Tornado Networks is not a national broadband platform waiting to be discovered. It is a local answer to a harder Pakistani question: who keeps a small city connected when national backbones, mobile networks, regulator rules, electricity reliability and household budgets all press on the same last mile? The company sits in Bahawalnagar, a Punjab district centre far from the glamour of Karachi, Lahore and Islamabad fibre marketing. Its public materials describe wireless internet, point-to-point and point-to-multipoint expertise, home broadband, TV bundles, CCTV cameras and biometric attendance systems. That mixture can look untidy from a metropolitan broadband lens. In a smaller access market, it is the business logic. The same person who sells a link may also install a camera, fix a router, climb a roof, answer a phone call and collect a bill.

The subject is Tornado Networks (Pvt.) Limited, the company tied to the public website tornado.com.pk, the Bahawalnagar address at Police Line Road, Nizampura West, Street 12, and the phone number 0333-2277111. The Pakistan Telecommunication Authority's list of CVAS licensees identifies Tornado Networks (Pvt.) Limited under licence DIR(L)/CVAS-723/PTA/2015, granted on 17 November 2015 and expiring on 17 November 2030, with Punjab jurisdiction, Data CVAS, Internet Services, commencement status issued on 5 September 2017, and Mamoon ur Rasheed named as chief executive officer. APNIC and public routing records connect the same company name to AS139718, also called TORNADO3-AS-AP. That identity chain matters because "Tornado" is an easy name to confuse with unrelated telecom, hosting, security or software businesses in other countries. The company analysed here is the Bahawalnagar Pakistani ISP.

The public website is both helpful and revealing. It says Tornado Networks Private Limited is a wireless internet service provider with expertise in PTP and PTMP technology and lists familiar small-ISP equipment ecosystems such as Ubiquiti, MikroTik, D-Link, TP-Link and Alfa. Its internet page advertises monthly plans at 1 Mbps for Rs 999, 2 Mbps for Rs 1,999, 4 Mbps for Rs 2,999, 6 Mbps for Rs 3,999 and 8 Mbps for Rs 5,999, each described with unlimited data, video calling, unlimited downloads, unlimited upload and 24/7 support. The home page also markets "up to 100+ Mbps", 99 percent or 99.999 percent uptime language, satellite TV, free Wi-Fi router language, and the claim that Tornado is backed by multiple service providers. The footer names Wateen, Jazz, Ufone and PTCL through partner logos, although the site does not disclose the commercial terms or technical form of those relationships.

That range of claims should be read carefully. The low-speed plan table and the "up to 100+ Mbps" headline can both be true in a broad service catalogue, but the public page does not tell readers how many customers buy each speed, where the highest-speed service is available, how much of the access network is wireless rather than fibre, how much traffic is contended, or what the practical service-level promise means during power cuts and upstream congestion. The site also carries unfinished filler text in several sections, which lowers confidence in the marketing copy as a complete representation of the business. Still, the commercially useful pieces are clear: Tornado sells connectivity to households and small businesses, uses wireless access as a core technology, offers direct contact by mobile and landline, and presents itself as a local support provider rather than a distant utility.

The real network record is stronger than the website polish. APNIC RDAP records show AS139718 registered in October 2019, active, with Tornado Networks (Pvt.) Limited as the registrant. APNIC records also associate the company with the IPv4 block 103.144.10.0 to 103.144.11.255, a /23 of 512 IPv4 addresses, and an IPv6 /48 at 2001:df1:a880::. Hurricane Electric's BGP Toolkit shows AS139718 originating two IPv4 prefixes, 103.144.10.0/24 and 103.144.11.0/24, with two originated and announced IPv4 prefixes, no originated IPv6 prefixes, two RPKI-valid originated routes, and three observed IPv4 peers. BGP.Tools similarly classifies the AS as active, small and Pakistani, with two IPv4 prefixes and no visible IPv6 origination. The network is small, but it is not only a reseller brochure. It has its own public autonomous system, number resources and externally visible routes.

The IPv6 detail is a useful test of Tornado's technical maturity. APNIC assigns an IPv6 /48 to the company, and public search snippets for Tornado's Facebook page refer to a first IPv6 deployment in Bahawalnagar. Yet the BGP views reviewed for this article do not show AS139718 originating IPv6 routes. That gap does not prove the company has no IPv6 anywhere; it may reflect timing, internal tests, delegated customer trials, a route withdrawn from public visibility, or data limitations. But it does show a difference between resource allocation, social signalling and current public internet routing. For customers, the practical question is not whether a provider once tested IPv6. It is whether IPv6, route security, DNS, support scripts and customer equipment all work consistently on live plans.

