The bundled invoice
The useful way to begin with TECHNet Informática is not with an autonomous system number, a legal filing or a fibre slogan. It is with a monthly invoice in Capim Grosso. TEN Internet's public home page advertises a 600 Mb residential plan at R$79.90 a month, with TenTV, Deezer Premium, installation included subject to availability, the first monthly payment only in the following month and a router supplied on a comodato lending basis (https://teninternet.com.br/). On the same page the entry 300 Mb plan is R$69.90, while the 1 Gb plan is R$109.90. The important fact is not that these numbers are unusually high or unusually low. It is that the broadband line is being sold as part of a thicker local bargain: a fibre connection, entertainment add-ons, a supplied device, a nearby support habit and the expectation that someone in the provider's world knows the customer when something breaks.
That expectation is worth money because the local market is not empty. Radar da Telecom's Capim Grosso fixed-broadband page, which presents Anatel and IBGE-derived data and says it was updated on 3 July 2026, lists 10,938 active fixed-broadband accesses, 13 active operators and 32.9 percent penetration against a population estimate of 33,235 (https://www.radardatelecom.com/banda-larga/capim-grosso-ba). It ranks Ten Internet first in the municipality with 5,266 accesses and 48.14 percent share, followed by SKY with 4,097 accesses and 37.46 percent, and Mr. Ynet with 1,164 accesses and 10.64 percent. That is leadership, not insulation. A household that is angry enough to switch has alternatives. A small business that loses patience has other names to call.
The arithmetic of the R$79.90 line gives the first clue to the economics. If every one of Ten Internet's 5,266 Capim Grosso accesses shown by Radar were paying exactly the advertised 600 Mb residential price, the gross monthly access revenue would be about R$420,753 before taxes, discounts, delinquency, plan mix, equipment, wholesale capacity, staff, pole or route costs, payment fees and support labour. That is not a statement of actual company revenue; the public evidence does not disclose current ARPU, churn, enterprise mix or discounting. It is a scale marker. At a price below R$80, a single avoidable truck roll, a mishandled installation or an unresolved Wi-Fi complaint can consume a meaningful slice of a month's contribution from one customer. In such a business, technical support is not a courtesy attached to broadband. It is part of the product's margin.
This is the governing argument for TECHNet Informática under the TEN Internet name. The company is valuable if it can make its customers think the service is local, remembered and repairable while also operating with enough network scale to avoid being a thin reseller. It is vulnerable if its service becomes a commodity speed claim in a market where price cards can be copied and customer irritation travels quickly through WhatsApp, shop counters, Instagram, church groups and municipal workplaces. In Capim Grosso, 600 Mb is not the whole product. The defensible product is the memory of who installed it, who answers, who can troubleshoot and how quickly the provider turns a complaint back into a working connection.
A Technet name inside a TEN brand
The public identity trail is unusually useful because it shows both continuity and rebranding. TEN's own history page says the journey began in 2000, when Technet started in Capim Grosso, Bahia, offering radio-internet plans; it says the company served 14 Bahia municipalities by 2015 with residential and enterprise connectivity, became a regional fibre pioneer and later transformed into TEN Internet with coverage in 20 cities (https://teninternet.com.br/quem-somos/). That self-description should not be treated as an audited market history. It is still commercially important because it states the company's own claim about how an IT and radio-access operation became a fibre ISP with a younger consumer brand.
The legal and registry evidence points to the same operating centre. BrasilAPI's CNPJ record for 03.989.716/0001-51 names TEN INTERNET LTDA, fantasy name TEN INTERNET, active status, opening date 14 August 2000, main CNAE "Serviços de comunicação multimídia - SCM", address at Zozimo Amancio 174 in Capim Grosso, Bahia, and R$170,000 of registered capital (https://brasilapi.com.br/api/cnpj/v1/03989716000151). The same record lists KAKTOS PARTICIPACOES LTDA as a partner and Saulo Emerson de Oliveira Sousa as partner-administrator. Radar da Telecom's company page repeats the CNPJ, active Receita Federal status, Capim Grosso headquarters, opening date, capital amount, SCM activity and partner names, while also linking to ReclameAqui as a place to consult consumer complaints (https://www.radardatelecom.com/empresa/ten-internet-ltda). Those records do not explain day-to-day control by themselves, but they give a real corporate body behind the consumer brand.
