How are we doing? A telecom CXO wants to know?

Yes, they know it. OR do they? At least they think they know, they claim they know, they believe they know, and they better believe OR who are they fooling ? So they must know, and they are convinced they know. BUT there is much more to know, and here is where Data Analytics chips in, not to insult them or downplay them, but to empower them. They know something and they must know much more… there is no end to knowing in Telecom billing.
Every operative level location head or regional head does check at the end of the day:
1. Daily Acquisition report of new connections on PRE-PAID and on POST-PAID.
2. Daily Churn report of the old connections on PRE-PAID and on POST-PAID.
3. The Daily MNP port-in report on how many subscribers are coming in from other Vendors with the first number generated by a different Vendor.
4. The Daily MNP port-out report subscribers are going out with first numbers generated in the company and are now to other vendors.
These four reports are coming separately, the data sets do not have any single field common and therefore becomes difficult to fathom the meaning at the elementary level. So the CXOs actually concentrate on the aggregated figure on a daily basis, and then what they do at their own kitty level is to have a triplet of and similarly . And then they make their own decisions.
The Challenge however is to relate the apparently un-relatable :-
1. {day, prepaid connection, pre-paid disconnection}
2. {day, post paid connection, post paid disconnection}

A higher level of challenge would be to combine the Daily MNP port-in with #1 and Daily MNP port-out with #2, that means the data sets should have the quadruplet
1. {day, pre paid connection, MNP port-in , post paid connection}
2. {day, post-paid disconnection, MNP port-out, post paid disconnection}
These two are like credit and debit of a daily run P/L account or income-expenditure statement.
A derived level would be to have a triplet like
{day, total new acquisition, total churn or disconnection}

Just to figure this out a CXO needs to have many datasets to run through different levels of simulation runs. A Data Analyst tool should help a CXO in bringing together these apparently un-relatabale data sets and compare them through bringing in the columns and then aggregating them.
On a Weekly basis an analyst would have 7 rows and can put them in normal spreadsheets and them run a dashboard like presentation on them. What if the CXO wants to have partial data or the same data separated in terms of region, separated in terms of Vendors[ to or from], in terms of even time period of the day? Then of course along with the above structure one needs to capture the region, the timestamp of transaction, the “other” Vendor. Generally this is not so difficult to capture with the raw data because the raw data comes along with that. One has to pre-plan that capture. The little challenge however is therefore not to go straight to a day wise aggregation but to keep the raw data and categorize them based on the dimensions. Here the operation [pre-paid connection/disconnection, or MNP port-in/port-out] is the fact and the region, vendor being the dimension. Time stamp acts both as a fact and a dimension due to its need. The specific Data Analyst tool must have the capability to use the same column or measure once as a fact and immediately after as a dimension, which basically means a kind of measuring one aggregate with the granularity of the same, that way a set of values within the column may be aggregated against a range of the same values. Here the industry needs Data Analysis tools a little different from the normal ERP based DSS tools.
Return of Investment is all that CXOs make themselves concerned in, and still they might fall quite a way short of the value of the investment. The price of a return on investment gives a static picture, the value shows the direction in which the future prices would show up. A simple connection/disconnection aggregate and their net value gives a daily return on the investment. We actually get a OPEX/CAPEX ratio apportioned on a daily basis, that can be aggregated over and the pay back calculated, but to know the future of the ROI – if the return is going to come acceleratingly or deceleratingly remains in the dark. One can surmise at best by looking at the trend value- this is simply speculating on the future based on the past values- this is quite a good practice indeed and still is not assertively comprehensive. The measurement of VALUE of a service actually gives a better idea about how a service-product would fare. For example if it is found that a particular Value Added Service [VAS] is attracting more customers then even by reducing the price of the particular VAS the market can be fast captured as opposed to other VASs. This way of market penetration actually would bring in more ROI in the future as this number would be diverging. A diverging measure may be fine tuned with respect to a specifically filtered dimension and introduced in the market with a competitively low price to capture the market and a good ROI is possible to be garnered.
The opposite strategy would be based on the hypothesis of consumer behaviour that a VAS is essentially and meant to be short lived. With that in mind the highlighting of that particular VAS as an add-on and jacking the price to quite a high mark would be the skimming of the market till such time the competition catches it up, with increasing competition the prices may be reduced and made competitive. A measure of Value of ROI therefore is the summed up return during the life time of VAS [Till the VAS does not give any added benefit or attraction from the user]. In fact there is a new measure that is becoming popular which is the VALUE [thus obtained] / average of the fleeting prices of the VAS] and if the ratio is positive the campaign is VALUE-WISE. These types of analysis can only be done by newer types of DATA ANALYSIS tools where absolutely apparently different data sets can be brought into comparison and then mapped in a dashboard.
What if I now assert and posit that IDEAL-ANLYTICS is one such unique tool in the market that was conceived with these challenges to cope with – that is why it is catching up in the market place as something very smart and trendy.. Watch out for even more surprises and thrills to come – down the time line..