How Can Dell Get Into Big Data?

I’m going to start this post by summarizing it in three lines:

  1. Dell was right to not buy Aster Data
  2. Dell should not buy ParAccel
  3. Dell should…

On second thought, let’s leave a bit of intrigue for the end of the post (now do your best to not jump ahead for the spoiler!).

Recently, there’s been a lot of suggestion that Dell should use software acquisitions as a way to get into the Big Data market.   For example, in January, GigaOM’s Derrick Harris listed 5 Cloud Software Vendors Dell Should Buy, with Aster Data topping that list (a recommendation he reiterated a month later, following Dell’s Q4 earnings call).  His line of reasoning, echoed by a number of other analysts, seemed rational: Dell wants to get into the Big Data market.  Dell has a reseller agreement with Aster Data.  Therefore, Dell should buy Aster Data.

But it didn’t.  And that was exactly the right move.  Now, ParAccel is trying hard to position itself as a prime target, but Dell isn’t biting.  Here’s why.

Understanding Dell’s Acquisition Strategy

Dell is a company that seems acutely aware of both its strengths and its limitations.   As it seeks to expand into higher margin offerings, this awareness has had a clear influence on its acquisition strategy.  In 2010, Dell made four software acquisitions: Kace, Scalent Systems, Exanet and Ocarina Networks.  All four of these companies were very similar, in that their products manage or otherwise interface directly with hardware that Dell sells.  Put another way, these products are “one degree away” from Dell’s existing core competencies.

This is a key characteristic that has been fundamental to Dell’s ability to monetize these acquisitions for one simple reason:

Dell does not have a significant enterprise software sales force.

In contrast to HP and IBM, Dell’s sales force is probably the most hardware-centric in the industry. Such a sales force could reasonably be expected to sell hardware provisioning software from Scalent.  But a complex data analytics platform is a completely different animal.  This is exactly why Dell complemented its acquisition strategy with targeted partnerships driven by its specialized Data Center Solutions group.  Dell knows the limitations of its broader sales force, yet it’s astute enough to recognize that product synergies exist and that through careful partnering, there’s a road to revenue.

Why HP + Vertica is not Dell + Aster Data

Although I have my doubts about the likelihood of the HP/Vertica acquisition succeeding, at least it makes sense on some levels.  HP understands software and it knows how to make software acquisitions.  It also has some experience in the data warehouse market (even though Neoview failed miserably, its sales force still benefits from that experience).  In contrast, Dell has not made a significant software acquisition that wasn’t one degree away from their core competencies, and is only just starting to learn about the Big Data market.

Acquiring Aster Data would undoubtedly have led to some quick product wins, most obviously a true Dell/Aster Data appliance.  But then what?  Aster Data’s sales force was relatively small and primarily US-based, so to truly scale worldwide, they would have effectively needed to build an overlay sales force from scratch.   While not completely out of the question, this would have taken significant time and effort, during which time Teradata, IBM/Netezza, and EMC/Greenplum would have left Dell/Aster Data in their dust.

Why HP + Vertica is Definitely not Dell + ParAccel

At this point, it should be pretty obvious why I think acquiring ParAccel would be a really bad idea for Dell.  ParAccel is far earlier in its development trajectory than Aster Data was, both in terms of product development/maturity and sales acceleration.  Moreover, it’s not clear yet whether or not ParAccel’s product actually scales to “Big Data”.  So if buying Aster Data wasn’t the right move, acquiring ParAccel definitely isn’t.

So What Should Dell Do?

If Dell is serious about making a big leap into Big Data, there’s only one route with a clear path to success:

Dell should acquire Teradata*

Think about it:

  1. In 2009, Teradata was viewed as a prime acquisition target for SAP.  Their market cap has increased significantly since then, but even at today’s level ($9.1B) it’s still within Dell’s reach.
  2. Teradata currently uses Dell hardware for some of their appliances and, as mentioned earlier, Dell has a reseller agreement with recently-acquired Aster Data, so there are already synergies.
  3. Purchasing Teradata would give Dell a large, worldwide sales force with experience in data warehousing, data analytics, and hardware appliances (not to mention a substantial services arm to complement its Perot acquisition).
  4. Oh, and by the way, Teradata’s very profitable.

The third point is the reason why such an acquisition could succeed where acquiring a smaller software vendor would fail.  With Teradata, Dell would gain an immediate ability to sell into existing and future Big Data customers anywhere in the world.  Add to that the cost benefits gained by combining an appliance vendor with its supplier (similar to what Netezza gained from its IBM acquisition) and on paper you’ve got a great match.  I’ve already stated my belief that Teradata + Aster Data is one of the top three Big Data contenders going forward.  Add Dell into that mix and..?

Now, do I think Dell will do it?  Hard to say.  This would be a huge move for a company that’s notoriously conservative when it comes to acquisitions (recall how quickly it got cold feet when battling HP for 3PAR).  That said, if Dell wanted to make a big splash, it’s hard to think of anything bigger.

Bacon Wrapped Data’s odds: 5-1.

* For any avoidance of doubt, this statement is entirely my own opinion and is not based on any non-public information garnered while at Aster Data.
This entry was posted in Aster Data, Dell, ParAccel, Teradata. Bookmark the permalink.

One Response to How Can Dell Get Into Big Data?

  1. Pingback: No Comment | blog.cieslewi.cz

Leave a Reply

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

*

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>