Summary

  • The TCS letter of intent and Tejas’s statement that supply began move the BSNL 4G RAN work beyond a bare announcement, but they do not establish delivery completion, acceptance, installation, integration or on-air service at 18,865 sites.
  • The government’s one-lakh-site snapshots are a useful dated operating baseline, not a reconciliation of this add-on order; the missing bridge is a public, auditable chain from procurement to working service.

A supply start is a milestone, not a coverage map

A network order becomes politically legible as soon as a large number can be repeated. “18,865 add-on sites” sounds like a count of places where customers will soon see a stronger signal. But in the 16 October results release, Tejas Networks describes something narrower: it says that in Q2FY27 it received a Letter of Intent from Tata Consultancy Services to supply 4G radio access network equipment for 18,865 “add-on sites” on Bharat Sanchar Nigam Ltd.’s pan-India network, and that it started supplies for the order during that quarter.

That disclosure is meaningful. A named buyer-side integrator, a defined equipment scope and a reported start to supply are more concrete than a generic technology ambition. They indicate movement through a procurement chain. They do not, on their own, tell a reader how many units left the factory, how many arrived at a site, how many passed acceptance, how many were installed or integrated, or how many are carrying traffic. The release does not publish the order value, a site list, delivery schedule, acceptance terms or cash receipts.

The distinction matters because public discussion often collapses procurement and service into one verb: “roll out”. A letter of intent is a commercial instrument; supply is a logistics event; installation is a field activity; integration is a network change; “on-air” is an operating status. Each has a different accountable party, evidence record and failure mode. Reporting the first two as if they establish the last one turns an input into an outcome.

The government baseline answers a different question

The Ministry of Communications’ 11 March 2026 reply said 97,906 of BSNL’s planned one lakh indigenous 4G sites had been installed and 96,103 were on-air as of 28 February. A 5 February reply gave the earlier snapshot: 97,672 installed and 95,511 on-air as of 15 January. Those dated figures show that installation and operational status are already tracked separately at program level. Between the two published snapshots, the totals rose by 234 installed sites and 592 on-air sites. That arithmetic is not evidence that the Tejas order caused those changes, nor does it identify any of those sites as part of the 18,865 “add-on” scope.

“Add-on” should remain in quotation marks until the contract scope is documented. The word could refer to equipment added to an existing network footprint, a commercial scope defined against a prior order, or another procurement distinction. The available release does not explain which. It is therefore unsafe to add 18,865 to the one-lakh target, subtract it from that target, or call it 18,865 newly connected places. The two numbers may describe different layers of a programme.

A useful public reconciliation would preserve the terminology in the contract while mapping each milestone to a stable site identifier: ordered, shipped, received, accepted, installed, integrated, on-air and, where measurable, carrying traffic. If the new equipment upgrades an existing site, the record should show the before-and-after configuration rather than count the location twice. If a site is a genuinely new footprint, it should be identified as such. This is ordinary operational accounting, not a demand for disclosure of sensitive network topology.

Company growth and network delivery are separate ledgers

Tejas’s results provide a second reason to watch conversion carefully, without treating financial pressure as proof of failure. The company reported Q2FY27 revenue of ₹406 crore, compared with ₹262 crore a year earlier, while its quarterly net loss narrowed to ₹207 crore from ₹307 crore. For the first half, it reported revenue of ₹808 crore and a net loss of ₹409 crore, versus ₹464 crore and ₹501 crore in the prior-year period. Its CFO said the order book stood at ₹2,315 crore at the end of Q2, against ₹1,529 crore at the end of Q1, and disclosed net debt of ₹4,384 crore, gross debt of ₹4,835 crore and cash of ₹451 crore.

These figures describe the supplier’s broader business, not the economics or performance of the BSNL order. The release does not assign a value to the LOI or say how much of the order book it represents. Nor does an expanding order book establish recognized revenue, accepted equipment or cash collection. Investors and public-service operators need separate ledgers: one for commercial backlog and cash conversion, another for physical network delivery and service readiness.

For BSNL, the operational question is whether equipment procurement translates into usable 4G capacity where subscribers need it. The government’s snapshots provide a national count but not a site-level service record. A site can be installed yet not on-air; an on-air site can have limited coverage or capacity; a coverage assertion does not by itself describe call completion, data performance or congestion. The article does not claim that any of these weaknesses has occurred at the 18,865 sites. It identifies the measurements needed to determine the outcome.

A public evidence chain can protect all sides

The roles in the chain are distributed. TCS is named as the source of the LOI; Tejas is the equipment supplier; BSNL operates the mobile network; the Ministry communicates the national programme status. A credible dashboard would identify which party attests to each transition and timestamp it. Supplier dispatch records can substantiate shipment. Receiver acknowledgements and acceptance certificates can substantiate receipt and compliance. Site work orders can substantiate installation. Network-management records can substantiate integration and on-air status.

Independent drive tests, complaint data or service metrics can address user experience.

No single number can substitute for this chain. The count of installed sites is not the count on-air. The count on-air is not the number of sites that have passed a published quality threshold. An order for equipment is not a statement of coverage. This is especially important when an initiative is described as indigenous: domestic design and manufacture are significant industrial facts, but they do not exempt a deployment from service-level accounting.

A sensible update would report progress in cohorts rather than promise a date unsupported by evidence: for example, how many sites have been accepted, installed and brought on-air during the period, with the denominator and definition held constant. Where disclosure must be aggregated for security, the method can still be public and the underlying audit can still be available to the operator, government and independent reviewers.

The standard is not to presume that the order will fail. It is to refuse to treat a commercial milestone as proof of a public-service result. Tejas has reported that supply started. The next evidence should show what was supplied, what BSNL accepted, what entered operation and what changed for users. Until then, 18,865 is a procurement scope, not a verified count of new live services.

Sources