OSS Line

Big Data for Telco – Where is the Data?

If you’ve worked in OSS for more than 5 minutes you know that data, the availability of it, and its accuracy, is a bit of a thing. So, if you’re thinking about ‘doing’ some big data as part of an OSS or BSS project what data related challenges are you likely to face, and what is the reward for your efforts?

A few months ago I attended Informa’s Telco Big Data and Real Time Analytics event in London. My assumption had been that there would be a great deal of focus on customer data – Pretty much every analytics themed telco session I had been to in the past focused on the classic topics of determining which customers are likely to churn (a nicer way of saying ‘leave you’) and what products you should try to up-sell to each customer. Typically this is based on call/billing records, purchase history, payment records, and CDRs. All relatively easy to acquire data and available no matter what OSS/BSS/hardware vendors’ systems you have.

Network, OSS, operational data is much less consistent, much harder to acquire. So I was pleased to find that these issues were covered by a number of presenters at Telco Big Data, including case studies from Sprint and Telefonica.

Telco data is divided in to two broad categories, familiar to any data analyst:

• Structured data, which is easily processed due to it being in a consistent format such as XML, spreadsheets, databases and is usually available over system interfaces. Examples include: Call records; billing records; electronic records (IP, service); location records; inventory; MIBs
• Unstructured data, which has no consistent, logical structure making it hard to derive insight from its content without sophisticated, specialist processing. Examples include: Voice calls; texts; blog posts; web content; file downloads; apps; media content

You could also categorise these as ‘data’, the actual content flowing across the network, and ‘meta-data’, data describing the properties, sources, costs, etc. relating to the content data.

The structured meta-data is usually found in OSS and BSS systems. In OSS, the data quality and format will vary greatly between CSPs as each has their own software vendors (different inventory databases) and hardware vendors (different NMS interfaces, device MIBs, and stat collection capabilities).

The cost of acquiring OSS data is often high. There’s the ‘integration tax’, referring to the development and data cleansing needed to extract and use available data. And, often, OSS vendors and hardware vendors will charge a license fee to allow use of their APIs to get at the data that’s there, sometimes costing several hundreds of thousands of dollars.

In the unstructured data, telcos have access to real social data that the likes of Facebook and Google can currently only dream of – relationships, family, work patterns and so on are available, as is accurate location data. It was suggested at the Telco Big Data event that with relatively simple analysis it is even possible to determine the sex and age of customers, even if they aren’t named on the household bill.

Of course, monitoring and analysing data in this way quickly leads to privacy questions. I’m not going to attempt to discuss them here.

If big data analysis is done with appropriate care and respect for the customers privacy, it can pay-off for the service provider. At the Telco Big Data event, Sprint claimed they were already generating $10 million in revenue by selling, externally, market insight about their customers; Many other CSPs were claiming double-percentage-point increases in product sales by better targeting; Guavus, a vendor of telco-focused analytics software claimed a reduction of network equipment costs, for one customer, from $1bn to $54m based on better understanding of how the network is being used.

What stood out at Informa’s Telco Big Data event was the relative maturity of the solutions being implemented, the quality of the case study results and the pragmatic approach to identifying and leveraging available data sources. Unlike emerging technologies such as SDN, big data is ready for telco now, and has proven cost-cutting and revenue-increasing benefits.

Ps. Some would argue the correct title should be “where are the data”. I checked. Use of ‘data’ in the singular is valid, it’s just that many in-house journo/publishing style guidelines standardise on using it only in the plural. 

 

Ontology at T-Mobile Czech

Ontology, provider of really cool OSS data search and data integration technology, scores a live customer in T-Mobile Czech.

It says here that Ontology has been deployed by the Czech T-Mobile op co.

Ontology's products provide a means of pulling together data from multiple sources and joining it up to give a single end-to-end view of, for example, your network inventory. It does this using clever semantic search capabilities that I don't understand, but I guess it's a bit like automagically working out that two objects in two separate data sources share the same name, or location ID, or whatever, and merging the two object models in to one complete model.

