Spider Net and the Microeconomics of Bangladesh’s Small-ISP Layer: Route Visibility, Retail Trust, and Survivability Under Upstream Dependence

Spider Net and the Microeconomics of Bangladesh’s Small-ISP Layer: Route Visibility, Retail Trust, and Survivability Under Upstream Dependence

Spider Net is not a company that can be understood through a conventional corporate profile. Its public website contributes little. Its brand footprint is modest. Its corporate-control record is incomplete. Yet its network traces are unusually instructive. Spider Net appears in APNIC registration records as a Bangladesh local internet operator associated with ORG-SN17-AP, AS138651, a Chattogram address, a /22 IPv4 allocation, and an IPv6 /32 allocation. Its visible routing is smaller than its registered address inventory. Its retail-control evidence is stronger in a customer app than in a marketing site. Its relationship graph points toward the practical dependency chain of Bangladesh broadband: local access license, APNIC resources, upstream transit, NTTN backhaul, IIG connectivity, payment rails, field repair, and customer trust.

The central finding is that Spider Net looks economically less like a standalone “telecom company” and more like a compact bundle of scarce local rights and operational relationships. The bundle contains a BTRC upazila/thana ISP identity in Chattogram Sadar, an APNIC-registered network identity, a small routed IPv4 edge, a dormant or underused IPv6 resource, a mobile-app billing and support channel, and a local office/contact surface. That bundle can produce real cash flow even when the company has little public media presence. It can also become fragile if any one of several non-retail dependencies fails: route sponsorship, IIG terms, leased transmission, license renewal, customer support quality, payment-channel reliability, or field repair density.

The evidence is thin on formal ownership, audited financials, litigation, financing, and management. It is stronger on network registration, public BGP visibility, licensing geography, customer-control tooling, and wholesale-dependency structure. That asymmetry is the point. In Bangladesh’s fragmented broadband market, the economic substance of many small networks is not visible in press releases or investor filings. It is visible in who originates their prefixes, which upstreams see them, whether RPKI is valid, whether customers can pay by bKash and open support tickets, whether the operator is listed in ISP association and BTRC records, and whether the license category confines the operator to a small geography.

Identity: a small Chattogram ISP, not a clean corporate group

The most defensible canonical identity is “Spider Net,” a Bangladesh network operator and local ISP identity tied to Chattogram. APNIC’s organisation record for ORG-SN17-AP gives the organisation name as Spider Net, the country as Bangladesh, the address as “7 No. Nabab Sirajjuddoula Road, Kotowali,” the phone number as +8801837686343, and the contact email as info@spidernetctg.com. The APNIC record classifies the organisation as a local internet registry, not as a corporate group with subsidiaries or a public-company reporting perimeter.

The same identity appears in BTRC’s upazila/thana ISP license list. The official BTRC PDF row names “Spider Net,” places it in Chattogram Sadar, gives the address as “7 No. Nabab Sirajdulla Road, Kotowali, Chattagram,” and lists license number 14.32.0000.702.47.043.22.376. The row also shows license validity to 31 July 2023 and next renewal date 1 August 2023, inside a document titled as an upazila/thana license list “as on 18-12-2024.” That creates a renewal-status ambiguity rather than a clean active-license conclusion: Spider Net is clearly in the official licensing universe, but the public row reviewed does not, by itself, prove that the license was current after the listed validity date.

The ISP Association of Bangladesh member listing also contains a “Spider Net” profile, membership number B-074, with BTRC license category “Upazila/Thana,” an Anderkilla/Chittagong address, the email azizuddin2007@gmail.com, and the same mobile number that appears in APNIC records. The same ISPAB page also lists separate entities with confusingly similar names, including Spider Mesh in Bogura and Spider Networks in Kuakata/Patuakhali. This matters because the public market contains multiple “Spider” broadband labels in Bangladesh, and only the Chattogram/Kotowali/AS138651 evidence ties to the target studied here.

No public source reviewed establishes that Spider Net is a subsidiary of Coronet Corporation Limited, Orange Communication, Orange Bangladesh, Exabyte, Plusnet, Windstream, or any other network operator. The relationship clues are routing-policy and network-market clues, not corporate-control proof. BGP.tools shows AS138651 in several AS-SET contexts, including Coronet, Orange, Windstream, Plusnet, iTel, Summit, and other Bangladesh network sets. Such AS-SET memberships can mean that an upstream or route-policy maintainer is prepared to accept and propagate a customer’s prefixes; they do not prove ownership or merger.

This distinction is economically important. A local ISP’s operating control may be independent even when its route policy is embedded in larger providers’ AS-SETs. Conversely, a small ISP may be commercially dependent on a larger IIG, aggregator, or neighboring ISP without any public M&A transaction. In Spider Net’s case, the public record proves a local network identity and suggests several wholesale-routing relationships, but it does not prove a parent company.

What the registered network proves

Spider Net’s APNIC address records show a meaningful but small resource base. The IPv4 allocation is 103.135.136.0–103.135.139.255, a /22 containing 1,024 IPv4 addresses, with netname SPIDERNET-BD, organisation ORG-SN17-AP, country Bangladesh, and status “ALLOCATED PORTABLE.” The record points to MAINT-SPIDERNET-BD and the IRT-SPIDERNET-BD abuse contact.

The IPv6 allocation is 2404:7ec0::/32, also under SPIDERNET-BD and ORG-SN17-AP. APNIC’s IPv6 record gives the same Chattogram/Kotowali address context and an abuse validation date in January 2026. The existence of a /32 IPv6 allocation shows that Spider Net has the registry capability to operate IPv6 at scale, at least on paper. It does not prove active IPv6 retail service.

The ASN is AS138651, named SPIDERNET-AS-AP. BGP.tools reports Spider Net as operating in Bangladesh, originating two IPv4 prefixes and no IPv6 prefixes. The two visible originated prefixes are 103.135.136.0/24 and 103.135.137.0/24, each shown as RPKI-valid. The same page reports two observed upstreams: AS137526 Plusnet Inc and AS139009 Windstream Communication Limited. Hurricane Electric’s BGP Toolkit independently shows AS138651 originating two IPv4 prefixes, no IPv6 prefixes, two observed BGP peers, 512 originated IPv4 addresses, and RPKI-valid originated IPv4 routes.

