools “alone” – which are in essence analytics with a weak backbone of metadata and semantic layer - are not ideal for an enterprise solution. They require a lot of manual work behind the scenes to keep the data fresh and accurate. A mammoth undertaking for IT that would drive continuous TCO issues.

Farnaz Erfan - Product Marketer