Of course it's much harder than that as data quality issues tend to get in the way, even if the data can technically be joined-up. And I've worked with telcos wanting to consolidate multiple inventory data sources only to find there are no common 'keys' between one data source and another: They know they have E1 services terminating in the switch centre; They know they have E1 services on their SDH rings; But they can't match one object record to the other.

From the PR - Pavel Stamenov, Head of Access Network Configuration Department at T-Mobile CZ.  “The solution allows our users to work with the transmission planning spread sheets uninterrupted, with zero or minimum downtime prior its final migration to the network inventory system.”

What's interesting about that is it is not simply providing an umbrella data model, but it's also enabling business processes and a migration path from spreadsheets to the Amdocs network inventory system. Sound's a lot like Ontology is now delivering solutions that are similar to those that Celona used to offer.

Now, I don't know how much business Ontology does, but I do know how hard it is for smaller OSS vendors to get a customer to agree to do a press release. Great to see a company like Ontology, with some really innovative technology, getting deployed and getting recognised for it.

NetCracker Selected by Andorra Telecom

NetCracker will deliver a comprehensive OSS/BSS solution to Andorra Telecom.

It says here, "Their selection of NetCracker underscores the industry trend toward utilizing a single, unified platform for BSS and OSS."

Andorra Telecom is the principal  telecomuncation supplier in the Principality of Andorra, a European microstate with a population of 85,000.

The press release can be found here.

Cisco to Buy Cariden

It seems like most of my posts these days are about hardware vendors buying software vendors. Here's another acquisition in the OSS industry's (not so mini) acquisition bubble.

Cisco is buying Cariden for $141 million. Cariden is a small organisation focused on IP/MPLS network planning, a few tens of employees, and revenues estimated to be in the low 'teens (millions of $). That is a pretty serious multiplier considering Cariden's modest revenues, eclipsing Opnet's purchase by Riverbed that was announced a few weeks ago.

Buzz-words are already being cited as reasons for this acquisition: SDN blah blah; Cloud blah blah. But it all comes down to a need for good ol' OSS to manage the increasing complexity of IP/MPLS networks and the requirements for hardware vendors to be able to accurately propose new networks, capacity upgrades, and optimisation to their telco & enterprise customers.

Good old Light Reading has a little more detail on this purchase.

Coming Up: Telco Big Data Summit 2012

I'm jolly excited to be heading to Informa's Telco Big Data and Real Time Analytics Summit next week.

It says here:

"Cutting through the hype, the Telco Big Data and Real Time Analytics Summit will bring the telecommunications industry together to discuss data management best practice and strategy – debating when, where, how and why..."

It's a trade show. There will be hype. But I actually think of all the hot topics of the last couple of years, Big Data is the most interesting and one the most likely to have a signifcant impact on the OSS industry.

What I will be looking forward to finding out is if anyone else is considering Big Data analytics for the network, or will it mostly be revenue assurance and customer relationship management? We will see...

Report to follow post-event.

In the mean time, if you are attending this event and fancy meeting up to talk OSS, drop me an email.

 

Riverbed to buy Opnet

The OSS mini-bubble continues with another acquisition of OSS assets by a hardware vendor. This time it's Riverbed, who provide boxes for network monitoring and performance optimisation, buying Opnet, suppliers of software for network and application management.

Riverbed are paying approximately $1 billion for Opnet which, based on expected revenues of $198 million this year, means a reasonably healthy 5-times multiplier. That's a lot better than some of the recent OSS exits, most of which looked a lot more like fire-sales. But Opnet has relatively good revenues: Sizable, but still lumpy and unpredictable like any business selling to telco and large enterprises. Opnet has also been successful in moving in to data-centre, WAN and service management with their Application Performance Management products, giving them a portfolio that has a broader market than pure OSS/carrier-class product suites.

In fact, you could argue this isn't really an OSS acquisition but rather an IT acquisition: Most of Opnet's value today is not in their original network performance management products as the company's focus has shifted to application performance management in recent years.

You can find a bit more info in their PR here.

 

Ericsson to buy ConceptWave

This week Ericsson announced intentions for another OSS/BSS acquisition, this time at the more customer-facing end of the stack.

