Dell and HP have been in sharp contrast over the last few years. Almost every Dell acquisition has gained market share, profit and strength, while HP destroyed Palm (by expressly ignoring the plan put in place to save it) in a few short months after destroying Voodoo Computers previously. Executive retention at Dell has been strong, while HP has visibly lost key executives (up to 25 percent from the Autonomy unit alone) from each of the acquired properties, and software company acquisitions are all about the people.
Let’s look at the future of the Quest Software acquisition and why it should succeed and could fail this week.
Acquisitions: Dell vs. HP
The companies take two very different approaches to acquisitions. Dell brought in Dave Johnson who had been instrumental in fixing IBM’s acquisition practice. He could, after coming over to Dell, step even more sharply away from IBM policy to create a next-generation acquisition model. HP continues to practice the popular, but largely discredited, practice of trying to actually merge an acquired company into an existing large enterprise.
But as Dell’s own software unit got larger, the risk of overlap increased and its ability to continue a model that kept the acquired companies very independent declined. With Quest, Dell will now need to start putting in place a more formal management structure and will need more product rationalization (eliminating redundant products), which will create retention problems and drag. This will then form the big test: Can this unique and successful model that Dave Johnson has created scale to encompass the large software entity that will result?
Keeping the Core Strategy
Why I think this change will result in some executive bleeding — not everyone will tolerate having their authority or status diminished — I also think Dell is well positioned to make this work. This is because at the core of Dave Johnson’s acquisition strategy is the identification and protection of key intellectual property and human assets. What makes most acquisitions fail is a process that places the value of the merger over the value of what was acquired. It becomes more important for the acquired company to look like another part of the acquiring company than it does to retain the core assets that were bought. As a result, with most mergers, the cost vastly exceeds the value because the value is mostly destroyed by the merger process.
Johnson flipped this on its head and the first part of Dell’s process is to identify the assets and develop a strategy that will protect them. If there is a conflict between the merger process and protecting the asset, unlike the more common path where the merger process wins, the asset value wins. So unlike other firms that overpay, Dell actually ends up getting more value than what it initially paid.
The reason I’m very conscious of what makes Dave Johnson’s process special is that I, too, worked for IBM and one of my jobs was as a member of the Internal Audit Tiger team, which was missioned to clean up acquisitions. Most were train wrecks largely because folks executing the merger had no motivation to protect the assets; they were simply missioned to get it done. This would be like a doctor doing a heart or limb transplant who was measured on speed but not measured on whether the part or patient lived or died. And yet that is how most mergers are done.
The reasons behind this are largely behavioral. You have titles in the new company that don’t align to titles in the old one. For instance, you could have a VP in the new company without any direct reports, but a policy in the old company that says a VP must have an organization of 1,000 people, so the new VP, who may be critical to the business, is forced out. You may have a process that works fantastically for the acquired company but that is in conflict with a far less efficient one with the old company, but the old firm doesn’t want to retrain, so the acquired firm has to learn and implement an obsolete process. In one instance where I saw this implemented, average engineer turnover went from 10 to 200 percent a year and was at the core of the division’s failure.
So the need for consistency, particularly when it comes to individual status, overrides the need to protect the asset and the asset is destroyed (anywhere from 70 percent to 90 percent of the time, and likely these numbers are understated). This is the institutional insanity that Dell’s process was designed to overcome and why other firms have ignored this lesson even in the face of Dell and EMC’s (RSA/VMware) successes with it.
Dell’s next step is one of blended integration and this is where we find out if the need to protect the assets and the need to protect titles and processes can be as elegantly balanced as they were when separated. The reality is that titles and actual power generally don’t align and processes just have to work well; they don’t have to be consistent. This means that, as long as Dell stays focused on protecting the assets, the resulting organization can be successful. But human behavior is behind why these efforts often fail and it is often unwise to bet against foolish human behavior in what is a largely political process. This means that success is not certain and that is what will make what comes next for Dell and Quest Software very interesting to watch.