The network identity preserves the older name. PeeringDB's public record for ASN 52995 names the network "TECHNet Informática", gives the website as teninternet.com.br, classifies the network as Cable/DSL/ISP, records an open peering policy, shows South America scope and lists an IRR set of AS52995:AS-TENINTERNET-ALL (https://www.peeringdb.com/api/net?asn=52995). Registro.br RDAP for AS52995 names the resource ASN52995, country BR, with registrant TEN INTERNET Ltda, the same CNPJ 03.989.716/0001-51, and related address resources including 177.47.224.0/20, 177.93.248.0/21 and 2804:4d8::/32 (https://rdap.registro.br/autnum/52995). The old Technet name and the newer TEN brand therefore appear to be not competing identities but layers of one operating story: a legacy local provider that rebranded while keeping public routing records tied to the TECHNet name.
There is one small caution in the website identity. TEN's footer and shop pages show the Capim Grosso address at Praça Zózimo Amâncio, 174, with CEP 44695-000, while the history page's "Dados cadastrais" section displays the same street address but a different CEP, 44820-003 (https://teninternet.com.br/quem-somos/ and https://teninternet.com.br/localizacao/). That discrepancy is not material to the broadband thesis, but it is the kind of small public-record mismatch that a lender, buyer or large customer would want cleaned up in diligence. Local trust is built from precise details. A provider that sells repairability should make its own public coordinates easy to reconcile.
The shopfront is part of the network
TEN's most interesting asset may not be a speed tier; it may be the public geography of its stores. The location page lists a wide spread of Bahia outlets and districts: Capim Grosso, Jacobina, Pindobaçu, Quixabeira, Saúde, Caldeirão Grande, Caém, Várzea da Roça, Mirangaba, Santa Luz, Campo Formoso, Carnaíba, Gavião, São José do Jacuípe, Junco, Itatiaia, Várzea do Poço, Várzea Nova, Ourolândia, Serrolândia, Sisalândia, Lagoa 33, Mairi, Taquarandi, Lages do Batata and other named localities, often with the same 0800 and WhatsApp-centred support apparatus (https://teninternet.com.br/localizacao/). A national reader may see a list of retail addresses. In a regional ISP, those addresses are part of the access network's social infrastructure.
The reason is simple. In a city such as Capim Grosso, a failed installation, a billing dispute or an intermittent Wi-Fi problem is not just a ticket. It is a person who may know the technician, pass the store on the way to work, discuss the problem with neighbours and compare the experience with a rival's promise. Physical stores raise cost: rent, staff, cash handling, local management and the discipline of keeping hours. But they also lower certain risks. They make a provider reachable to customers who do not want to solve everything through an app. They give small business owners somewhere to escalate. They let the provider sell availability street by street rather than merely postcode by postcode.
TEN's own support language leans heavily into that model. The home page and history page advertise WhatsApp, 0800 022 2290, chat online, stores and a second-copy boleto path as customer-service channels (https://teninternet.com.br/). The company's public testimonials are self-selected, not an independent satisfaction survey; they should be read as marketing copy. Yet the themes TEN chooses to display are revealing: agile attendance, attentive service, technical confidence and "desde sempre" local loyalty. Marketing departments do not create value by themselves, but they tell us what the company believes it can sell. TEN is not only selling "ultra velocidade". It is selling the comfort of a provider that will be present when the connection matters.
This is where the IT-to-ISP crossover matters. A company that began with radio internet and local technology support can carry a different service memory from a company that arrived as a pure fibre overbuilder. In many Brazilian towns, the provider that installed the first office network, configured routers, supported a small retailer's card terminal or fixed a school connection becomes the default person to ask before a formal tender or household subscription. That familiarity can be economically powerful. It lowers acquisition cost, improves conversion from inquiry to installation and allows the provider to sell adjacent services without starting from zero. But it is also fragile. If the support promise is stretched across too many towns, the old familiarity becomes a liability because customers remember what the company used to be able to do.
