Before moving to Amazon Web Services, it took ITV ’21 days to make a single firewall change’

April 17, 2015

That’s more of an indictment of ITVs technical leadership than an endorsement of aws. What were the leaders doing? http://www.computing.co.uk/ctg/news/2404193/before-moving-to-amazon-web-services-it-took-itv-21-days-to-make-a-single-firewall-change

Advertisements

Why outsource wireless and not landline networks?

May 6, 2009

Telcom Ramblings asked a very pertinent question: Outsourcing Networks: Why Wireless But Not Wireline?

That is a very good question and I went off and did some thinking about why this is so.  There are a few points I think which may be relevant:

  1. The wireline customer edge is quite high touch. Provisioning, testing, turnup, IP address assignment etc. The wireless customer edge is quite low touch – buy a phone, get a billing relationship established, or turn on a prepaid account, and you’re in business.
  2. Margins are higher for wireless. On the wireline side, the network is running so close to the edge you cannot insert a middleman.
  3. Wireless is more contained with very well defined interfaces to the rest of the world. Wireline’s interface to the rest of the world is a lot more flexible, which is useful, but evolves and changes quickly. Outsourced companies by definition are not going to be fast reacting as in-house. You will need to develop training material, rework processes, etc. to fit into the statement of work for the outsourced vendor. If they react very fast, they have to have a dedicated team for the operator, and then they lose the economies of scale, raising cost.
  4. Wireless is differentiated in a few things: services offered to the customer, price, VAS. With well developed interfaces, you can focus on things like ringback tones, voice-mail, carrier decks. With wireline, there aren’t as many value added services you can deliver – most networks focus on pricing, customer service and reliability of networking. Those are your core competencies. They differentiate you. Those are going to harder to outsource.
  5. Infrastructure as a competitive advantage. There are wireline operators whose infrastructure is a competitive advantage. This is true for wireless as well (verizon wireless for example has made a business out of infrastructure), but for the majority of people outsourcing their network, most of the marketing collateral talks about things other than infrastructure. For a wireline operator, your business is defined by infrastructure: where you are, what you can deliver, how good your connectivity is to the rest of the world.

This doesn’t mean that wireline cannot be outsourced. There are several networks (Bharti/China Telecom/Saudi) among others, that are either looking at, or are outsourcing operations to vendors. Some of it is driven by the fact that you cannot hire competent enough folks in volumes necessary to run a good shop.  For some operators, wireline is not their core competency and they are filling in the gaps. If you need to get a checkbox on an RFP, then outsourcing is good enough. Some of it is driven by force multiplication: You are doing a build-out or expansion and you need to bring in temporary help to enable you to get over the hump. At the end of the day, the temps go away and you do ongoing adds/moves/deletes with a much smaller in-house staff.

In conclusion, while I believe outsourcing wireline is possible, it is going to be tougher than it first appears. I believe the same to be true for wireless as well. In the long run, the base cost of providing service is going to be within a few percentage points, done in-house or outsourced to a vendor who is not shifting cost around – e.g. making up a service loss by margin on hardware sales.