The gap between the APNIC /22 and the currently observed AS138651-originated /24s is one of the most important economic facts in the record. Spider Net holds or is registered for 1,024 IPv4 addresses, but AS138651 is visibly originating 512 IPv4 addresses through its own ASN. BGP.tools shows 103.135.138.0/24 originated by AS150774 EXABYTE LTD and described as Spider Net, while AS150178 EXABYTE LTD is shown originating 103.135.139.0/24, also described as Spider Net. APNIC-derived route-object material for 103.135.138.0/24 also shows route objects pointing to Exabyte-linked ASNs, including AS150178 and AS150774, with descriptions that include EXABYTE LTD and Md. Aziz Uddin at the same Kotowali address.

There are three economically plausible interpretations. First, Spider Net may operate the 136/24 and 137/24 blocks directly while allowing two other /24s to be routed by, leased through, or technically managed with Exabyte-related networks. Second, a larger upstream or aggregator may be carrying portions of Spider Net’s address space for customer, NAT, backup, or transition purposes. Third, the allocation may reflect legacy registry ownership while the real traffic-serving role of parts of the /22 has moved to other ASNs. None of these interpretations can be proven from BGP alone. But all of them point to the same economic structure: the address asset is separable from the retail brand, and route control can migrate before legal or branding records do.

The absence of IPv6 announcements is also revealing. A /32 IPv6 allocation is a large technical asset for a small ISP. Yet both BGP.tools and Hurricane Electric show no IPv6 prefixes originated by AS138651. This suggests either limited IPv6 deployment, no public IPv6 route visibility from AS138651, or IPv6 service carried in some other way not visible through the reviewed AS. Economically, the consequence is that Spider Net likely remains dependent on scarce IPv4 and carrier-grade NAT for much of its retail service, unless IPv6 is deployed privately or through another operator. For a small ISP, IPv6 is not merely a technical modernization issue. It affects customer support load, gaming and remote-work complaints, static-IP monetization, abuse attribution, and upstream bargaining.

Route visibility as a form of market capital

For a small ISP, an ASN is partly a technical identifier and partly a market credential. AS138651 makes Spider Net visible to upstreams, route collectors, abuse desks, and enterprise customers. It also allows the operator to maintain portable address space, originate routes, and implement RPKI. This produces a modest but real form of bargaining capital. A reseller without its own ASN and portable space can be cut off, renumbered, or absorbed more easily. A small ISP with portable addresses and valid RPKI can switch or multi-home upstreams with less disruption, at least in theory.

Spider Net’s current route visibility is narrow but not trivial. Two /24s are visible from AS138651, both RPKI-valid, and no downstream ASNs are reported by IPinfo. IPinfo classifies the ASN type as ISP, reports 512 IPv4 addresses and no IPv6 addresses under AS138651, shows no hosted domains on the ASN, and lists Plusnet and Windstream as peers/upstreams. BGP.tools and Hurricane Electric converge on the same basic picture: a small stub-like Bangladesh ISP with two IPv4 BGP neighbors, no visible IPv6 origination, and no public downstream cone.

This matters because route visibility disciplines the commercial model. Spider Net does not appear to be a wholesale transit provider. It is not using its ASN to build a large downstream cone. It is not visibly hosting websites at scale. IPinfo reports zero hosted domains on AS138651, and Host.io shows spidernetctg.com itself resolving to an Akamai Connected Cloud/Linode-style shared hosting environment rather than being hosted on Spider Net’s own ASN. The network’s economic center is therefore unlikely to be datacenter hosting, transit resale, or enterprise cloud connectivity. It is more likely fixed access: home broadband, small-business internet, local packages, bill collection, and field service.

The RPKI-valid status of the originated prefixes is a positive technical signal. In a market of many small broadband labels, RPKI validity reduces route-hijack risk and indicates that either Spider Net or its technical counterparty maintains the route-origin authorization process. It is not a guarantee of service quality. It is a signal that the operator is not merely a street-level cable reseller with no registry competence. That distinction affects upstream trust and customer trust differently. Upstreams care that routes are valid and contacts are reachable. Retail customers care that the line works during peak hours and that someone answers when the fiber is cut.

The customer-control surface is stronger than the website

Spider Net’s website is not a strong public channel. Host.io identifies spidernetctg.com as resolving to 172.234.24.211 under AS63949 Akamai Connected Cloud, with an OpenResty “Redirecting…” title and no meaningful backlink footprint. A direct marketing-site interpretation would make the company look inactive or underdeveloped. That would be misleading if taken alone.

The more informative customer-channel evidence is the Android app “Spider Net,” developed by SoftifyBD. The Google Play listing describes a utility that lets customers see downloaded and uploaded data usage since last connectivity to the server, request package changes, test router connectivity, open support tickets, message the technical team, pay monthly bills through bKash without extra charge, view payment history, receive disruption/offers/news notifications, and reconnect automatically after nonpayment disconnection once payment is made. The listing shows an update date of 29 December 2025 and 50+ downloads.

The app’s functions reveal the real operating problem of a local ISP. Spider Net’s economics depend on reducing the cost of small transactions: bill reminders, payment reconciliation, nonpayment disconnection and reconnection, support-ticket triage, package-change handling, and basic customer diagnostics. A mobile app does not prove scale, but it proves the operator has invested in a retail control plane. The most commercially significant app feature is not branding; it is the automatic reconnection mechanism after bKash payment. That feature compresses the cash-conversion cycle. It reduces field visits for collections. It also turns prepaid or quasi-prepaid disconnection into a lower-friction enforcement tool.