The store map therefore cuts both ways. It demonstrates reach beyond a single headquarters town, which is positive for scale. It also implies a distributed operating burden. Capim Grosso may be the anchor, but a regional footprint forces the company to manage field technicians, spare routers, fibre repairs, power events, payment collection and local escalation across many communities. A consumer brand can make that look simple. The economics are not simple. A support-heavy regional ISP only earns its reputation if the service desk and field teams remain faster than the irritation curve.
AS52995 is a control surface, not decoration
The technical records improve the case because they show that TEN is not merely a storefront reselling someone else's access with a local wrapper. Registro.br RDAP ties AS52995 to TEN INTERNET Ltda and the same CNPJ as the company record, with registration on 24 May 2011 and last change on 25 April 2025 (https://rdap.registro.br/autnum/52995). RDAP also links the autonomous system to IPv4 blocks 177.47.224.0/20 and 177.93.248.0/21, and to the IPv6 block 2804:4d8::/32. RIPEstat's AS overview reports AS52995 as "AS52995 - TEN INTERNET Ltda" and marked as announced (https://stat.ripe.net/data/as-overview/data.json?resource=AS52995). Its announced-prefixes view showed 29 prefixes over the 19 June 2026 to 3 July 2026 observation window when retrieved (https://stat.ripe.net/data/announced-prefixes/data.json?resource=AS52995).
PeeringDB adds the commercial-network layer. It records 300 suggested IPv4 prefix limits, 100 suggested IPv6 prefix limits, 100-200 Gbps traffic, a balanced traffic ratio, open peering and four IX.br exchange presences (https://www.peeringdb.com/api/net?asn=52995). The PeeringDB netixlan record lists operational 10G ports at IX.br São Paulo, Salvador, Fortaleza and Rio de Janeiro, with both IPv4 and IPv6 addresses on each public exchange record (https://www.peeringdb.com/api/netixlan?net_id=16332). BGP.tools, a route-intelligence source rather than a company filing, describes AS52995 as an active eyeball network registered on 24 May 2011, originating 22 IPv4 and 7 IPv6 prefixes, peering with 76 other networks and using five observed upstream carriers including Lumen, VGRNET, Lumiar, CLIENT.CO and VOANET (https://bgp.tools/as/52995).
These records should be read carefully. They do not prove that every customer gets a flawless path to every destination. They do not disclose paid transit terms, packet loss, private interconnection, congestion, last-mile faults or service-level compliance. But they do show the presence of a real routing surface. For a regional ISP, that matters. An operator with its own AS, allocated resources, multiple exchange presences and route visibility has levers that a simple local reseller lacks. It can peer closer to content, manage upstream diversity, tune traffic exchange and reduce dependence on a single wholesale path. It can also fail in more sophisticated ways, but at least the knobs exist.
The geographic pattern is especially important. A Bahia-focused residential provider with exchange presence in Salvador has a nearby regional interconnection option; São Paulo remains Brazil's major traffic exchange centre; Fortaleza and Rio add more route diversity and content reach. The 10G figures in PeeringDB are port-speed records, not utilization evidence. Still, they are enough to say that TEN's economics are not only street-level. The company has a regional access footprint tied to a broader Brazilian internet exchange map. That combination is the attractive part of the story: a provider close enough to know customers, but technically developed enough to operate as a recognisable network.
The caution is supplier dependence. BGP.tools' observed upstream list is useful, but observed routing does not equal signed contract depth. A diligence team would still ask for transit agreements, commits, renewal dates, price per Mbps or Gbps, burst terms, DDOS arrangements, outage histories, route-policy documents and incident reports. RIPEstat's neighbour view showed 38 unique neighbours and 26 uncertain neighbours at the 3 July 2026 observation point (https://stat.ripe.net/data/asn-neighbours/data.json?resource=AS52995). That is healthy as a visibility signal, but uncertainty in route inference is not a substitute for contract evidence. Public routing records prove a network exists. They do not prove the cost base.
