The 100 Gbps Optical DWDM market is finding its footing, growing a  massive 85% Y/Y in the first quarter of 2015, driving revenue up to over one billion dollars.  Long haul applications continued to drive the majority of the demand for 100 Gbps and we estimate that approximately two thirds of wavelength shipments were for long haul applications versus metro applications in the quarter.

Price erosion intensified following a few quarters of better than average price declines—the average selling price (ASP) of a 100 Gbps wavelength declined a little more than 20% in 1Q15.  The main factors behind the higher declines and our predicted future price erosion are:

  • the growing interest in 100 Gbps wavelengths for metro applications
  • introduction of lower priced pluggable CFP2-ACO (analog) components
  • aggressive competition among vendors to obtain new network footprint

A select few vendors have captured a high share of this lucrative market with the early development of coherent optical DWDM line cards using internally developed DSP technology.

The top five 100 Gbps DWDM vendors by shipment share and revenue share in the trailing four-quarter period ending 1Q15.

Vendor Shipment Share Revenue Share
Alcatel-Lucent 13% 15%
Ciena 14% 16%
Huawei 27% 27%
Infinera 16% 10%
ZTE 12% 13%

 

While most would think that a vendor’s shipment share would be the same as their revenue share, it doesn’t seem to be the case for 100 Gbps DWDM for good reason.   The reason being that not all 100 Gbps DWDM line cards are identical; they have not yet been “commoditized.”  Hence, 100 Gbps DWDM line card prices are determined by the market based on a number of factors, a few of which are:

  • region of sale
  • mix of metro, long haul, and submarine applications
  • product cost
  • pricing strategy

Therefore, varying vendor prices due to factors such as these make the difference between shipment share and revenue share results.

Beyond 100 Gbps…

We anticipate a dramatic improvement in demand for >100 Gbps wavelengths such as 200 Gbps wavelengths this year as additional vendors introduce their flexible modulation line cards.  In 2014, Alcatel-Lucent was the only vendor with availability and volume shipments of 200 Gbps capable line cards.  We believe both Ciena and Cisco have started to ship their 200 Gbps capable line cards in 1Q15 and that first revenues will be recognized in 2Q15. Other vendors with 200 Gbps capable line cards include Huawei, Xtera, and ZTE.

Having just returned and digested the information gathered from the small cell event in London last week, it is becoming increasingly clear to me that small cells are doing just fine and the outlook is brighter than ever, even if there were signs of skepticism and pessimism during the event.

The pessimism at the show, which was more confined to the non-macro vendors, stemmed from the fact that non-residential small cell revenues are apparently lower than some expected, and as a result, the future is now more uncertain. In other words, the business case was better three years ago when non-residential small cells were mostly an unproven vision, while the business case today is now suddenly more murky even though the technology has been proven in the field and early adopters and innovators are starting to deploy commercial small cell networks, and larger carriers are for the most part including small cells in the discussion every time they are issuing new RAN contracts. If it does not make any sense, then it is because it should not.

Now is more important than ever to review Geoffrey Moore’s classic – Crossing the Chasm. Some of the feedback I received during the show sounded too familiar…

  • Management is getting frustrated, revenues were supposed to be much greater by now, we are not sure if management will want to invest in LTE-LAA because the return has been so poor on LTE.  Management wants to see some return on existing investments before making new investments
  • We are running into difficulties running up against the big RAN guys. The verification process is just too intense. We are now looking at moving into areas that the macro RAN vendors don’t prioritize, such as rural areas, etc.
  • We are not sure why we lost that contract. Our price was significantly lower than that of the incumbent macro vendor.
  • The macro LTE network has done much better than we expected and maybe small cells volumes will not accelerate until 5G
  • We are accelerating training across the board. Our entire enterprise channel will know how to sell small cells.

The reality is that while small cells are gaining traction, commercial volumes are still smaller than some expected and deployments are primarily confined to the innovators and early adopters. And even if we estimate the market grew 3-fold in 2014 and will more than double in 2015, there will be some bumps in the road before volumes will scale and the early majority and late majority carriers deploy large scale deployments.  But the technology is maturing, demand is improving, and the gap in the business case between macro and small cell is narrowing as the carriers complete work on their macros.  And more importantly, new entrants such as SpiderCloud and RAN vendors with smaller macro footprints including both Nokia and Alcatel-Lucent are winning contracts even when the macro is supplied by Ericsson or Huawei which means that new entrants should not be forced to only focus on niche markets, though in some cases of course it makes sense to be realistic.

Now is not the time to get discouraged and assume the business case will not work until 5G.  And if you have had some success, now is not the time to waste money training the entire sales force. Now is not the time to focus solely on existing customers and satisfy every single item of their demanding wishlists. It is equally important to look beyond and understand the demands of the early and late majority type customers that might be deploying small cells in the next one to three years. Now is not the time to focus solely on price wars – this is a big change for carriers and performance assurance coupled with a future proof roadmap will generally carry a lot more weight inside the chasm than price competition.

We have not changed our 5-year projections since we started making small cell projections three years ago suggesting small cells would account for five to six percent of the total RAN market by 2017, and despite the signs of pessimism during the show last week, I am more confident than ever that this goal will be met.

In short, some frustration is understandable. But confusing unrealistic expectations with the enormous opportunities that remain on the other side of the chasm can be extremely costly.