Scale is the next hard fact. Hurricane Electric reports 512 originated IPv4 addresses for AS139718. BGP.Tools ranks the network far below the country's major national and metro providers. APNIC Labs' June 2026 customer-population estimate places AS139718 around rank 153 in Pakistan, with roughly 2,659 estimated users, about 0.01 percent of the measured country total. That estimate should not be mistaken for a subscriber count. APNIC's method estimates user population from measurement samples and can understate or overstate a small ISP depending on address sharing, mobile offload, NAT, test distribution and how customers are routed. Even with that caveat, the order of magnitude is valuable. Tornado is a local network in a country where the top measured access networks serve millions.

Pakistan's national statistics explain why a small provider can be locally important despite that scale. The Pakistan Economic Survey 2025-26, using PTA figures, says Pakistan reached 207.22 million telecom subscriptions by March 2026, including 160.9 million broadband users and more than 4.29 million fixed-broadband connections. It also reports 92 percent cellular population coverage, 3G/4G signals to more than 81 percent of the population, and 2.7 million FTTH subscribers by March 2026. The contrast is the whole market story. Mobile broadband is enormous; fixed broadband is still much smaller. A household in Bahawalnagar may have mobile data from Jazz, Zong, Ufone or Telenor, but that does not remove the need for home Wi-Fi, small-business connectivity, CCTV backhaul, school links, payment terminals, desktop work, streaming and local troubleshooting.

The Universal Service Fund's older Multan Telecom Region project adds historical texture. That region included Bahawalnagar, Bahawalpur, Dera Ghazi Khan, Khanewal, Leyyah, Lodhran, Multan, Muzaffargarh, Rahim Yar Khan, Rajanpur and Vehari. USF described an area of about 100,000 square kilometres with roughly 24 million people, and noted that narrowband internet had been available through multiple operators across the districts before the broadband project. PTCL, Wateen, Worldcall, Wi-Tribe and Multinet appeared in the bidding history. PTCL ultimately had an all-11-district broadband contract target, while Worldcall's final footprint excluded Bahawalnagar and Rahim Yar Khan. This is not current evidence of Tornado's market share, but it explains the terrain: southern Punjab connectivity has long depended on a mix of national incumbents, subsidised expansion, local wireless operators and uneven district economics.

Tornado's likely customer is therefore not a single archetype. The first segment is the ordinary household that wants a stable connection for schoolwork, social media, messaging, entertainment and remote administrative tasks. The second is the shop, office, clinic, school, hostel, warehouse or small factory that needs Wi-Fi, cameras, attendance machines, repairs and someone local to call when the service is down. The third is the edge customer outside easy fibre reach, where wireless CPE and point-to-multipoint coverage may be more practical than waiting for a national fixed-line upgrade. The fourth is the bandwidth-sensitive user who may keep mobile data as a backup but still wants fixed or fixed-wireless internet because phone packages, hotspot limits and tower congestion are not a substitute for a home or office link.

The unit economics are tight. A Rs 999 monthly 1 Mbps customer does not leave much room after taxes, payment friction, customer-premises equipment, installation time, router support, rooftop alignment, upstream capacity, tower rent or mast maintenance, staff, complaint handling, diesel or battery backup and bad debt. The 4 Mbps plan at Rs 2,999 and the 8 Mbps plan at Rs 5,999 improve gross revenue, but they also raise customer expectations. If a subscriber expects video calls, streaming and always-on service, Tornado has to provision enough upstream and local access capacity to avoid evening congestion. Wireless installation creates a different payback curve from pure FTTH: CPE may be reused, rooftops can be reached quickly, and coverage can expand from a tower, but line-of-sight problems, interference, weather, power supply and truck rolls can eat margin. Collection discipline and churn matter as much as the nominal plan price. A customer who leaves after a subsidised install can destroy the economics of that account; a customer who stays because the local technician solves problems can become profitable despite low speed.