The price ladder sells calm, not only capacity
The residential price card creates a narrow but telling revenue ladder. TEN's public home page offers 300 Mb at R$69.90, 600 Mb at R$79.90 and 1 Gb at R$109.90, all wrapped in a consumer bundle that can include TenTV, Skeelo, Deezer Premium, Wi-Fi Premium or out-of-home Wi-Fi depending on the plan presentation (https://teninternet.com.br/). The R$10 step from 300 Mb to 600 Mb is small relative to the nominal speed jump, which suggests a familiar ISP tactic: make the middle plan feel like the obvious choice. The R$30 step from 600 Mb to 1 Gb then monetizes heavier users without asking the entire base to pay gigabit prices.
That ladder only works if the provider controls the surrounding cost. Customer premises equipment matters because the modem-router is lent on comodato rather than sold outright; the provider therefore carries an asset, retrieval and replacement problem. Installation matters because the public offer says installation is included subject to availability; the provider must decide which addresses are cheap enough to connect, which drops are worth promotional subsidy and how to handle customers at the edge of the network. Support matters because a 600 Mb line that tests poorly over a bad home Wi-Fi layout becomes a service complaint even when the fibre itself is healthy. Billing matters because a R$79.90 plan has little room for long delinquency, manual collection or repeated card failures.
The company tries to lift the ceiling through enterprise products. Its business page advertises a 1 Gb "Plano Exclusive" at R$199.90 a month, with fixed IP, priority service, a Wi-Fi 6 Mesh router, public IP and support seven days a week from 07:30 to 00:00; it also lists a 200 Mb business plan at R$99.90 and a 500 Mb plan at R$139.90 (https://teninternet.com.br/internet-empresas/). The page says dedicated links are available and presents fibre, dedicated internet and radio as possible business access types. Separately, TEN's "outras soluções" page markets dedicated link and TenSeg camera monitoring with remote mobile access, cloud recordings and alarm integration (https://teninternet.com.br/outras-solucoes/).
The enterprise page is commercially more important than its modest prices suggest. A small retailer, municipal office, school contractor or clinic may not need a national enterprise telecom contract, but it may pay for a static address, priority support, camera monitoring or a dedicated link if the provider is already known. This is where local technical memory turns into margin. A company that knows the customer's building, prior trouble calls and payment habits can sell a broader support bundle than a price-comparison page can capture. It can also overpromise. Fixed IP, dedicated access, cameras and alarm integration create expectations that the provider can protect the customer during outages, security incidents and power problems. The higher-margin product is also the product that punishes weak operations fastest.
The contract page adds another useful clue. TEN publishes a page for contracts and regulations, including Technet-branded SCM/SVA documents, multimedia communication contracts, comodato contracts, Anatel authorization material, security guidance and service contracts for camera monitoring and vehicle tracking (https://teninternet.com.br/contratos/). The page does not by itself prove current subscriber terms, and individual PDFs would need separate review for legal reliance. But the existence of those documents supports the view that TEN's business is not only a consumer landing page. It has the paperwork footprint of a regulated communications provider selling recurring services, equipment lending and adjacent monitoring products.
Capim Grosso is a fortress with doors
The strongest market number is the local one: 5,266 fixed-broadband accesses and 48.14 percent share for Ten Internet in Capim Grosso, according to Radar's Anatel-derived municipal page (https://www.radardatelecom.com/banda-larga/capim-grosso-ba). A provider with nearly half a town's fixed-broadband accesses can spread local advertising, support labour, shopfront awareness and maintenance over a large base. It can be the default answer in neighbourhood conversation. It can decide which streets deserve faster upgrades because it has enough households nearby to justify the work. It can also attract every competitor's attention because the share is visible and valuable.