The app’s data-safety disclosure also states that data is not encrypted in transit and cannot be deleted, according to the Google Play listing. For a small ISP, this is not a trivial footnote. Customer self-service systems create trust when they work, but they also enlarge the operator’s digital risk surface. Billing records, outage notices, router connectivity checks, and support tickets can become sources of privacy and reputational risk if poorly secured. Small networks often modernize billing before they professionalize security governance; Spider Net’s app evidence should be read in that context.

The channel evidence suggests that Spider Net’s customer relationships are local and operational rather than media-led. Customers likely discover the operator through neighborhood availability, technician referral, building-level wiring, word of mouth, local Facebook or phone contact, and the practical fact that the line can be installed quickly. The app then supports retention and collections. In such a market, a weak website is not fatal. It is a missed trust signal for enterprise customers, but retail broadband in dense neighborhoods is often sold by field presence and price rather than by search-engine credibility.

Geographic footprint and operating layer

Spider Net’s strongest geography is Chattogram. APNIC gives Kotowali/Chittagong. BTRC’s upazila/thana list places the licensee in Chattogram Sadar. ISPAB gives an Anderkilla/Chittagong address. IP geolocation vendors also place sample Spider Net IPs in Chittagong, though IP geolocation should not be treated as precise operating-footprint evidence. IPGeolocation.io places 103.135.137.104 in Chittagong and associates it with AS138651 and Spider Net. AbuseIPDB’s WHOIS/geolocation view for 103.135.138.197 labels the usage type as fixed-line ISP and associates the domain with spidernetctg.com, but it places the sample IP in Cox’s Bazar/Chittagong; that should be treated as a third-party geolocation signal, not a confirmed service-area map.

The BTRC license category is more economically important than the exact geolocation of individual IP addresses. Bangladesh’s ISP guidelines define licensing categories including nationwide, divisional, district, and upazila/thana. They also restrict last-mile implementation: a licensee may generally provide last-mile internet service within about three kilometers in metropolitan areas and six kilometers elsewhere, subject to local-authority orders and BTRC instructions. This regulatory geography pushes small ISPs toward dense, local access economics. It is difficult for an upazila/thana operator to become a national brand without new licensing, acquisitions, or wholesale/partner arrangements. The valuable question becomes not “How many cities does Spider Net serve?” but “How many profitable buildings, lanes, small businesses, and household clusters can it serve within its authorized and physically reachable territory?”

The operating layer also appears to be retail access rather than wholesale transit. There is no evidence of downstream AS customers. There is no evidence of a PeeringDB public profile for AS138651 from the reviewed searches. IPinfo reports no hosted domains. The route table shows a small access network with upstream dependence, not a regional backbone. The app shows customer billing and support, not wholesale provisioning. The ISPAB and BTRC categories show local access licensing, not IIG or NTTN licensing. Taken together, these facts place Spider Net in the last-mile and near-last-mile economics of Bangladesh broadband.

Bangladesh’s regulatory stack turns every local ISP into a dependency manager

Bangladesh’s broadband rules structurally separate local access from upstream layers. The 2020 ISP regulatory and licensing guidelines say an ISP licensee may provide internet, data communication, and IP-based services, and may lease or sub-lease transmission network from licensed NTTN operators. The guidelines also say ISP licensees must connect to licensed International Internet Gateways for leased internet bandwidth and connect to National Internet Exchange facilities for domestic inter-operator traffic.

That framework makes upstream dependence a designed feature, not an accident. A local ISP such as Spider Net cannot simply become a vertically integrated national fiber, international gateway, and retail access operator by preference. It buys or leases crucial inputs. It depends on IIGs for international bandwidth. It depends on NTTN or equivalent transmission arrangements for fiber/backhaul. It depends on NIX arrangements for domestic traffic exchange. It depends on BTRC license rules and tariff directives. It depends on local authorities for physical access, road work, pole routes, and practical last-mile installation conditions.

This dependency stack creates a margin squeeze. Retail price is visible and politically sensitive; wholesale bandwidth and transmission costs are less visible but determine gross margin. If retail customers expect “unlimited” service at a regulated or market-converged monthly price, peak-hour usage growth must be absorbed by buying more capacity, tolerating congestion, or tightening contention ratios. The cheapest way to survive is local density: many customers per route meter, low truck-roll frequency, high payment collection, and low churn.

NTTN pricing illustrates the cost mechanism. A 2021 Financial Express report on BTRC-approved NTTN tariffs said that up to 10 Mbps of regular capacity bandwidth would cost Tk 200–300 in metro areas and Tk 400–500 outside metro areas, with cheaper rates at higher bandwidth ranges. IIG pricing also affects the small ISP’s margin. A 2024 Business Standard report said Bangladesh internet gateways proposed reducing minimum selling prices for ISPs, including a proposed reduction from Tk365 to Tk215 per Mbps per month for ISPs buying up to 500 Mbps in Dhaka and from Tk399 to Tk265 outside Dhaka, with larger-volume buyers receiving lower proposed rates. The pattern is economically clear: larger buyers receive better unit costs, while small networks must either aggregate demand through wholesalers or accept weaker procurement leverage.

The regulatory environment also signals consolidation pressure. BTRC’s broadband connectivity report includes a policy suggestion to reduce the number of ISP licenses and combine or recycle more than 2,000 ISP licenses. The Daily Star reported in 2022 that 286 ISPs were to be disconnected and quoted the ISPAB president saying more than 40 percent of those 286 either did not operate or operated at very limited scale. A 2024 Financial Express report said BTRC issues ISP licenses in four categories and that license issuance had been suspended from March 1 of that year while new guidelines were developed, with ISPAB citing oversaturation in specific areas as a reason for stopping new ISP licenses earlier.

For Spider Net, this means that survivability is not only a question of whether customers like the service. It is also a question of whether the regulatory system continues to tolerate many small access licensees, whether renewal records are clean, and whether the operator can maintain enough scale to justify its separate existence.