Tornado's plan prices also sit in an uncomfortable national reference frame. PTCL's public DSL table shows 10 Mbps at Rs 2,999, 20 Mbps at Rs 3,299, 30 Mbps at Rs 3,999, 50 Mbps at Rs 6,299 and 100 Mbps at Rs 10,999, with GPON, VDSL and DSL technologies under its national broadband brand and availability in more than 2,000 cities and towns. A simple speed-per-rupee comparison makes Tornado's 4 Mbps and 8 Mbps wireless plans look expensive. That comparison is incomplete. PTCL availability, copper quality, local repair speed, fibre rollout, installation delay, customer support and actual throughput can vary by area. A small provider can charge a local premium if it installs quickly, answers the phone and restores a link when a larger provider's process is slow. But the benchmark still constrains Tornado: if customers can get reliable fibre or DSL at materially better speeds, the small wireless provider has to win through responsiveness, coverage gaps, relationship and bundled services.

Upstream dependence is where local trust meets national infrastructure. Tornado's website says it is backed by multiple service providers and displays Wateen, Jazz, Ufone and PTCL logos. Public BGP data shows observed peers including CMPak Limited, Login.Me (Pvt) Ltd and Shahram Telecom Private Limited in the Hurricane Electric view. These are not the same thing as full supplier contracts, and public BGP views cannot show every transport, resale, private interconnect or physical backhaul arrangement. They do show that Tornado is not an island. Its customer experience depends on larger networks, available transit, local transport paths, radio backhaul, power at towers and the health of Pakistan's wider domestic and international routes. A fault beyond Bahawalnagar can still become Tornado's customer-service problem if the household sees only that "the internet is slow."

This dependency is not theoretical in Pakistan. The country has seen repeated internet slowdowns from submarine cable faults, backbone outages, traffic management controversies and power-related service degradation. Cloudflare's Q3 2025 disruption summary described a near-complete outage at Pakistani backbone provider PTCL in August 2025, with traffic dropping sharply. Dawn reported in October 2025 that users often face disruptions from persistent submarine-cable faults, including Red Sea cable cuts affecting peak-time performance for some users. Arab News reported in January 2025 that Pakistan warned of internet disruptions after a fault in the AAe-1 submarine cable near Qatar. These examples are national rather than Tornado-specific, but they define the operating environment. A small ISP cannot control the Red Sea, PTCL's national backbone or a distant cable repair. It can only decide how much redundancy to buy, how clearly to communicate, and how quickly to separate a local fault from an upstream event.

Power reliability is the second dependency. Wireless access networks need powered radios, switches, routers, backhaul links, customer devices and sometimes powered towers or repeater sites. A household also needs electricity for the Wi-Fi router and client devices. Pakistan's broader telecom sector has repeatedly treated power backup as a resilience problem, with operators using batteries, generators and, increasingly, solar at selected sites. For a small provider, every additional hour of backup has a capital cost and a maintenance cost. Batteries age, generators need fuel, solar systems need upfront funding and rooftops have physical limits. The customer experiences this in simple terms: when the lights go out, does Wi-Fi stay up, does the mobile backup work, and does anyone answer the phone? Tornado's local promise is only as strong as its answer to that sequence.

Regulation shapes the business more than a casual buyer might realise. Tornado's CVAS licence gives it a formal place in the PTA framework, and its expiry in 2030 matters because it reduces one identity risk. But PTA's current Class VAS page says it has commenced district-level class licensing for internet services, while provincial and nationwide class licences for voice and data internet services are suspended or stopped for new applications. It also advises existing Class Licensees for Data (Internet Services) seeking upgrade and renewal to obtain a Fixed Local Loop licence. The wording is important for small ISPs. A provider that began in the older CVAS regime may face different requirements as it grows, upgrades, renews or wants a more infrastructure-heavy footprint. Compliance is not only paperwork; it can affect expansion timing, fees, legal certainty, service descriptions and investment appetite.

There is also the informal regulator: the customer who knows the office address. Tornado's public contact details are local, direct and personal. Its Facebook page snippets show more than 8,000 likes and 93 check-ins, with the same mobile and landline numbers. WISPAP's board-member page describes Mamoon-UR-Rasheed as the founder and chief executive officer of Tornado Networks Pvt. Ltd, working in Bahawalnagar, and says he began providing internet services about ten years before that profile text. That association profile is promotional rather than audited, but it matches the pattern visible elsewhere: Tornado is not just a brand; it is tied to a named local operator and a town-level reputation. In a market where service complaints become social conversations, this can be an asset or a liability. The person who answers the complaint is part of the product.

