Bye-Bye Neoview?

Mark Smith, the CEO of Ventana Research, writing for Information Management on the 8th of December, has a detailed article about HP in which he states that “the company has decided to leave the BI market, which includes analytics and data warehousing software. It will do this by shutting down its marketing, sales and product efforts with HP Neoview and related technologies.”

While I am not aware of any formal announcement to this affect I have no reason to doubt Mark’s comments. Indeed, I have to say that I am not surprised. The signs were not encouraging from the beginning. For example, I was asked by an HP executive before the launch of Neoview whether I thought the product should be in the hardware (disk drives) division! I think that reflects HP’s long term attitude towards software and, while there is lots of hardware in a data warehouse (whether it is an appliance or not) it is ultimately a software product.

I was similarly consulted as to who HP should think about partnering with for Neoview. I suggested that SAS might be one such company. The executives I was discussing this with agreed that that might be a good idea. And SPSS (now IBM) I suggested. The response to this was: who? Now you have to applaud HP for recognising its shortcomings and asking for help in a field where they were clearly not experts but you would have thought that they might be a little more educated than that.

So, the signs were not good from the beginning. And then there was the product itself. This was hugely over-engineered for fault tolerance and not enough for performance. In particular, using a fabric-switched architecture for connecting to disk, as opposed to direct attached storage, showed a fundamental misunderstanding of the most basic requirement for data warehousing, which is performance.

What’s more I happen to know that HP licensed a data warehouse from another vendor (that will remain nameless) in order to do benchmarks against its product. According to that other supplier it was beating Neoview hands down. Of course, they would say that. But I was always inclined to believe them if perhaps not in such generous terms.

This is not to say that there aren’t or weren’t some very good features in Neoview but then the company also had a marketing problem. Namely, that it couldn’t decide who it was competing with. It started out talking about competing with Netezza (now IBM) but it soon became apparent that it couldn’t compete with Netezza on either price or performance so it switched its target to Teradata: thinking that it could match it for performance and beat it on price.

But it couldn’t do the former and failed to recognise that Teradata itself was being forced to react to the emergence of Netezza, Vertica, Greenplum (now EMC) et al with improved ease of use, price-performance and so on, so Neoview failed there also. As a result its run rate for acquiring customers was pitiful.

I am not the first to say this but HP’s philosophy has always been that software is a way to sell more tin—and if you’re focused on tin then it should come as no surprise when you become a graveyard for any software businesses you acquire. So, if this is the death knell for Neoview I cannot pretend to be surprised. Still, I have been saying for some time that there are too many vendors in the market. There are still too many but at least there is apparently one less.

Philip Howard is Research Director (Data Management) at Bloor Research. Data management refers to the management, movement, governance and storage of data and involves diverse technologies that include (but are not limited to) databases and data warehousing, data integration (including ETL, data migration and data federation), data quality, master data management, metadata management and log and event management. Philip also tracks spreadsheet management and complex event processing.