Business model: local density, not brand premium

Spider Net’s likely revenue logic is standard for a small fixed broadband ISP: monthly retail subscriptions, installation charges, router or ONU/CPE economics, package upgrades, possible public/static IP fees, small-business plans, and perhaps local dedicated or semi-dedicated lines. The app’s functions confirm package-change requests, monthly bill payment, support tickets, payment history, data-usage visibility, and reconnection after nonpayment. The public evidence does not establish exact tariff plans, subscriber counts, ARPU, or churn.

The cost stack is more inferable. A small ISP must pay for upstream internet capacity, domestic transmission/backhaul, last-mile materials, building access, technician labor, splicing and repair, routers or ONUs, local collection, power backup, billing software, license fees, registry maintenance, and support. It must also carry the risk of customer equipment damage, unpaid bills, cable cuts, power instability, neighborhood disputes, and peak-hour congestion. In this model, the headline subscription price is less important than the ratio between average monthly revenue and four operational variables: peak bandwidth consumed per active subscriber, field-support cost per subscriber, churn and bad debt, and route/building density.

The small-ISP profit engine can be expressed simply. A dense lane or apartment cluster becomes attractive when one fiber route, one splitter/OLT path, and one technician visit can support many paying customers. A dispersed customer base is less attractive because every installation and fault consumes labor and materials. A price-sensitive customer base is viable only if service is standardized, collections are automated, and support is locally fast. The app’s bKash payment and automatic reconnection features directly address the collections side of this problem.

The visible IP inventory also shapes the revenue model. If AS138651 visibly originates only 512 IPv4 addresses while APNIC Labs’ customer-population estimates for SPIDERNET-AS-AP show an estimated user population in the low-to-high tens of thousands during 2025–2026, the network is almost certainly not assigning unique public IPv4 addresses to all end users. APNIC Labs’ Bangladesh AS-population pages show SPIDERNET-AS-AP estimates such as 11,533 users on 7 August 2025, 14,881 users on 12 January 2026, 18,468 users on 3 May 2026, and 17,387 users around late June 2026; these are measurement estimates, not contractual subscriber counts.

Those estimates must be used cautiously. APNIC Labs is estimating user populations from measurement systems, not publishing Spider Net’s accounts. Still, the order of magnitude is commercially useful. It implies a retail-eyeball network much larger than its public IPv4 count, consistent with carrier-grade NAT and shared public addressing. For small ISPs, CGNAT is economically rational because IPv4 addresses are scarce, but it creates service-quality frictions. Customers who need gaming, CCTV, remote access, VPN stability, or inbound connections may require static IPv4 or port-forwarding workarounds. Static public IPs then become a monetizable scarce product. Abuse handling also becomes harder because many customers may share a public IP at different times.

Pricing power is weak. Bangladesh broadband retail pricing has been subject to “One Country, One Rate” style tariff interventions since 2021, when BTRC fixed maximum broadband prices such as Tk500 for 5 Mbps, Tk700–Tk800 for 10 Mbps, and Tk1,100–Tk1,200 for 20 Mbps or above, according to local press coverage of the policy launch. Later local reports in 2026 described revised ceilings such as Tk500 for 30 Mbps, Tk1,000 for 100 Mbps, and Tk3,000 for 250 Mbps, though the reviewed sources for that update are press reports rather than a directly retrieved BTRC tariff PDF. The broader point is stable: retail broadband is politically regulated and intensely competitive. A small ISP cannot rely on premium pricing unless it serves a constrained building, business niche, or unusually service-sensitive customer base.

Supplier power and route dependency

Spider Net’s observed upstreams are Plusnet Inc and Windstream Communication Limited. These providers are not merely names in a table. They represent the buyer-supplier relationship that determines uptime, latency, congestion, and cost. A two-upstream configuration is better than a single upstream, but it is not deep resilience. If one provider raises prices, has a route leak, experiences congestion, or changes commercial terms, Spider Net has limited visible alternatives unless it has dormant or off-table arrangements through other IIGs or aggregators.

The AS-SET relationship clues broaden the dependency map. BGP.tools shows AS138651 listed in AS-SETs associated with Coronet IIG, Orange, Windstream, Plusnet, Summit, iTel, and other Bangladesh providers. Coronet’s own public website presents it as an IIG and IP Transit company in Bangladesh offering dedicated internet access, IP transit, MPLS, IPLC, and related services. BGP.tools shows Coronet Corporation Limited, AS149765, with upstreams including Bharti Airtel, Reliance Jio, Hurricane Electric, BSCCL, and Fiber@Home Global, and with a larger downstream/peer footprint than Spider Net.

Orange appears as another relevant network-market clue. BGP.tools identifies AS137453 Orange Bangladesh and AS135341 Orange Communication as Bangladesh eyeball networks with broader IPv4/IPv6 prefix sets and upstream relationships that include Exabyte, Windstream, Summit, Level3/CenturyLink, Earth Telecom, and each other. Exabyte is more directly implicated in Spider Net’s address-space pattern because BGP.tools shows Exabyte ASNs originating 103.135.138.0/24 and 103.135.139.0/24 with Spider Net descriptions. PeeringDB lists Exabyte AS150774 as “Exabyte IIG,” network type NSP, with IPv4 and IPv6 support in its public profile.

The economic interpretation is not that Spider Net is owned by these entities. The interpretation is that Spider Net sits inside a local wholesale web. In Bangladesh, the small ISP’s procurement leverage is often indirect. It may buy from one upstream today, have route objects maintained by another, appear in AS-SETs of several, and migrate parts of address space across ASNs as commercial conditions change. This route-policy flexibility can improve survivability, but it also reveals dependency. The customer may see only “Spider Net.” The packet path may involve a larger IIG, a national fiber provider, a route sponsor, and an international transit chain before it reaches the global internet.

Buyer power, churn, and the trust problem

Retail customers have high buyer power when several local ISPs can serve the same building or lane and when mobile data is an adequate fallback. They have lower buyer power when the building has one practical cable route, one operator has already wired the premises, or the user depends on a known local technician. In Bangladesh’s dense urban broadband markets, both conditions can coexist. A customer may have many advertised options but only one or two reliable installers at the exact address.