It says here, “ConceptWave’s unique offer is to provide an end-to-end catalog-driven suite of order fulfillment automation software with ConceptWave Order Care® and ConceptWave Rapid CRM™”.

This purchase by Ericsson is all about strengthening their BSS product suite with Customer Relationship Management and their BSS-to-OSS process integration with Order Management.

ConceptWave sits nicely on top of the Telcordia OSS suite that Ericsson completed the purchase of at the very start of this year. This more complete OSS/BSS stack competes better with the likes of Amdocs' long-standing product suite and NEC’s suite that has also recently been plumped-up by acquiring Convergys’ BSS assets.

Common Sense Won’t Sell Your OSS Products

The OSS market has changed... Part 4 of 4. This series of four blog posts is a synopsis of my contribution to a discussion I chipped in to on LinkedIn’s ‘OSS Gurus’ group.

Economics? RoI? Plain common sense? These won’t sell your OSS products.

So I talked about business benefits, CAPEX and OPEX savings last time. And you probably assumed the OSS buyer will primarily be driven by one or both of these KPIs when selecting OSS products and running OSS projects, right?

In lean times you’d think cost saving would be A Big Deal. If you can save the telco $100 million per year for just $500,000 investment, they’d bite your arm off. Not always. Rarely, in fact.

Pure economics common-sense and individual telco manager's KPIs are not always aligned. If they're measured on ARPU (average revenue per user), then they'd be buying solutions that increase revenue, not drive down costs. A manager/director in Retail, Products, Sales or Marketing will likely be measured primarily on customer growth numbers and ARPU, not CAPEX and OPEX savings.

Yes, someone cares about cost in the telco, and I think most of those guys would be buying more OSS if they had the budget. But they don’t have the budget if the business is losing customers or revenue. It tends to be the successful telcos that spend more on OSS. Obvious really:

1. They have more money to spend across the whole business

2. They have growing customer demands and service fulfilment pain that high-lights any OSS weaknesses

3. What’s the point in investing in OSS at the back-end unless you have customers coming in at the front-end?

It comes down to the CEO who decides where the bulk of the budget goes: To sales/marketing initiatives or network efficiencies? And again, she may be driven by something other than common-sense long-term economics because that's business, right? The financial market says you're a more valuable telco if you have stuff like:

1. An LTE license

2. Growing smartphone up-take

3. Increasing ARPU...

So that's where the budget goes this year, then a bit later on you get.... Profit!

Of course a CEO will know that you need a margin on that ARPU, but if she can dictate that cost-saving OSS initiatives must *also* have an ARPU up-side then that's where the budget will go before a pure cost-saving project.

In summary, you’re doing OSS so you’ve got a cost-saving story. That’s a given. Where’s the upside for the rest of the business? The City doesn’t give the CEO a pat on the back for making her planners and engineers jobs easier.

The Business Case for OSS

The OSS market has changed... Part 3 of 4. This series of four blog posts is a synopsis of my contribution to a discussion I chipped in to on LinkedIn’s ‘OSS Gurus’ group.

The business case for OSS is less clear than ever.

Are OSS customers driven by CAPEX, OPEX or revenue? Every pundit I’ve talked to has a different opinion.

Some say revenue is king. I agree that being able to demonstrate how an OSS solution generates revenue is A Good Thing. But most of the folks buying OSS, being back-office, are actually more concerned about their day-to-day job of Getting Stuff Done and not blowing their quarterly budget.

So we try and sell a vision at the CxO level instead, right? And there's some visionary CxOs out there who understand processes from BSS->OSS->Network need to evolve, and that enabling OSS enables new services and revenues, but the other 90% of CxOs will delegate all OSS decision making to the Manager of Network Engineering.

So you’ll be pitching CAPEX and OPEX savings to most of your buyers and key users. And which is most compelling is down to the local economics, accounting practices and the latest cost-saving edict from the CEO. There’s no universal truth about whether CAPEX or OPEX savings trump the other. Your solution better offer both and your sales guys are able to have an intelligent conversation about the challenges and business objectives of the customer to quickly focus the pitch on their specific issues.