SKY's 4,097 accesses and 37.46 percent share make the competitive point. TEN does not face a tiny long tail; it faces a significant second operator. Mr. Ynet's 1,164 accesses and 10.64 percent share add a local challenger below the top two, while the same ranking includes Start X Net Telecom, Friends Telecom, OI, Starlink Brazil, Claro, Hughes and smaller names. That mix matters because substitution has different forms. A customer can move to another fixed provider, to a satellite product, to mobile or fixed-wireless substitution, or to a national brand with a call centre and promotion. The more options exist, the less a provider can rely on customer inertia alone.
Radar's same page says fibre in Capim Grosso is available from 50 Mb to 1 Gb, with 4G/LTE up to 100 Mb, 5G up to 1 Gb and coaxial cable up to 300 Mb listed as available technologies. That does not mean every address has every option. It does mean the speed claim is not enough. A consumer comparing "up to 1 Gb" products may care more about installation speed, Wi-Fi setup, billing, support response and whether the provider has a store nearby. This is why TEN's local support claim is not an adornment. It is the way a regional operator tries to keep a commodity product from being bought only on headline speed and price.
There is also a national context. Radar's company page says its data are from Anatel, Cadastro de Prestadoras and BrasilAPI, and it does not present any state-level access loss for Ten Internet in the visible "Estados com perda de base" section at retrieval (https://www.radardatelecom.com/empresa/ten-internet-ltda). That should be read only as a third-party market signal, not as an audited growth statement. Still, it sits comfortably with TEN's own multi-location footprint and the local lead in Capim Grosso. The company looks like a regional Bahia operator with enough density in its anchor market to matter, but not enough scale nationally to dictate market terms.
The most dangerous interpretation would be to treat the 48 percent Capim Grosso share as a moat by itself. Market share without customer love can be sold down quickly when a rival cuts price, funds installations or targets apartment buildings and small-business clusters. Market share with support memory is different. A local customer who believes a provider's technician knows the house, the router, the last outage and the billing history has more reason to stay through a rival promotion. TEN's job is to make that memory operationally true, not merely nostalgic.
The day the support desk fills
The failure scenario for TEN is not dramatic sabotage. It is an ordinary regional-network bad day that becomes a churn event. Imagine a backhaul fault, upstream congestion or power problem affecting several Bahia towns during a workday. Residential customers start messaging WhatsApp. Business customers with card machines and cloud systems call the 0800 number. Stores in Capim Grosso and surrounding towns receive walk-ins. A school office wants to know whether online systems are safe for the afternoon. A small retailer asks whether camera monitoring is still recording. Field technicians do not yet know whether the fault is a local fibre cut, a power issue, a router problem, an IX/backhaul problem or a supplier incident. The provider's public promise of fast, attentive support suddenly turns from marketing into a labour test.
In that scenario, TEN's scale helps and hurts. The company has enough local presence to communicate, triage and physically visit customers where needed. It also has enough customers that a correlated fault creates a flood. The business-page promise of priority support from 07:30 to 00:00 is commercially useful in ordinary times (https://teninternet.com.br/internet-empresas/). During an area incident, it creates a hierarchy problem: who receives the first call back, the residential customer who pays R$79.90, the small business paying R$199.90, the dedicated-link customer with a separate contract, or the store manager whose payment terminal is down? A support-heavy regional ISP must have an answer before the incident, because customers form their judgement during the first hour.
The migration path is equally concrete. In Capim Grosso, Radar lists SKY, Mr. Ynet, Starlink Brazil, OI, Claro and other operators alongside TEN (https://www.radardatelecom.com/banda-larga/capim-grosso-ba). A frustrated household may not wait for a root-cause explanation if a rival can install quickly. A business may split connections or move the primary link if a failure interrupts sales. A public office may demand service records before renewing. The provider can recover from one outage if it communicates well, offers credible repair information and preserves trust. It cannot repeatedly ask customers to accept a local premium while behaving like an anonymous commodity provider.
This is why the technical evidence matters to the customer-facing thesis. AS52995, four IX.br presences and multiple observed upstreams should, in principle, give TEN more ways to route around some problems than a provider with one wholesale dependency (https://www.peeringdb.com/api/netixlan?net_id=16332 and https://bgp.tools/as/52995). But the customer does not buy BGP architecture. The customer buys the absence of confusion. If TEN's network control translates into faster diagnosis, better route diversity and fewer vague explanations, it supports the brand. If it does not, it remains an impressive record that customers never feel.