For Spider Net, trust is produced operationally. The app enables support tickets and disruption notices. The ISPAB and APNIC records provide phone and email contacts. The BGP records show the network is real and visible. But the weak website, limited public reviews, small app download count, and ambiguous license-renewal row all limit externally visible trust. A customer deciding between a national ISP and Spider Net would not find the same public assurance signals.

This creates a local-trust equilibrium. A small ISP can survive without national brand credibility if field response is superior. Fast repair after a fiber cut can matter more than a polished website. A technician who lives near the service area can be more valuable than a call center. A bKash reconnection after payment can reduce customer frustration more than a corporate portal. But this equilibrium is fragile. If service quality deteriorates for two billing cycles, if evening speed collapses, or if support tickets go unanswered, churn can be rapid because the customer’s switching cost is often just a new installation fee and a phone call to a rival.

The app’s 50+ download count should not be interpreted as the subscriber count. Many customers may not install the app, may pay through other channels, or may share accounts across households. But the low public download count is still a signal: Spider Net’s digital self-service channel is not operating at the visible scale of a mass-market national ISP. That is consistent with a local access network whose moat is geography and technician relationships, not consumer brand recognition.

Competition and substitutes

Spider Net competes against four classes of alternatives. The first is nearby local ISPs with similar BTRC categories and similar cost structures. The ISPAB member pages around Chattogram show numerous upazila/thana broadband providers and small operators with local addresses, emails, and phone numbers. These operators compete on installation speed, monthly price, peak-hour quality, payment convenience, and technician responsiveness.

The second class is larger regional or national ISPs. Larger networks can procure upstream capacity more cheaply, invest in better NOC functions, advertise more professionally, and sometimes offer bundled services. They may still perform poorly at hyperlocal repair if their field force is stretched or if a particular neighborhood has access constraints. The competitive contest is therefore not simply large versus small. It is procurement scale versus local density.

The third class is mobile broadband. Mobile operators are not always substitutes for fixed home broadband because unlimited home use, gaming, streaming, and work-from-home usage favor fixed connections. But mobile data is a powerful churn threat during outages. If a household can survive on mobile data for a week, it may tolerate a bad fixed ISP less patiently. If mobile packages become cheaper or 5G fixed-wireless access becomes stronger, small fixed ISPs lose some of their captive demand.

The fourth class is informal or semi-formal resale. In fragmented markets, a customer may buy internet from a local cable operator, building reseller, or neighboring network that depends on an upstream ISP. Regulatory enforcement and licensing cleanup affect this layer. If BTRC tightens enforcement, licensed entities like Spider Net may benefit from reduced informal competition. If enforcement is uneven, informal resellers can undercut licensed ISPs by avoiding compliance costs.

The supplier side is similarly competitive but concentrated by function. Upstream IIGs and larger NSPs compete for ISP bandwidth, but the small buyer’s leverage is constrained by volume. NTTN/backhaul options may be limited by geography, last-mile civil constraints, and practical route availability. Payment rails such as bKash reduce collection friction but create dependence on a third-party payment ecosystem. App vendors like SoftifyBD reduce software-development cost but create vendor dependence for billing and support workflows.

Ownership, financing, and control ambiguity

The public record does not identify Spider Net’s beneficial owner, capitalization, bank debt, shareholder structure, or management team. It does reveal recurring contact clues. APNIC’s organisation and IRT records show Spider Net contacts and the +8801837686343 phone number. ISPAB lists the Spider Net email as azizuddin2007@gmail.com and the same mobile number. APNIC-derived route-object material for 103.135.138.0/24 includes a route object described with “Md. Aziz Uddin” and the same Kotowali address. These clues suggest an individual administrative or proprietor-style control point, but they do not prove legal ownership.

This ambiguity is normal for small ISP intelligence. The economic assets of a local broadband operator are often not held in a tidy corporate shell. They may include a BTRC license, local right-of-way relationships, customer lists, outstanding receivables, installed fiber, OLTs and routers, IP resources, technician relationships, and upstream contracts. Financing may come from retained earnings, supplier credit, informal capital, customer installation fees, or local partners. None of that is visible in APNIC or BGP records.

Bangladesh’s ISP guidelines restrict license transfer and assignment without prior approval, which limits simple sale or pledge of the license as if it were ordinary inventory. In practice, consolidation may occur through customer migration, operational support, route sponsorship, address-space routing changes, or brand continuity under a larger network, rather than through a clean public acquisition. That is why Spider Net’s split routing through Exabyte-linked ASNs is more than a technical curiosity. It is a possible early sign of operational dependency, shared route management, or partial migration of network assets, even though it is not evidence of M&A.

The unresolved control question changes valuation. If Spider Net is an independent local ISP with active paying customers, its value lies in the customer book, local access plant, license, IP resources, and field team. If it is partly operationally absorbed by a larger provider, its standalone margin may be lower, but its customer relationships may still have acquisition value. If it is mainly an address/resource holder with a reduced retail base, the IPv4 and ASN assets dominate the economic story. The updated app and APNIC Labs traffic estimates argue against pure dormancy, but they do not settle the ownership question.

Abuse, reputation, and security signals

No reviewed source establishes a major Spider Net outage, litigation, procurement dispute, license cancellation, or named cybersecurity incident. The public footprint does show weak abuse and reputation signals at the IP level. CleanTalk’s page for 103.135.136.99 identifies the network as 103.135.136.0/24, AS138651, organisation Spider Net, website spidernetctg.com, and lists a network spam rate of 8 percent while showing AS spam rate of 0 percent. IPGeolocation.io’s page for 103.135.137.104 associates the IP with Spider Net and AS138651, assigns a threat score of 75, and labels the IP as proxy, residential proxy, VPN, anonymous, and known attacker, while not classifying it as spam or bot.

