Any big announcement is going to generate a lot of comments. In tech, or almost any other space, M&A is a big announcement, particularly if it involves well-known players. So it was with the rumor that HPE was buying Juniper, and I joined the comment frenzy in my blog yesterday. But later yesterday, HPE made the deal official, and offered some comments on it. The press release is HERE, and it gives us an opportunity to assess what HPE says it wants to do with, and expects to get from, the Juniper acquisition. It’s also generated some other commentary, and so I propose to take another look at how it might, and must, evolve.
The first directional comment in the release comes in the second paragraph: “The combination of HPE and Juniper advances HPE’s portfolio mix shift toward higher-growth solutions and strengthens its high-margin networking business, accelerating HPE’s sustainable profitable growth strategy.” This is expanded in the following paragraph, which says the deal with double HPE’s networking business. I think that the combination of these two paragraphs are at least potentially supportive of my point yesterday, which is that strategic influence in the enterprise network space has decisively shifted toward the computer/software vendor, creating an opportunity HPE could now exploit.
But…there’s also the quote from HPE CEO Antonio Neri: “This transaction will strengthen HPE’s position at the nexus of accelerating macro-AI trends, expand our total addressable market, and drive further innovation for customers as we help bridge the AI-native and cloud-native worlds, while also generating significant value for shareholders.” This is no doubt the source of the Wall Street and media assertions that the deal is all about, or at least mostly about, AI. If that were true, it would be an astonishing justification given that HPE could acquire AI assets without buying a network company.
Finally, we have a Light Reading analysis that says the deal could be bad news for the telcos. The article points out that HPE’s stated focus is really mostly about enterprise issues, and all the network equipment vendors have (as I’ve noted in past blogs) been backing away from the telco markets because of enduring and apparently persistent issues with profit per bit. But even if telco market growth is limited, the market accounts for about a third of Juniper’s revenues. If it’s true that in the eyes of HPE, the telco world is an also-ran, then there’s a lot of money that could be lost if telcos react that view as we could expect they would.
How can we balance these three pieces of news? Let’s take a shot.
I’ve heard from HPE people, including some pretty high up the organizational chain, who agree with my overall theme that the deal would be a powerful step in engaging network equipment purchase decisions and linking them to HPE application and IT skills. Most also agree that this is going to be a heavy lift for a combined company that hasn’t been a marketing powerhouse. You can’t get sales organizations to drive new mission statements; they have to sell and make quota. It’s all marketing, win…or lose. I’ve not seen or heard anything that changes my view here.
Some push on AI, in light of that broader goal, is reasonable. AI is Juniper’s secret sauce, the thing that enterprises are most likely to recognize as a differentiator. If HPE wants to exploit its IT-vendor strategic influence to push Juniper equipment, it makes sense to tout the thing that enterprises already accept as a differentiator.
The AI story, I think, may also be a bit of playing to the crowd. Most companies don’t want to take a complex story to either Wall Street or the media, given that neither audience is really particularly equipped to absorb it. There’s a line in “The Man Who Shot Liberty Valance” that could apply here: “This is the West, sir. When legend becomes fact, print the legend.” AI as a story, perhaps more a legend, has legs and so it makes sense to ring the AI bell in a press release. That’s particularly true when you consider my criticism of both HPE and Juniper’s marketing/positioning acumen. If you’re a great marketeer you can afford to go out into the world and make people care. If not, you have to start with something they already care about.
OK, what about our last citation, the Light Reading story that says the HPE/Juniper deal could hurt telcos. If that one has a germ of truth then HPE is in big trouble because a loss of even a decent chunk of telco revenue would doom any chance of the deal being accretive to HPE’s revenues any time soon. But here I think that rather than having a potential gain at risk to substandard execution, we have a potential loss whose risk may be overstated.
Juniper and arch-rival Cisco both see a revenue/profit decline in the service provider or “telco” space. They’ve said it on earnings calls, and certainly everyone on Wall Street knows it. They’ve also indicated that more emphasis on enterprise sales is the answer. Given all this, why would anyone be surprised if HPE talked up enterprise network opportunities and the product elements that were linked directly to them, elements like AI? The HPE statements don’t raise the risk level, they just reflect what everyone in the industry already knows.
Might the HPE/Juniper deal actually lower the operator risk? With regard to AI and the enterprise-centricity of the value proposition, aren’t telcos worldwide trying to find business services to sell? Isn’t is possible that AI elements in services could make those services more exciting, and raise their value to enterprises, and in doing so let them sustain higher prices and margins for service providers? This isn’t automatic, but it’s at least credible.
And that’s not all. Shall we trip down memory lane to Network Functions Virtualization (NFV)? HPE was an aggressive early player in the NFV space, having hired the first head of the NFV ISG. You can argue that NFV was an early recognition that the future of network services, and thus of network infrastructure, was the marriage of hosted features and connection hardware, which means computers, software, and router/switches. You could argue that selling all these things makes it easier to create that unified future, especially given the market-wide drive toward the single source.
Suppose that HPE/Juniper is motivated not by what I believe to be a stellar opportunity, but by the more typical short-term, tactical, number-crunching, workforce-shedding, bean counting? Suppose there is no awareness of the stuff I have talked about yesterday and today? OK, that’s a possibility, maybe even probable. Yes, it asks us to accept that a major-company CEO would make a $14 billion mistake, but I’m sure that’s been done, and worse. But as long as we’re playing the “what if” game, suppose that the adage “Judge an enemy by capabilities not intent” is true. There’s no shortage of belief that this is a stupid move, but that’s not the only possibility. This deal plays to market trends that have been visible, and in fact becoming dominant, for well over a decade. Those trends create a reality that’s something that needs to be taken as seriously as the challenges the deal faces.
I think that whether HPE recognizes the strategic shifts that favor IT vendors in defining the future of enterprise networking, or is simply responding to things like IBM’s success with hybrid cloud and IBM’s strong and broad strategic influence. Whichever is true, adding networking to their portfolio makes HPE into perhaps the ultimate one-stop tech shop. It also gives HPE a bunch of assets, assets like Juniper’s superior SD-WAN technology, AI, superior data center switching, and strong service provider product line, that HPE could exploit better than Juniper has.
The the “could” as opposed to “will” here remains the key in my view. Enemy capabilities aren’t always exercised, and $14 billion is a lot of money to spend if all you’re doing is moving defensive pieces around the market board. This is, as I said yesterday, a major marketing challenge for a combined company that’s not demonstrated great skills there for a long time. There is nothing we can say yet about the chances of success, and so all I can promise is to follow the progress of the deal and report objectively, drawing on enterprise and telco comments, what I believe is going to happen. And I do promise just that.