Regulation is becoming part of the cost base
Brazil's fixed-broadband market has been unusually favourable to regional ISPs, but the regulatory environment is becoming less tolerant of informality and poor records. Anatel's Internal Resolution 449, dated 27 June 2025, approved an action plan to combat unfair competition and regularize fixed-broadband SCM provision; the text expressly notes the role of small providers in broadband expansion, while also pointing to concerns over regulatory obligations, sector data, consumer security and regular authorisation (https://informacoes.anatel.gov.br/legislacao/resolucoes-internas/2030-resolucao-interna-449). That is not a TEN-specific sanction. It is the sector's direction of travel.
The same direction appears in Anatel's pole-use data collection. Anatel states that, starting 1 December 2025, it began collecting information on contracts for shared pole infrastructure between telecom providers and electricity distributors, with the goal of updating records and building a positive register of regular providers from the standpoint of shared electric-sector infrastructure (https://www.gov.br/anatel/pt-br/dados/infraestrutura/coleta-de-dados-contratos-de-uso-de-postes). It says the submission is mandatory for all SCM providers that use poles shared with electric distributors, regardless of size. For a regional fibre operator, pole access, contract regularity and route documentation are not paperwork footnotes. They affect expansion cost, outage repair, saleability and the risk that an overbuilt route becomes legally or physically contested.
The public evidence reviewed here does not establish TEN's pole-contract status, pole-route condition or any regulatory breach. It does establish that TEN is an SCM provider by CNPJ activity, publishes Anatel authorization and contract material on its own site, and operates a substantial fixed-access base in at least its anchor municipality. For a lender or acquirer, that combination would make infrastructure documentation a central diligence request. A clean route book and pole-use file can turn a local provider into a financeable platform. A messy file can turn the same subscriber base into a discounted, repair-heavy asset.
Regulatory formalisation may also help stronger operators. If Anatel's push reduces informal competition, providers with documented authorisation, visible contracts, public support channels and network resources may benefit. But formalisation costs money. It demands better records, legal attention, engineering documentation and timely reporting. TEN's advantage, if it has one, is that it already looks more formal than a hobby provider: active CNPJ, AS52995, public contracts, visible stores and PeeringDB/RDAP records. The remaining question is whether the internal route, support and customer data are as orderly as the public-facing pieces.
The institutional customer is the quiet swing factor
The household invoice is the cleanest starting point, but the more interesting margin may sit in small institutions. TEN's public materials do not name a school district, hospital, public office or manufacturer as a current customer, so none should be asserted. Yet the business page and support footprint show the kind of product set a regional provider would use to sell into those customers: fixed/public IP, priority attendance, a long daily support window, Wi-Fi equipment, dedicated-link availability, camera monitoring and the ability to speak to someone nearby (https://teninternet.com.br/internet-empresas/ and https://teninternet.com.br/outras-solucoes/). A household may compare a 600 Mb plan with a rival's promotional offer. A small institution compares the provider's memory of its building, the cost of downtime, the ease of reaching support and the practical pain of changing router, firewall, payment, camera or phone arrangements.
This is where TECHNet's origin story has economic force. A provider that began as a local technical operator is more likely to have accumulated what might be called configuration memory: where the router sits, which rooms have poor Wi-Fi, which business needs a fixed address, which customer has old camera equipment, which street has difficult drops, which account has a recurring payment problem, which branch has unreliable power. None of that is visible in RDAP or PeeringDB. But it is exactly the knowledge that lets a regional provider turn a low-priced access line into a broader support relationship. If the company can keep that memory organised and available to staff, it reduces repeat diagnosis and raises switching costs. If the knowledge remains informal in a few employees' heads, it becomes a key-person risk.