The market signals point in the same direction. Search snippets from Tornado's Facebook posts refer to a tower installed in Chishtian in 2017, to "fastest internet service in Bahawalnagar," and to a claim that Tornado first deployed IPv6 in Bahawalnagar. A Facebook group question from 2022 asks whether any optical-fibre internet service exists in Bahawalnagar New City. These are scattered signals, not audited operating data. They suggest a market where coverage announcements, tower locations, neighbourhood questions and technical bragging carry commercial weight. Customers are not only comparing national package tables; they are asking whether a provider reaches their street, whether the link works for games and video calls, whether the installer can come quickly, and whether the service survives common local problems.

This is where the trust premium becomes economic. Tornado cannot out-scale PTCL, Cybernet, Nayatel, Jazz, Zong, Ufone or Wateen. It can only sell a narrower proposition: local access where large providers are weak, a technician who can visit, wireless deployment that avoids slow trenching or copper limitations, and adjacent services that make the relationship stickier. CCTV is a good example. A shop that buys cameras from the same provider may also keep the internet link because the camera feed, router, power supply and mobile viewing app are all part of one practical problem. Biometric attendance equipment works the same way for a school, factory, clinic or local office. These services are not glamorous, but they turn a commodity access bill into a wider local technology relationship.

The cost side is just as local. Wireless networks avoid some civil-work costs, but they create a field-support business. Someone has to survey line of sight, mount equipment, align antennas, waterproof cables, replace damaged power supplies, configure routers, handle interference, update firmware, answer calls after storms, and explain to a customer why the problem is Wi-Fi inside the house rather than the upstream link. If Tornado operates any towers or repeaters, site rent, electricity, batteries, access rights and security matter. If it leans on third-party backhaul, supplier concentration matters. If it promises 24/7 support, staffing and response discipline matter. The local network may be small, but the operational surface is broad.

The demand side is price-sensitive. Pakistan's telecom growth has been built on affordability, but that also means low ARPU pressure. The Economic Survey says telecom revenues were Rs 1,075 billion in FY 2024-25 and Rs 837 billion in July-March FY 2026, with sector investment of US$567 million in the same July-March period. Those are national numbers; they do not reveal Tornado's revenue. They do show a market where operators are investing while consumers remain highly price-conscious. A Rs 1,000 to Rs 6,000 monthly internet bill can be material for a household or small shop. That creates a constant tension: customers want unlimited usage and fast support, while providers must pay for upstream capacity, people and equipment in a currency environment where imported gear can become more expensive.

Competition comes from several directions at once. PTCL is the obvious incumbent, with fixed-line reach, Flash Fiber branding, DSL, GPON and national scale. Mobile operators are substitutes for some households, especially when 4G packages are cheap enough or when fixed service is unreliable. Nayatel, StormFiber, Transworld, Wateen and other fibre brands shape national expectations even where they may not be present on a given street. Local cable operators and small CVAS licensees can compete neighbourhood by neighbourhood. Informal resellers and poorly documented access sellers can also disturb price discipline. Starlink or other satellite broadband is not yet the central mass-market substitute for most Bahawalnagar households, but satellite changes the imagination of customers at the edge: if terrestrial providers fail repeatedly, a self-installed alternative becomes part of the conversation for farms, remote offices and higher-income users.

Bahawalnagar's geography reinforces that competitive pattern. A dense urban block can support a cheaper, higher-capacity access build because one technician visit may connect several nearby homes and because backhaul can be shared across enough paying users. A spread-out edge settlement or road-side cluster is different. A wireless link may be the rational first build because it avoids trenching, pole negotiations and long payback on fibre drops. But wireless is more exposed to roof access, alignment, interference and customer-side power. The provider has to decide which addresses deserve new coverage, which should wait for demand to form, and which are likely to generate too many support calls for too little monthly revenue. That is a capital-allocation problem in miniature. It is not visible in national broadband penetration figures.

