Now that I have your attention, I’ll move to the actual topic. Ever since I’ve started heading up the Business Assurance product line at Subex, one question has inevitably been asked by almost everyone I meet: what will RA products look like <insert preferred number> years from now?
I find that letting your imagination run wild actually helps you find the perfect, ideal solution. From that point on, it’s a matter of identifying the key roadblocks and challenges and trying to overcome each one – or finding an insurmountable one, which means we go back to redefining the end-state.
In my thought journeys so far, I have three candidates that I believe are the greatest things since sliced bread (if sliced bread was business assurance). Each of the following options are based on three distinct challenges which I would love to overcome, namely:
- control/influence over the whole BSS/OSS roadmap,
- a genius level understanding of cyber-ecosystems, and
- my personal favorite, the complete trust of all the subscribers of my telco customer.
Do feel free to post in the comments which one is your personal favorite, or better yet, your own thought journey. Now onto the great reveal! My top three candidates for the utopian future of RA systems are:
1. The Self-Correcting Telecom Stack
In a nutshell, this idea involves OSS/BSS systems which talk to each other on an ongoing basis. So if a customer of the network frequently suffers dropped calls, they automatically get upgraded to a priority slice with the CRM sending across a message that says: “Sorry, my bad! No more dropped calls for you today!”.
Another fun example is if a service deteriorates from 100 Mbps to 25 Mbps for some customer, then it automatically sends a message to the billing platform to charge at a lowered rate until normal service resumes, with the message to the customer being: “Sorry John, my network planning team slipped up in terms of guesstimating the actual peak capacity. We’ve reduced your billing rate and given you 20GB top-up for each day you see this”.
I’m talking about network switches and charging platforms and partner settlement systems actually talking to each other in terms of assuring integrity and correctness. This approach would truly bring back the “customer” to the term “customer-centric”! There is only one hitch: to achieve this today I would need to be bigger than Huawei and Ericsson combined.
2. Botnet-Based Auto-Correction
Let’s assume for the moment that my being bigger than Huawei takes a little more effort than keeping your new year resolutions. In that scenario what else can we do? I like to take a leaf out of the IT operations world by using intelligent agents on all network elements. We can call them bots, because they do more than monitoring key KPIs. The idea is similar to the previous one, except instead of each element having a built-in communication heartbeat protocol, we have a botnet deployed across all the operations.
What happens in the aforesaid botnet? Every single update to key configuration or profile parameters is communicated in byte sized messages through the botnet to all relevant up/down-stream elements. For example, in case of a dynamic tariff plan change, the bot at the provisioning element communicates with the bot residing on the charging element as well as the bot on the network side. The charging system bot verifies that for each particular subscriber, the rate plan has been changed and the network side bot then confirms that the QoS allocation for that subscriber has in fact been upgraded.
We can also have PCEF/PCRF bots to verify aspects of the fair usage policy, max up/download limits etc. What a wonderful world we would live in if mismatches can only occur when the botnet is down!
I admit that building an intelligent, communicative and corrective botnet is not how I currently spend my Sundays. Is there no way forward for BA software? Fret not, there is one more idea up my sleeve. What if, with the consent of all the subscribers in an operator environment, we send BA agents on the phones of our subscribers?
The BA agent essentially monitors usage on the network, up/download volumes, services available, billing/charging profile hashmaps and essentially any network related faults. Imagine all these device-bots (or as I like to call it, the Dendrones) then communicate with the central server (or as I like to call it, the BRAIN). Discrepancies are identified on a per device/profile/event basis, and corrective actions are executed in event-time.
This approach would show telcos how they appear from the outside, which may be sorely missing in most BA departments today. Most importantly, this would set aside the complexities of the network and focus on the deliverable: predictable, consistent and quality service to our subscribers. Tres MAGNIFIQUE!!!
Of course, these are the musings of someone who wants BA tooling to become simple enough to be comprehensive, and yet light enough to allow for dynamic expansion with minimal time to market for new controls. Do let me know in the comment section which option makes sense to you, or conversely, let’s speak about why something wouldn’t work. Either way this is a topic which I believe has crossed all our minds at some point or the other. Collectively, we might find a new thought journey which makes sense to all of us.
This article was originally published on LinkedIn; see here. It has been reproduced with the author’s permission.