These should be treated as vendor-specific signals, not adjudicated facts. Residential broadband IPs often appear in proxy, scraping, credential-stuffing, malware, or spam datasets because infected customers, resold access, or third-party proxy SDKs use end-user connections. A single IP label does not prove company complicity. But abuse reputation is economically relevant. If Spider Net’s public IPs are used by residential proxy networks or compromised devices, upstreams may receive complaints, customers may face CAPTCHA friction, and the ISP may incur support costs. CGNAT can amplify the problem by making abuse attribution harder.

The app’s data-safety statement that data is not encrypted in transit is another risk signal, though it is self-reported through Google Play and may not describe every backend path. If accurate, it would be a governance issue for customer privacy and account security. Small ISPs often adopt billing platforms to reduce labor costs before they have formal privacy engineering, incident response, or vendor risk management. The next stage of small-ISP professionalism is not only more bandwidth; it is better identity, billing, and abuse controls.

Alternative hypotheses and what each changes economically

The first hypothesis is that Spider Net is an active independent Chattogram local ISP. This is supported by APNIC registration, BTRC/ISPAB identity, visible AS138651 routes, RPKI-valid prefixes, the customer app updated in late 2025, and APNIC Labs population estimates. Under this hypothesis, the economics are straightforward: Spider Net earns retail access revenue and survives by keeping local density high, wholesale input costs low, payment collection automated, and support responsive. The main risks are regulatory renewal, upstream costs, churn, and field failures.

The second hypothesis is that Spider Net remains a retail brand or address holder while parts of its network function are carried by larger operators. This is supported by the two Spider Net-described /24s originated by Exabyte ASNs, by route-object material referencing Exabyte-linked origins, and by AS-SET membership in larger provider sets. Under this hypothesis, Spider Net’s standalone gross margin may be lower because more economics are captured by upstream or operational partners. But survivability may improve if the larger partner supplies route stability, procurement leverage, or technical support.

The third hypothesis is that Spider Net’s license or retail operation is partially stale, while network resources and some customer base continue through successor or partner arrangements. The BTRC license-list row showing a 2023 validity date in a 2024 list creates this question, though it does not prove expiration or closure. Under this hypothesis, the core value shifts from active retail cash flow to migration rights, customer relationships, route objects, IP resources, and whatever local plant remains usable. The updated app and APNIC Labs estimates make a fully dormant interpretation less likely, but the public record cannot exclude partial operational transition.

The fourth hypothesis is that Spider Net is a small but resilient local operator whose public website is simply irrelevant. This is common in local access markets. The weak Host.io footprint of spidernetctg.com would then say little about customer economics, while the app, phone contact, neighborhood presence, and visible routes carry more weight. Under this hypothesis, analysts should not over-penalize weak web marketing. The relevant indicators are route continuity, app activity, support responsiveness, payment behavior, and upstream changes.

The unresolved facts are commercially material. A current BTRC renewal certificate would reduce regulatory risk. A customer count would determine whether APNIC Labs estimates are over- or under-stating the paying base. A wholesale invoice or upstream contract would reveal procurement leverage. A management confirmation would distinguish proprietor-controlled operation from partner-supported migration. A current route policy from Exabyte, Coronet, Orange, Plusnet, or Windstream would clarify whether the route clues are historical, live, or merely permissive.

Survivability over the next 12 to 36 months

Spider Net can survive if it remains useful at the neighborhood level. The small-ISP model is not obsolete in Bangladesh because local access is operationally granular. Large networks still need building entry, local repair, customer education, and collection. A small ISP with a dense service area, known technicians, automated payments, and acceptable upstream quality can defend a customer base even against larger brands. Spider Net has enough public evidence to be treated as a real network identity rather than a directory artifact.

But survivability is conditional. The first condition is regulatory clarity. The BTRC license-list ambiguity is not fatal, but it is material. A local ISP with uncertain renewal status is vulnerable to enforcement, consolidation, and customer trust erosion. The draft FTSP framework published by BTRC in 2025 proposed a technology-neutral fixed telecom service provider regime with nationwide and district FTSP categories, a 10-year license term, and stated goals including fair competition, SME participation, affordability, and quality. If that framework or a similar successor consolidates licenses, Spider Net’s value will depend on whether it can convert, partner, or be absorbed on favorable terms.

The second condition is upstream bargaining. Spider Net’s visible AS has only two observed upstreams. If wholesale bandwidth prices fall, small ISPs can benefit, but only if competition does not pass the savings entirely to customers. If wholesale terms tighten, small networks feel the squeeze first. The presence of Exabyte-originated Spider Net prefixes and route-policy ties to larger providers can be interpreted as a hedge: Spider Net may have more practical routing options than the two direct BGP peers suggest. It can also be interpreted as dependency: parts of the network may already require larger operators to remain visible.

The third condition is peak-hour quality. IPinfo’s network-activity view characterizes AS138651 as a consumer ISP with evening and late-hour usage patterns. This is consistent with household broadband economics. Evening peaks force the operator to choose between buying more capacity, accepting congestion, or managing customer expectations. In a low-price market, the difference between survival and churn is often the ability to keep video, gaming, and remote work tolerable between 8 p.m. and midnight.

The fourth condition is field operations. The app can reduce billing and support friction, but it cannot splice fiber, replace damaged ONUs, negotiate building access, or prevent local cable cuts. Small ISPs win when the local technician arrives quickly. They lose when every outage becomes a multi-day dispute. Field operations also determine capital efficiency: the same OLT, route, and technician can serve a profitable cluster or an uneconomic scatter of customers depending on local density.

The fifth condition is abuse and reputation management. Residential proxy and spam signals at the IP level do not prove wrongdoing, but they create cost. Upstream complaints, customer CAPTCHA issues, and shared-IP blacklisting can damage perceived quality. Better subscriber attribution, CGNAT logging, malware-notification workflows, and abuse-desk responsiveness are not luxuries for a small ISP; they protect upstream access and customer trust.

The sixth condition is IPv6. Spider Net has a /32 IPv6 allocation but no visible IPv6 origination from AS138651. Deploying IPv6 would not immediately remove the need for IPv4, but it would reduce future pressure on CGNAT, improve some application paths, and signal technical maturity. Failure to deploy IPv6 keeps the operator dependent on scarce IPv4 and makes static-public-IP demand harder to satisfy.