The price spread illustrates why institutions matter. TEN's 600 Mb residential headline of R$79.90 is less than half the R$199.90 business 1 Gb offer shown on the enterprise page (https://teninternet.com.br/ and https://teninternet.com.br/internet-empresas/). The business plan also carries a heavier service promise. On paper, the extra R$120 a month can pay for priority attention, a better router, more complex support and a fixed or public IP feature. In practice, the provider must earn that extra money by preventing the business customer from feeling abandoned when the link becomes operationally important. A small enterprise account that calls often can be less profitable than a quiet household; a quiet enterprise account that trusts the provider can be much more valuable than its price suggests.
The same pattern applies to camera monitoring and security-adjacent services. TenSeg is presented as camera monitoring with remote real-time mobile access, cloud recordings and alarm integration (https://teninternet.com.br/outras-solucoes/). Such a product creates a different relationship from ordinary broadband. The provider is no longer just carrying entertainment, schoolwork and messaging. It is touching the customer's sense of physical security and business continuity. That can be attractive because it deepens the account and gives the provider another reason to stay in contact. It can also raise reputational risk. A broadband outage is irritating. A camera system that fails during a break-in, a shop dispute or a cash-office incident can become a more emotional grievance.
Public complaint and review surfaces should be handled with restraint. TEN's site displays favourable testimonials and points users toward Google reviews; Radar's company page points to ReclameAqui for complaints (https://teninternet.com.br/ and https://www.radardatelecom.com/empresa/ten-internet-ltda). These are not reliable statistical samples. Satisfied customers are prompted, angry customers self-select, and the loudest cases can distort the median experience. But in a regional ISP, customer chatter is not irrelevant merely because it is messy. It is part of the acquisition and retention environment. A local provider can spend years turning support into a reputation and lose a surprising amount of that value if the town begins to repeat a simpler story: nobody answers, the router is blamed, the technician does not come, or the bill is confusing.
For that reason, the best management indicator would not be a single review score. It would be a set of operating ratios: repeat trouble calls by town, first-contact resolution, installations completed on the promised date, median time to repair, business-customer escalation count, CPE replacement rate, outage communications time, credit issuance and churn in the 90 days after a major incident. Those are not public in the available record. Still, the public structure tells us why they matter. TEN has chosen a high-touch brand. It has stores, WhatsApp, chat, 0800 support, business support hours and nearby physical offices. A low-touch competitor can disappoint customers on price alone. A high-touch competitor disappoints them against its own promise.
What a buyer or lender would underwrite
A buyer, lender, large enterprise customer or regulator would pay attention to five things. First, the Capim Grosso access base: 5,266 accesses and 48.14 percent share in a 10,938-access municipality is a real anchor (https://www.radardatelecom.com/banda-larga/capim-grosso-ba). Second, the identity continuity: TEN Internet Ltda's active CNPJ, 2000 opening date and SCM activity align with the company's own history of Technet beginning in 2000 and becoming TEN Internet (https://brasilapi.com.br/api/cnpj/v1/03989716000151 and https://teninternet.com.br/quem-somos/). Third, the network surface: AS52995, announced prefixes, IPv4 and IPv6 resources, PeeringDB traffic range and four IX.br 10G records suggest technical control rather than pure resale (https://rdap.registro.br/autnum/52995, https://stat.ripe.net/data/announced-prefixes/data.json?resource=AS52995 and https://www.peeringdb.com/api/netixlan?net_id=16332). Fourth, the store and support footprint suggests local acquisition and retention capacity (https://teninternet.com.br/localizacao/). Fifth, the enterprise page and adjacent monitoring products hint at higher-value services beyond household access (https://teninternet.com.br/internet-empresas/ and https://teninternet.com.br/outras-solucoes/).
The same diligence team would discount what the public record does not answer. It would want town-by-town subscribers, ARPU, gross margin, churn, delinquency, installation cost, support tickets per thousand customers, truck rolls, median repair time, enterprise revenue, dedicated-link contracts, camera-monitoring economics, CPE inventory, pole-use agreements, route maps, fibre lease obligations, upstream/transit contracts, peering utilization, incident logs, cyber controls and customer-complaint trends. It would refuse to underwrite a support premium from testimonials alone. TEN's public testimonials and review prompts are market signals, not audited satisfaction evidence. Radar's market pages are useful third-party signals, but they are not a substitute for raw Anatel submissions and internal billing ledgers. The value case is good enough to investigate; it is not complete enough to price precisely.