Most customers won’t believe business benefits and return-on-investment figures given to them ‘cold’. If you can get the customer to provide just a few metrics up front along with their top two or three cost/revenue KPIs, then you can weave that in to the business benefits and present something much more credible. The good news is that even with conservative input figures you can generate very impressive RoI figures for most OSS products. If you can’t show an RoI in under six months your solution is overpriced or you’re not solving a problem anyone cares about.

Moving on from 1990s OSS

The OSS market has changed... Part 2 of 4. This series of four blog posts is a synopsis of my contribution to a discussion I chipped in to on LinkedIn’s ‘OSS Gurus’ group.

Systems are going to have to change from those Big OSS products developed in the 90s.

Big OSS inventories focus on reservation and allocation of, usually, fixed resources like SDH/SONET timeslots, DWDM wavelengths, subscriber ports, DDF/ODF ports etc, etc. What traditional inventory is less able to do is model shared resources that vary over time (like IP and the more dynamic MPLS protocols) because in order to capacity manage these you need a more complex traffic/service models, observed utilisation, etc.

Traditional ‘offline’ inventories need to be augmented with additional products to support this (and I'm sure Amdocs/Ericsson/NEC/Oracle have their offerings), or you assume 'the network is the inventory' and make OSS an analytics-centric, rather than database-centric, application.

Inventory is an essential part of an OSS stack, but the area for innovation and exciting new products/projects will not be interested in which timeslot is allocated to which customer. The exciting stuff will be dynamic analytics of the current network state, service profiles, traffic flows, what-if growth analysis, predictive failure analysis, service feasibility/profitability analysis...

The Big OSS products developed in the 1990s and deployed in the 2000s can be categorised as 'inventory-centric fulfilment' solutions. I believe the next ten years will see a shift back to ‘the network as the inventory’ for real-time, utilisation, and performance data. Business-critical OSS solutions will then comprise solutions driven by ‘big data’ network analytics in order to make sense of this data and deliver actionable, proactive insight in to the network state and customer experience.

Where are all the New OSS Projects Going to Come From?

The OSS market has changed... Part 1 of 4. This series of four blog posts is a synopsis of my contribution to a discussion I chipped in to on LinkedIn’s ‘OSS Gurus’ group.

Where are all the new OSS projects going to come from? I don’t think there'll be many 90s/00s style OSS transformation projects, even with the changes occurring with LTE (tech change has, in the past, driven OSS change). But I do believe there's a big opportunity for niche solutions, from start-ups or otherwise, to address the challenges/opportunities of the more modest changes we're seeing in telco during the 2010s.

The economics of telco make such innovation difficult though, particularly for start-ups: 18 month sales cycles; lack of willingness to adopt term license or SaaS license models; and the demand for CoTS products that are highly customised to legacy business processes.

No OSS product will be successful without a compelling need by the telco. LTE might drive a need for some solutions, as will fixed-line operators rolling out WiFi, the steady migration of backhaul and transport to IP/Ethernet over OTN/DWDM, and the eventual shift of telcos from offering bit-pipes to differentiated services and content delivery platforms.

The opportunity for OSS lies not in managing the bits n’ bytes of the latest network technology, but in enabling telcos to offer viable, profitable products on a cost-effective network.

NEC to Acquire Convergys BSS Software

NEC announced their intention to acquire Convergys' BSS assets for $449million (USD).

The acquisition is intended to compliment NEC's OSS products (from the Netcracker and Subex purchases) and their bag of BSS bits they have been using to promote the Netcracker brand as a complete OSS/BSS solution. Neither NEC nor Netcracker had a great deal of brand-recognition in the BSS space, but Convergys will bring this in spades. NEC/Netcracker will be in a far better position to compete directly with Oracle and Amdocs, both of which have a complete OSS/BSS suite.

The deal will include Convergys' Information Management business, home of their BSS products. This leaves Convergys to focus on their less telco-specific relationship management and call centre software.

By the way, that's $449m from 2011 revenues of $329m by the IM business, an unexciting 1.36 multiplier.

A bit more about this story in NEC's own press release and some more business figures in the Light Reading story (where they amusingly persist in trying to get us all to call OSS/BSS software 'SPIT').