The most likely economic path is not dramatic growth into a national ISP. It is either disciplined local survival, operational partnership, or absorption into a larger provider’s access and route ecosystem. Spider Net’s value is in being locally embedded and technically registered. Its vulnerability is that each layer above and below the retail customer is controlled by someone else.

Evidence ledger

1. APNIC WHOIS, organisation ORG-SN17-AP, “Spider Net.” URL: https://wq.apnic.net/apnic-bin/whois.pl?searchtext=ORG-SN17-AP. This is the primary registry evidence for the Spider Net organisation name, Bangladesh country code, Kotowali/Chittagong address, phone, email, maintainer, and APNIC organisation handle.
2. APNIC WHOIS, IPv6 allocation 2404:7ec0::/32, SPIDERNET-BD. URL: https://wq.apnic.net/apnic-bin/whois.pl?searchtext=2404:7ec0::/32. This proves a portable IPv6 allocation to Spider Net and gives abuse/IRT contact context.
3. AbuseIPDB WHOIS mirror for 103.135.138.197 / APNIC IPv4 record and route objects. URL: https://www.abuseipdb.com/whois/103.135.138.197. This provides the 103.135.136.0–103.135.139.255 allocation, usage-type/geolocation vendor context, and route-object evidence for Exabyte-linked origins on 103.135.138.0/24.
4. BGP.tools AS138651. URL: https://bgp.tools/as/138651. This is the main public BGP source for AS138651’s originated prefixes, RPKI status, observed upstreams, APNIC aut-num fields, and AS-SET memberships.
5. Hurricane Electric BGP Toolkit, AS138651. URL: https://bgp.he.net/AS138651. This corroborates AS138651’s two IPv4 prefixes, zero IPv6 prefixes, two BGP peers, and RPKI-valid originated routes.
6. IPinfo, AS138651. URL: https://ipinfo.io/AS138651. This corroborates the ASN name, ISP type, 512 IPv4 addresses, zero hosted domains, zero downstreams, Plusnet/Windstream upstream context, pingable IPs, and consumer ISP activity pattern.
7. Host.io, spidernetctg.com. URL: https://host.io/spidernetctg.com. This shows the domain’s hosted IP/ASN context, “Redirecting…” page title, shared hosting environment, and lack of backlink/redirect footprint.
8. Google Play, “Spider Net” app by SoftifyBD. URL: https://play.google.com/store/apps/details?id=bd.com.softifybd.spider_net. This is the main retail-operations evidence: usage viewing, package change, router test, ticketing, bKash payment, payment history, outage/offers/news notifications, automatic reconnection, update date, downloads, and data-safety disclosure.
9. ISP Association of Bangladesh members page, Spider Net and similar-name entities. URL: https://ispab.org/members/S?page=5. This gives Spider Net membership number B-074, upazila/thana category, Anderkilla/Chittagong address, email, phone, and distinction from Spider Mesh and Spider Networks.
10. BTRC ISP Upazila/Thana License List as on 18-12-2024. URL: https://objectstorage.ap-dcc-gazipur-1.oraclecloud15.com/n/axvjbnqprylg/b/V2Ministry/o/office-btrc/2024/12/29e9f4bf494145f5bfee76bd1a384ddc.pdf. This official list places Spider Net in Chattogram Sadar with license number 14.32.0000.702.47.043.22.376 and the Kotowali address; it also creates a renewal-date ambiguity.
11. BTRC ISP Regulatory and Licensing Guidelines 2020. URL: https://btrc.gov.bd/sites/default/files/files/btrc.portal.gov.bd/notices/c71cc8a5_2753_4bf8_a72b_8b120dc21077/ISP%20Guideline%202020%20.pdf. This is the core regulatory source for ISP service scope, license categories, last-mile distance limits, IIG/NIX interconnection obligations, and transfer restrictions.
12. Draft Regulatory and Licensing Guidelines for Fixed Telecom Service Provider, 2025. URL: https://btrc.gov.bd/sites/default/files/files/btrc.portal.gov.bd/notices/9478d892_7127_4686_8098_a3e4fe6d1ae5/Draft%20Regulatory%20and%20Licensing%20Guidelines%20for%20Fixed%20Telecom%20Service%20Provider.pdf. This is the main regulatory watchpoint source for a possible FTSP framework and license-category restructuring.
13. BTRC Bangladesh Broadband Connectivity Report. URL: https://objectstorage.ap-dcc-gazipur-1.oraclecloud15.com/n/axvjbnqprylg/b/V2Ministry/o/office-btrc/2024/12/2553c9a48743467faaa8b420c2e6ecb5.pdf. This source is used for consolidation policy signals, including the suggestion to reduce, combine, or recycle more than 2,000 ISP licenses.
14. The Daily Star, “286 ISPs to be disconnected.” URL: https://www.thedailystar.net/business/economy/news/286-isps-be-disconnected-3065936. This local press source gives context on IIG function, licensing categories, and limited-operation ISPs.
15. The Financial Express, “BTRC restarts accepting ISP licence applications.” URL: https://today.thefinancialexpress.com.bd/trade-market/btrc-restarts-accepting-isp-licence-applications-1713376640. This source gives context on license categories, suspension of new licenses, and oversaturation concerns.
16. The Financial Express, “BTRC okays tariffs for NTTN services.” URL: https://today.thefinancialexpress.com.bd/last-page/btrc-okays-tariffs-for-nttn-services-1628707314. This source is used for NTTN cost structure and the volume-sensitive nature of transmission pricing.
17. The Business Standard, “Internet gateways propose 40% tariff cut.” URL: https://www.tbsnews.net/bangladesh/internet-gateways-propose-40-tariff-cut-how-it-may-benefit-broadband-users-986811. This source gives IIG wholesale-price context and scale effects in bandwidth purchasing.
18. The Daily Star, “One Country, One Rate: Internet at flat rate across the country soon.” URL: https://www.thedailystar.net/toggle/news/one-country-one-rate-internet-flat-rate-across-the-country-soon-2105929. This source documents the 2021 retail broadband tariff framework and regulated retail-price context.
19. The Business Standard, “Broadband Internet: Rural users still pay higher, get lesser speed.” URL: https://www.tbsnews.net/bangladesh/telecom/broadband-internet-rural-users-still-pay-higher-get-lesser-speed-277135. This corroborates the 2021 One Country, One Rate price bands and rural/urban pricing concern.
20. NewsBangladesh, “New internet pricing enable high-speed access.” URL: https://www.newsbangladesh.com/english/information-technology/news/125398. This is a press source for reported 2026 broadband price ceilings; it should be treated as secondary pending direct BTRC notice verification.
21. Views Bangladesh, “BTRC sets new internet prices.” URL: https://viewsbangladesh.com/btrc-sets-new-internet-prices/. This is another press source for reported 2026 broadband price ceilings and contention-ratio discussion; it is used as secondary market context.
22. Coronet Corporation Ltd website. URL: https://coronet.com.bd/. This source identifies Coronet as an IIG/IP Transit company in Bangladesh and describes its DIA, IP transit, MPLS, IPLC, and related services.
23. BGP.tools AS149765, Coronet Corporation Limited. URL: https://bgp.tools/as/149765. This source provides Coronet’s upstreams, prefixes, and broader downstream/peer scale relative to Spider Net.
24. BGP.tools AS137453, Orange Bangladesh. URL: https://bgp.tools/as/137453. This source is used for Orange Bangladesh’s network size, upstreams, and relevance as a route-policy relationship clue.
25. BGP.tools AS135341, Orange Communication. URL: https://bgp.tools/as/135341. This source is used for Orange Communication’s network size and upstream/downstream context.
26. BGP.tools AS150774, EXABYTE LTD. URL: https://bgp.tools/as/150774. This source shows 103.135.138.0/24, described as Spider Net, originated by Exabyte AS150774.
27. BGP.tools AS150178, EXABYTE LTD. URL: https://bgp.tools/as/150178. This source shows 103.135.139.0/24, described as Spider Net, originated by Exabyte AS150178.
28. PeeringDB, Exabyte AS150774. URL: https://www.peeringdb.com/net/33169. This source identifies Exabyte AS150774 as Exabyte IIG / Exabyte Limited and provides public interconnection-profile context.
29. CleanTalk, 103.135.136.99 spam report. URL: https://cleantalk.org/blacklists/103.135.136.99. This is a weak but relevant IP-reputation signal for a Spider Net /24.
30. IPGeolocation.io, 103.135.137.104. URL: https://ipgeolocation.io/browse/ip/103.135.137.104. This is a third-party geolocation and security-label source associating a sample IP with AS138651 and Spider Net; it is not treated as dispositive.
31. APNIC Labs, Customer Populations Estimated per Network for Bangladesh. URL: https://stats.labs.apnic.net/cgi-bin/aspop?c=BD. This is used cautiously for estimated user-population signals for SPIDERNET-AS-AP, not for subscriber counts.
32. APNIC RDAP information page. URL: https://www.apnic.net/about-apnic/whois_search/about/rdap/. This explains RDAP as an alternative to WHOIS and confirms APNIC’s support for RDAP/autnum lookups, relevant to the starting evidence.

