Summary

  • FIXNET's contract-free proposition is economically coherent, but only as a disciplined regional access business. Its prices can undercut several visible national alternatives, yet the absence of lock-in means every unpaid installation, support contact, modem dispute and short-tenure customer lands on FIXNET's own margin.
  • The company is not just a thin marketing reseller. Public ASN, RIPE, PeeringDB and exchange records show an operating network around AS47288, Edirne, Istanbul, Bursa and Bulgaria-facing interconnection. The limit is that the last-mile access promise still leans heavily on Türk Telekom wholesale inputs outside FIXNET's own fiber footprint.
  • The judgment is clear: FIXNET's advantage is customer freedom plus regional network craft, not scale. It becomes a better business if it raises the share of on-net Edirne fiber and enterprise links; it becomes a weaker one if low headline tariffs attract high-churn households whose service costs cannot be recovered.

The payer starts with a simple calculation. A household in Edirne, Keşan, Kırklareli, Şarköy or any other address covered by Türk Telekom access can sign a long contract with the incumbent, try a national no-commitment challenger, or choose a smaller operator that says the customer can leave without a penalty. FIXNET sells that third choice. Its public pages make the offer explicit: taahhütsüz, kotasız, no Fair Use quota, no cancellation penalty, and a call-center-led relationship rather than a giant retail machine. In plain economics, the customer is buying an option.

The option is the right to cancel, move or switch if the line is poor, the price moves, the household moves, or a better fiber build reaches the building.

That option is not free. Someone has to pay for acquisition, provisioning, billing, support, wholesale access, upstream connectivity, credit-card processing, modem handling and the failed installations that never become durable monthly revenue. A long commitment lets a telecom operator finance those costs through time. A no-commitment tariff forces the operator to recover more of the cost either up front, through a higher monthly price, through a lower service cost, or through a customer base that stays voluntarily because the service is good.

FIXNET's strategic problem is therefore not whether Turkish customers like contract-free broadband. They do. The question is whether FIXNET can make a no-lock-in broadband book profitable when the customer has the freedom to leave before the operator has earned back the cost of winning and supporting that customer.

The answer is cautiously positive, but with a narrow lane. FIXNET has enough network evidence to be more than a price sticker. It has a registered autonomous system, public interconnection records, a RIPE membership presence, an open peering posture, a looking glass and visible routes. It presents itself as an Edirne-origin operator with Türk Telekom wholesale access, its own Edirne fiber and a data-center/backbone story that reaches toward Sofia and Frankfurt. Those facts matter.

They mean FIXNET has some control over routing, peering, IP resources and enterprise connectivity rather than being wholly dependent on another operator for every layer above the customer line. But the residential access proposition still depends on wholesale economics. The company says it serves customers wherever Türk Telekom infrastructure is available; it also says its own GPON fiber is in Edirne center. That is the boundary of the business. On-net customers can be a margin pool. Off-net customers can become a spread trade.

The spread trade is harsh. A wholesale-access ISP buys access on regulated and commercial terms, pays for its own aggregation and backhaul, adds support and billing, then sells the final package into a market where the incumbent, TurkNet, GIBIRNet, PoyrazNet and other challengers can all change prices. The spread is not the headline monthly fee. The spread is what remains after the wholesale port, transport, taxes, collection costs, payment friction, customer-care minutes, installation fallout, modem losses and churn are absorbed. In a high-inflation economy, the spread is also a moving target.

When Turkish CPI is running above 30 percent year on year, and when network equipment, routers, optics and much interconnection equipment are effectively exposed to hard-currency pricing, any operator that sells fixed-price comfort without a lock-in must decide who bears the inflation gap. If the customer is not locked, the operator cannot simply count on future months to repair today's underpricing.

FIXNET's public tariff stack shows why the model is attractive and why it is fragile. Its home pages show 35 Mbps VDSL at 499 TL, 100 Mbps VDSL at 549 TL, 35 Mbps fiber at 599 TL and 100 Mbps fiber at 699 TL under a no-commitment package presentation. Other public package pages and business pages show higher-speed fiber lines, including 200 Mbps, 500 Mbps and 1 Gbps tiers, with current campaign prices rising into the 1,150 TL to 1,650 TL range on some pages. The business internet page lists comparable access products for commercial users and also states a Türk Telekom connection and activation fee of 2,100 TL, charged on the first bill.

The generic application page shows a 150 TL connection fee. A live campaign application route shows a seven-month fixed-price package and a modem rental line. The details matter less than the pattern: FIXNET is using a mix of monthly tariff, connection fee, installment/payment handling and campaign structure to keep the customer-facing promise of freedom while trying not to eat the full activation bill.

That is the right instinct. No-commitment broadband fails when the operator treats a flexible monthly tariff as if it were a long-contract tariff without the contract. A new customer may need address qualification, number or service transfer checks, order entry, identity verification, customer outreach, wholesale provisioning, on-site work, modem configuration, call-backs, invoicing and the first support contact. Even where Türk Telekom does the physical access work, the retail operator owns much of the customer experience.

If the subscriber leaves after two or three months, a 499 TL or 699 TL tariff may not have recovered the acquisition cost, particularly if the operator absorbed a modem or technician visit. The more honest model is to charge enough up front, rent or recover the modem cleanly, and make the monthly tariff transparent. FIXNET appears to be moving in that direction, but the public site still gives a fragmented picture. A customer can see different fee references on different paths. That is not only a conversion problem.