The hardest underwriting question is whether support memory scales. A founder-led or locally known provider can often serve one anchor town brilliantly. Scaling that habit across 20-plus places is harder. The company must keep enough local touch to defend churn while standardising enough operations to avoid cost creep. Too much local improvisation produces inconsistent service and weak data. Too much central standardisation makes the brand feel remote. TECHNet/TEN's economic sweet spot lies between those extremes: formal enough to operate AS52995 and satisfy regulators, local enough that a customer still believes someone remembers the last problem.
Public evidence register
The key public evidence is compact but varied. TEN's home page supports current residential pricing, bundle construction, installation and comodato language: https://teninternet.com.br/. TEN's history page supports the 2000 Technet origin, Capim Grosso start, 2015 municipal expansion, fibre-pioneer claim and TEN rebrand story: https://teninternet.com.br/quem-somos/. BrasilAPI supports the active legal identity, CNPJ, opening date, SCM activity, address, capital and partner record: https://brasilapi.com.br/api/cnpj/v1/03989716000151. Radar's company page supports third-party company profile data and visible market-signal framing: https://www.radardatelecom.com/empresa/ten-internet-ltda. Radar's Capim Grosso page supports the municipal fixed-broadband access count, operator ranking, TEN share and competitor list: https://www.radardatelecom.com/banda-larga/capim-grosso-ba.
The network evidence is also public. Registro.br RDAP supports AS52995, TEN Internet Ltda as registrant and related IPv4/IPv6 resources: https://rdap.registro.br/autnum/52995. RIPEstat supports AS52995 being announced and the 29-prefix observation window: https://stat.ripe.net/data/as-overview/data.json?resource=AS52995 and https://stat.ripe.net/data/announced-prefixes/data.json?resource=AS52995. PeeringDB supports the TECHNet Informática network name, website, traffic band, open peering policy and four-exchange count: https://www.peeringdb.com/api/net?asn=52995. PeeringDB netixlan supports the São Paulo, Salvador, Fortaleza and Rio de Janeiro IX.br records and 10G port speeds: https://www.peeringdb.com/api/netixlan?net_id=16332. BGP.tools supports the active eyeball-network description, observed prefix counts, peer count and upstream list: https://bgp.tools/as/52995. TEN's business page and contracts page support enterprise pricing, priority-support claims, fixed/public IP language, dedicated-link availability and contract-document footprint: https://teninternet.com.br/internet-empresas/ and https://teninternet.com.br/contratos/. Anatel's 2025 regularisation and pole-use pages support the sector regulatory context: https://informacoes.anatel.gov.br/legislacao/resolucoes-internas/2030-resolucao-interna-449 and https://www.gov.br/anatel/pt-br/dados/infraestrutura/coleta-de-dados-contratos-de-uso-de-postes.
The fact that would change the judgement
The single fact that would most change the judgement is a town-by-town cohort file linking accesses, ARPU, churn, support load and gross margin. If TEN's Capim Grosso lead consists of loyal customers paying near-list prices, with low delinquency, low repeat support contacts, orderly pole records and a healthy enterprise layer, TECHNet/TEN looks like a high-quality regional platform whose local support memory deserves a premium. If the same access base is held by discounts, heavy CPE subsidy, overloaded support, high truck-roll frequency, weak pole documentation or expensive wholesale dependence, the company is still important but the margin is much thinner than the brand suggests.
Until that data is visible, the fair conclusion is balanced. TECHNet Informática, through TEN Internet, has real public evidence behind its identity, services and network. It has an anchor market where it leads rather than merely participates. It has AS52995, IX.br presence and enough store-level visibility to make support memory plausible. But its value is not the promise of "the best internet" in Bahia. Its value is whether, month after month, a R$79.90 household line and a R$199.90 business line are supported cheaply enough, repaired quickly enough and remembered locally enough that customers do not reduce the decision to the next rival's installation offer.