The same geography changes how a customer judges quality. In a metropolitan fibre market, the household may compare speed tests and bundle prices. In a smaller city, the first question is often availability: can the provider reach this street, this house, this rooftop or this shop? The second question is restoration: if the link drops after a storm, who comes? The third is social proof: who else nearby uses the service, and did the provider solve their complaint? Tornado's social presence, direct phone numbers and local address are therefore not only marketing details. They are part of the trust architecture. A customer may tolerate lower headline speed if the provider is reachable and honest about faults. The same customer may leave quickly if the provider goes silent, because the relationship is personal rather than abstract.

There is also a security and abuse-handling dimension to a small AS. A network with 512 public IPv4 addresses can develop a reputation quickly. If compromised customer routers, badly secured CCTV systems, infected PCs or spam activity create abuse complaints, upstreams may push the local provider to clean up the traffic. If the provider responds slowly, its routes, mail reputation and supplier relationships can suffer. APNIC's RDAP records show abuse contacts and recent validation for Tornado-related email addresses, which is positive as a public-contact signal. But public records do not show how the company handles malware calls, bot traffic, unlawful content requests, customer identity records or law-enforcement demands. In Pakistan, where internet governance, blocking orders and political disruptions can affect connectivity, even a small access provider needs disciplined operating habits.

Customer collection is another underestimated part of the economics. Small access providers often collect through cash, bank transfer, mobile wallets or local office visits, depending on the customer base. Each method has cost and leakage. Cash requires staff time and creates reconciliation risk. Digital payment is easier to scale but can be unfamiliar to some customers or carry fees. Late payment creates an awkward local relationship: the same provider who wants to keep a family or shop connected may also need to suspend service quickly enough to protect cash flow. A national operator can hide behind automated billing. A local provider may have to negotiate in person. That is one reason adjacent services matter. A shop that relies on Tornado for cameras, attendance equipment and internet may be more likely to pay because the relationship has more operational value than entertainment alone.

The business-customer opportunity is attractive but demanding. A school, clinic, money-transfer shop, retailer, rice-trade office or local government contractor does not simply buy a Mbps number. It needs working Wi-Fi, stable video calls, payment connectivity, CCTV access, file transfer, staff attendance records and some assurance that a human being will come when systems fail. These accounts can support higher monthly revenue than a home user, and they can justify custom installation or managed devices. They also raise the standard. A broken link at home is frustrating; a broken link at a shop can mean failed card payments, lost customer messages, a camera blind spot or a missed administrative deadline. If Tornado can serve such accounts well, it can escape the worst commodity pricing. If it cannot, the service mix becomes a support burden.

The website's TV-bundle language is a reminder that broadband competition is not only about internet access. In many markets, the household historically bought entertainment, telephone, Wi-Fi and local repair through overlapping channels: cable operators, satellite dishes, mobile shops, computer repair outlets and fixed-line operators. Tornado's claim around 150-plus or 200-plus channels may reflect this blended customer expectation, even if the public site does not provide enough detail to judge the offer. Bundling can reduce churn because customers do not want to rearrange several services at once. It can also expose the provider to content rights, device support, remote-control issues and customer complaints that are not really access-network faults. A local ISP earns margin by solving messy household technology problems; it also risks becoming responsible for every screen and router in the house.

Tornado's public routing footprint gives counterparties a way to ask sharper questions. Two RPKI-valid IPv4 routes are better than an opaque resale arrangement, because they show that the company can originate its own prefixes in the global routing system. The absence of publicly visible IPv6 origination, despite a registered /48, is not disastrous, but it is a missed opportunity if customers or business accounts begin to expect modern addressing. The observed peer set also raises a practical diligence question: how much of the route diversity is operationally meaningful from Bahawalnagar, and how much is a view from public collectors? A provider can look multihomed in public tables while still depending on one physical path out of town. Conversely, a provider may have private transport arrangements that public BGP pages do not show. Buyers and large customers would need physical-route evidence, not just AS names.

The regulatory risk is equally practical. PTA's public language around district-level class internet licensing and FLL upgrade or renewal advice points to a sector moving from permissive small-provider growth toward more formal infrastructure governance. That can benefit serious operators by making the market cleaner and reducing the space for informal resellers. It can also raise compliance cost for small firms. If Tornado wants to expand beyond wireless access into more fibre, enterprise, managed Wi-Fi, public-sector work or larger service areas, it may need more documentation, more fees, better engineering records and stronger consumer-handling processes. The key question is whether the company can professionalise without losing the local responsiveness that makes it useful.

