FiOS: Just before the GPON RFP

I have gotten us to the point where the GPON RFP will arrive. I want to recap a bit, because I think it might be useful to understand where we are.

On the deployment front, things have dramatically cleaned up. Verizon is doing about 10,000 ONT installations a day. Verizon has gotten the Motorola solution approved but the product is restricted to a small market share given all the progress that we had made. We were working with Verizon to lower the cost and time associated with installations. We were making lots of progress on installed quality and the reported failure rates were declining nicely. There were still challenges with first installation quality, especially when new areas were turned up. Things were going pretty well. We were very surprised with the high take rate that was being reported. But it was all good.

On the cost front, our friends at Vinci had done a tremendous job in both cost reduction and selling small price increases. ONT margins were still too low but they would poke into positive territory. There were two non-product issues here. Tellabs had a really simple model of cost allocation. They assigned overhead costs at a flat rate to every product. This penalized ONT margins as it was a (relatively) high volume, low mix product. On top of that cost to market and sell ONTs were essentially 0. If Verizon installed an OLT, they had to purchase ONTs to serve customers. So, ONT margins absorbed lots of costs that were actually related to other products improving their margins. On top of that, Tellabs used ONT pricing to leverage other products into Verizon - particularly the 7100. Just as ONT margins were getting to be okay, Tellabs gave Verizon a price break to get them to deploy the 7100. I always thought that the complaints about ONT margins based on both of these issues were rather hollow.

On the technology front, we had gotten Ethernet into the UMC and had met all the ONT and PON requirements of the RFP. Many of the features that we developed never actually got deployed and that delayed things that were useful to Verizon in practice. This kept tension between the organizations and things never were quite on the friendly basis that I have seen in other relationships. It was not a toxic relationship but it wasn't exactly wonderful.

When I posted about Broadlight, I talked about where the technology was going. Our Executives did not want to do GPON and spend all the development dollars. This made a lot of economic sense as (if you think about it) we had been investing in POTS for over 100 years, DSL for less than 10 years and BPON for under 5 years. Now the network would switch technologies mid-stream of a large rollout. Tellabs would have to spend a lot of R&D to make a new product that was low margin. We pitched an upgrade to BPON as an alternative. There was a version that had a 1.2 Gbps/622 Mbps that we could implement. That didn't do much for Verizon and they eventually rejected it.

Now finally on a technology note. Remember that Verizon thinks the UMC is limited to 622 Mbps, even though at this point it was doing 2.4 Gbps. The connectivity based on this went from 155 Mbps per OLT to 3x155 Mbps to 3x1.0 Gbps over 2 years. Verizon could upgrade to 3x2 Gbps at their leisure.  We looked at the OLTs when we got the opportunity and found that they had no congestion.  We never found a system that had ever stored more than 3 cells.  That has probably changed at this point but I think its telling how underutilized Access Systems truly are.

So, BPON was making money. It had a huge market share within Verizon. Things were stable. Tellabs stock was up. Access was doing over $1B in revenue total. 80% of that revenue was from Tier 1 carriers. The 2001 Business Plan had not just succeeded, it had worked better than anyone could have hoped. And then it all went wrong.


Jim Sackman
FocalPoint Business Coaching
Change Your Business - Change Your Life!
Business Coaching, Sales Training, Web Marketing, Behavioral Assessments, Financial Analysis
Systemize Your Business Free Webinar!