Zoomlar Beta Launched

Zoomlar - Easy-to-use spreadsheet analytics for forecasting trends and predicting outcomes.

A bit of news from my day job: I have been busy over the last three months preparing for the launch of a re-branded, re-packaged product aimed at SME, academics, government and utilities. It takes the clever software we've been using in telco OSS products and makes it applicable to a far wider set of users.

Zoomlar allows you to analyse small or very large amounts of data and solve difficult forecasting problems faster and more accurately. It can be applied to a variety of problems, that fall in to "time-based trending" or "what-if prediction" analysis.

Zoomlar has been designed to be very easy to use. It requires no specialist knowledge and no prior knowledge of analytics, statistics or algorithms. Anyone who can use a Microsoft Excel spreadsheet should be able to use Zoomlar in less than 30 minutes.

For the next few months we're running a beta programme. You can use Zoomlar for free during this time, and get online support.

Take a look. If you come up with any interesting use-cases, drop me a line.

Comptel Acquires Xtract

More evidence that OSS/BSS vendors and their customers are starting to really appreciate data analytics.

Comptel has announced its acquisition of Xtract (here), supplier of analytics products for telco (and retail/banking too).

The price paid by Comptel is modest to say the least: 3.1 million Euros paid, for Xtract's last year revenues of 2.4 million Euros. So not even a 1.5x multiplier. Either Xtract are motivated sellers, or it suggests that much of the analytics success is services-driven, rather than being off-the-shelf product sales; services companies usually achieve lower exit multipliers than product companies.

The deal is good news. Innovative OSS/BSS vendors are recognising that they can offer smarter solutions with analytics.

Most OSS/BSS products that grew out of the telco boom of the 90s/00s are process-oriented. They sit on a wealth of valuable data, using it to execute jobs, rules and tasks to do something useful (provision a DSL line, calculate a bill, etc). But all that data is ripe for analytics, either to perform additional tasks (like Xtract's customer churn prediction and service upsell suggestions) or to do existing tasks an order of magnitude better (like Aria Networks re-optimising the network for greater service resilience).

This is where the big OSS/BSS vendors are going to be differentiating their product range. Analytics is a natural bolt-on to existing OSS/BSS data sources, making it an attractive up-sell opportunity for vendors and a low-risk project for customers.  We can expect to see more acquisitions and product development in this space over then next couple of years.

 

Ericordia/Telicsson - Ericsson Completes Purchase of Telcordia

Back in June, after several month of speculation, news broke that it was Ericsson that intended to buy Telcordia. That deal has now completed, just a few weeks later than expected.

Celebratory Ericsson-branded mugs all-round for the Telcordia team.

According to the press release, Telcordia was aquired for $1.15billion for annual revenues of $739m.

Telcordia now joins Ericsson's Business Unit Mulimedia.

Comverse up for Sale

Is Comverse looking for a buyer of their BSS software?

Globes has a story about Comverse looking to sell their telco billing unit. This comes a few days after a number of people in the industry received invitations from investment bankers to discuss the potential acquisition of a "premier billing and OSS software company... in the US".

Globes suggests Oracle may be interested, which makes a lot of sense. More telco software to sell more hardware and software infrastructure alongside their OSS offerings.

NetBoss Acquires Sidonis

News from Management World Americas last week: NetBoss has completed acquisition of Sidonis, the start-up doing clever stuff in service assurance.

Sidonis are a small team (I could count them on my fingers) based in Bath, UK, who have developed something that NetBoss, and many other SA players, lack. Most SA solutions focus on collecting, filtering and prioritising network alarms, which is nice but only half the information that network ops need. It says here:

"Sidonis technologies include the industries most advanced software applications for determining both the impact on services and the root cause of failures in large, complex networks spanning multiple technology domains."

In the past many SA vendors had to integrate with relatively expensive and heavy-weight inventory solutions to 'enrich' alarms with service information or perform topology-aware analysis like root-cause determination. With this acquisition, NetBoss is building out a more complete, stand-alone SA solution with less dependencies on third-parties and reduced cost of integration.

The official word about NetBoss acquiring Sidonis is here.