Watchpoints

The first watchpoint is BTRC license status. A current renewal record, conversion into a district/FTSP framework, or removal from future official lists would materially change the risk profile. The 2024 upazila/thana list confirms the identity and license number but shows a 2023 validity date, so the next public license record is a high-value trigger.

The second watchpoint is route origination. If AS138651 begins originating 103.135.138.0/24 and 103.135.139.0/24 again, the standalone-network interpretation strengthens. If more Spider Net address space moves to Exabyte, Orange, Coronet, or another AS, the operational-partner or absorption hypothesis strengthens. If the two current AS138651 /24s disappear, Spider Net’s independent network value would decline sharply.

The third watchpoint is upstream change. A shift from Plusnet/Windstream to Coronet, Exabyte, Orange, Summit, iTel, or another IIG/NSP would reveal new bargaining conditions. A third direct upstream would improve resilience. A single upstream would increase outage and price risk.

The fourth watchpoint is RPKI validity. Spider Net’s visible originated routes are RPKI-valid. Any invalid or not-found route state would raise technical-governance risk, especially if address space continues to be split across multiple ASNs.

The fifth watchpoint is IPv6 activation. A public announcement of 2404:7ec0::/32, or customer evidence of IPv6 service, would reduce long-term IPv4/CGNAT pressure and signal higher technical maturity. Continued absence of IPv6 keeps the operator dependent on scarce IPv4 and static-IP monetization.

The sixth watchpoint is app activity. Continued Google Play updates, more downloads, stronger data-safety posture, and stable bKash/payment features would support an active retail-service interpretation. Stale app updates or broken payment/reconnection functions would be an early warning on customer-control operations.

The seventh watchpoint is abuse reputation. More residential-proxy, spam, or “known attacker” signals across Spider Net IPs would increase upstream complaint risk and customer friction. Improvement would indicate better subscriber attribution, malware response, and CGNAT governance.

The eighth watchpoint is wholesale tariff movement. Further reductions in IIG or NTTN input prices would help small ISPs only if competition does not force full pass-through to retail customers. Wholesale increases or stricter contention rules would pressure Spider Net’s gross margin.

The ninth watchpoint is regulatory consolidation. Any implementation of the FTSP framework, license recycling, revenue-share changes, or reduction in small ISP licenses would change the option value of Spider Net’s local license and customer book.

The tenth watchpoint is local field evidence. New customer complaints, Facebook posts, technician hiring, office changes, neighborhood expansion claims, or outage notices would be more valuable than generic website updates. For Spider Net, the economic truth is most likely to appear first in field operations, not in corporate announcements.