None of this should be overstated into a hidden national champion story. The evidence does not support that. Tornado is better understood as an example of the layer that sits between household demand and national infrastructure. That layer is full of small decisions: whether to stock spare radios before the monsoon, whether to replace a weak battery, whether to pay for another upstream path, whether to train a second technician, whether to document tower sites, whether to answer complaints publicly, whether to remove obsolete package claims from the website, and whether to tell customers the truth when a national upstream has failed. These decisions determine whether a local ISP becomes a trusted utility or a temporary workaround.

For Tornado, the strongest competitive defence is not headline speed. It is service memory. A national provider can advertise a bigger number, but a local provider may know which rooftop has clear line of sight, which lane loses power first, which shop needs camera access at night, which customer pays late but stays loyal, and which business owner will switch if the connection fails during working hours. That knowledge lowers some selling and support costs. It can also trap the company in labour-intensive work that does not scale. The same customer intimacy that wins the account can turn into a constant stream of low-margin calls.

The public evidence also exposes a credibility problem. Tornado's website contains useful contact and package information, but it also has generic website filler and counters that display as zeros for corporate customers, home users and uptime in the captured text. A serious business customer reading the site might wonder whether the public web presence reflects the operational quality of the network. That weakness is not fatal for a local ISP; many small networks are technically competent and commercially active while their websites lag behind. But it matters for trust outside the existing relationship circle. If Tornado wants to sell higher-value business accounts, backhaul, managed Wi-Fi, CCTV maintenance or enterprise-style service, the public evidence should look as disciplined as the field operation claims to be.

The most interesting upside case is that Tornado is a durable regional wireless specialist with a loyal Bahawalnagar base, a real AS, formal PTA licensing, visible number resources, practical know-how in MikroTik and Ubiquiti-style deployments, and enough local reputation to defend accounts against larger but slower operators. If it can buy or build reliable backhaul, maintain power resilience, improve its web and support systems, and turn CCTV and attendance services into recurring business relationships, it can remain valuable without ever becoming large. In this reading, Tornado is part of Pakistan's necessary middle layer: too small to move national statistics, but important to the streets and towns where national coverage maps do not guarantee customer satisfaction.

The bearish case is that the same evidence describes fragility. The AS is tiny. Public routing shows no current IPv6 origination. The website's published speeds are low against national benchmarks, while the broader marketing claims are not supported by detailed coverage or performance data. Upstream and backhaul contracts are undisclosed. Customer-review evidence is thin. Power resilience is unknown. The PTA licensing transition could add cost or uncertainty if the company needs upgrade or renewal under a different framework. Larger providers can improve local fixed service, mobile operators can make wireless home broadband more attractive, and customers can churn if support slows or prices feel high. A business built on one town's trust can lose value quickly if the person who answers complaints becomes unavailable, overworked or undercapitalised.

What would change the judgement? First, verified subscriber counts by service type: wireless, fibre if any, business links, home broadband, CCTV-maintenance customers and attendance-system accounts. Second, an updated coverage map for Bahawalnagar, Chishtian and any neighbouring service areas. Third, public speed and uptime evidence from measurement platforms, not only marketing text. Fourth, details of upstream, transport and redundancy arrangements, including whether the BGP peers visible in public views reflect primary suppliers or partial route visibility. Fifth, proof that the APNIC IPv6 allocation is live for customers. Sixth, complaint and restoration data showing how the company performs during power cuts, storms and national upstream events. Seventh, any PTA licence upgrade, FLL move, merger, acquisition, partnership or dispute. Eighth, signs that PTCL, Wateen, Jazz, Ufone, Zong or a local fibre operator has materially improved fixed access in Tornado's strongest streets.

The company is therefore best read as a local dependency business, not a bandwidth story alone. Customers depend on Tornado if it is the provider that can actually reach them, install today, answer tonight and repair tomorrow. Tornado depends on larger networks, regulator continuity, imported hardware, power backup and payment discipline. Pakistan depends on thousands of such local arrangements because national broadband penetration is not the same thing as fixed, reliable, supported connectivity in every town. Tornado Networks shows the economics in miniature: the last mile is not just a cable or a radio link. It is a chain of trust, power, routing, collection and labour, and every weak link eventually becomes a phone call to the person whose number is printed on the local ISP's website.

Evidence register