 

iCloud and The Death of the Asynchronous Network Model

Something else to blame on the over-the-top service providers (yes, those guys who actually make money from comms). New Cloud services are breaking the content-consumption model that ADSL, and network planning in general, is predicated on.

iCloud is not the first Cloud service to be sold to consumers. But, like so many things from Apple, it may well be the first to be adopted en masse, changing the way a large number of broadband subscribers use their bandwidth.

Various flavours of ADSL exist, supplying different levels of bandwidths to users. What they have in common is that they are asynchronous (The A in ADSL) providing significantly more download bandwidth to the customer compared with upload bandwidth to the Internet. This ratio can be quite large; a modern ADSL service from my UK-based supplier is 40Mbs down, 2Mbs up.

That’s great when you spend most of your time browsing the web or downloading video-on-demand. Until recently, uploading data was pretty rare; Just a few Word docs attached to emails or a few dozen photos uploaded to Flickr.

Cloud services, particularly back-up or file-sharing services like iCloud or Dropbox, make much greater demands on the upload bandwidth. The moment someone buys iCloud or Dropbox a heck of a lot of files need to be upload to ‘the Cloud’ for storage/back-up/sharing. All those MP3s, home-made videos, big PowerPoint slides... ADSL isn’t great at offering the customer good upload speeds, which may be something of a disappointment. More critically to the telco, their entire model for understanding customer bandwidth demand and their resultant network architecture breaks.

The network has been architected around ADSL. Recent investment in new architectures, like CDN, are aimed at alleviating performance and capacity issues for the old data-consumption usage pattern.

When iCloud takes off, pretty much every file, of every customer, will need to be uploaded to The Cloud. If 20% of your customers buy iCloud at the same time, thanks to an Apple marketing push, would your network be ready? Do you meter or charge for upload traffic? Do you have a service plan for consumers who sync a lot of data?

Remember how much of a hassle illegal peer-to-peer file sharing was for telcos a few years back? iCloud is about to legitimise comparable file-transfer traffic loads, with none of the efficiencies inherent in P2P, promoted by millions of marketing dollars...

Some More Tribold Investment

Yesterday, Tribold announced a funding round of Some Money for strategic investment purposes.

This follows $11 million back in January 2009 from Intel, DFJ Esprit and my local VC, Eden Ventures.

This time the value of the investment has not been disclosed. Nor has Eden contributed to this round, as far as I can tell from the PR.

Nic Brisbourne (who’s blog you really should check out), partner at DFJ Esprit, added: “As an existing investor in Tribold, we are confident of the company, its technology and its management. The involvement of Intel Capital also confirms Tribold’s potential for us. We’re also happy to have had the opportunity to increase our own investment in the company.”

Cisco Buys Bits of Comptel

It says here that Comptel is selling parts of its OSS portfolio to Cisco.

This is not the clear-cut Big-Vendor-Buys-OSS-Company that we are familiar with. No, in this case Cisco is acquiring a specific part of Comptel's OSS product portfolio: The Axioss suite picked up as part of Comptel's own acquisition of UK-based Axiom back in 2008.

Furthermore, some of the Axioss staff will move as part of the deal, but Comptel will remain in the OSS market with its other products and, perhaps most interestingly, will continue to offer support to their existing Axioss customers.

So, from Cisco's perspective it's a technology purchase, giving them the freedom to employ it however they wish, without the ball-and-chain of existing support contracts.

Cisco has a history of investing in OSS to augment the core iOS network management system. This investment has been both in-house development and acquisition. To my knowledge this is a first step-up to 'proper' fulfilment OSS, as opposed to stand-alone engineering tools.

Expect more of this in the next twenty four months: Anyone at recent events like TMW Dublin will have noticed how the traditional OSS vendors booths are getting smaller while hardware vendors like Huawei are Going Big.

Hardware vendors <3 OSS right now. This, along with the Ericordia deal, is the start of a second wave of OSS deals.

Update: The purchase price is 21.3 million Euros. Not bad in the current economy. Comptel acquired Axiom for 'up to' 21.6 million Euros, based on another slightly unusual deal that tied the price to their business performance for the year.