It is a margin-control problem, because unclear fee presentation creates disputes at the point where the business most needs clean cash collection.

The strongest part of FIXNET's price case is that the national benchmark is not obviously lower. Türk Telekom's public 100 Mbps fiber offer visible in the current campaign context is an 18-month commitment at 800 TL, with the modem excluded and the usual commitment breakage economics in the background. TurkNet's published tariff list shows a post-June 2026 no-commitment price of 949.90 TL for its main current 16 to 100 Mbps and Gigafiber rows, with older 699.90 TL rows closed to new sales.

GIBIRNet's July 2026 update lists 100 Mbps at 780 TL after a first-three-month promotional period, with activation and other service fees disclosed separately. PoyrazNet presents 100 Mbps around 600 TL with no commitment. Against that field, FIXNET's 549 TL VDSL and 699 TL fiber 100 Mbps figures are credible rather than implausible. They are low enough to attract a price-sensitive household, but not so far below every challenger that the only possible explanation is loss-making subsidy.

The problem is that low enough is not the same as profitable enough. A 100 Mbps VDSL customer at 549 TL is probably a wholesale-spread customer unless the address is inside FIXNET's own access footprint. That customer can compare the price to PoyrazNet, TurkNet, Türk Telekom, GIBIRNet and local alternatives. If the line works, the customer may stay because switching is annoying. If the line is unstable, the customer can leave, and the no-commitment promise makes leaving part of the product. The very freedom that wins the order increases the operator's churn exposure. The profit test is not the first invoice.

It is the retained gross margin after six, twelve and eighteen months, adjusted for support and bad debt.

The better customer for FIXNET is not necessarily the cheapest household. It is the user for whom FIXNET's regional network actually changes the experience. The company emphasizes Edirne's proximity to Europe, point-to-point links from Türk Telekom POPs to its Edirne data center, low international latency, own fiber in Edirne center, Metro Ethernet, VPN, server hosting, MPLS and VoIP. Public interconnection records support that this is not empty language.

PeeringDB lists AS47288 as a cable/DSL/ISP and network-services operator with open peering, balanced traffic, 100-200 Gbps traffic level and public peering at NetIX, RegPEX and TurkIX Sofia, each shown with 100G capacity. Facilities records put FIXNET in or around Edirne, Istanbul, Bursa and Bulgarian facilities, including Sofia and Haskovo. BGP data shows AS47288 with multiple upstreams, including Türk Telekom, RETN, Hurricane Electric and the TurkIX Sofia route-server environment. The network is small compared with national carriers, but it is visible and multi-homed.

That network evidence changes the valuation of the company. A pure resale ISP lives and dies on customer acquisition and a thin regulated spread. A regional network operator can add gross margin through peering, direct transit choices, enterprise circuits, static IP, hosting, voice and local support. FIXNET's public services page points toward that second model. Metro Ethernet is not a household product; it is a paid control product for firms that need bandwidth, symmetry, committed routing and support. VPN and point-to-point services push the company into multi-site business connectivity.

VoIP brings a small but useful voice-margin layer, especially for customers moving fixed numbers into IP service. Static IP adds a visible recurring add-on. None of these turns FIXNET into a national incumbent, but they improve the mix. The more the company can sell enterprise and on-net services around its Edirne backbone, the less it has to rely on low-price, high-churn residential access.

The operating boundary is still uneven. FIXNET's own corporate page says it signed a Türk Telekom wholesale contract and delivers internet services from points where Türk Telekom infrastructure exists. The same page says it can bring real fiber optic GPON to homes over its own fiber infrastructure in central Edirne and can provide Metro Ethernet from its own fiber. That distinction should drive every economic judgment. In central Edirne, the company may control a larger part of the access stack, even if it still buys upstream and interconnection from others.

Outside that footprint, the economics look more like VAE or resale, where Türk Telekom's access network is both the essential supplier and the most powerful substitute. This is why the company cannot price all customers as if they are equal. An on-net fiber customer, a Türk Telekom VDSL customer, a business static-IP user and a multi-site VPN customer do not have the same margin.

The public package pages partially recognize this. The 1 Gbps X Fiber offer is marked for Edirne center only and carries a strikingly lower price on some visible pages than comparable higher-speed national-style fiber rows. That is exactly what one would expect if the line is on a more controlled local access footprint. When the operator owns or materially controls the access layer, it can price aggressively because the marginal cost of extra traffic is different from a wholesale bitstream line. But the public tariff presentation needs discipline.

If an Edirne-only product is priced as a local showcase, it should be clearly separated from wholesale-based fiber products so customers do not read one footprint's economics as a national promise. Confusion here would create a sales boost at the cost of complaints, call-center load and reputational drag.

The cost side starts with Türk Telekom. BTK's reference access material shows why wholesale offers matter: operators with effective market power can be required to publish access terms, conditions and fees. The current BTK page lists updated Türk Telekom reference offers for IP-level data-stream access and buy-sell xDSL/FTTx wholesale sale in June 2026, along with other access and interconnection offers. Those documents are not just regulatory paperwork. They set part of the floor under FIXNET's off-net economics.

If wholesale input prices rise, if port or activation fees move, or if provisioning terms become less favorable, a no-commitment retail tariff must absorb the shock or reprice quickly. Contracted incumbents can hide some of that pressure inside long promotional structures. A smaller no-commitment operator has less room.

The market data intensifies that point. Turkey had 21.2 million fixed broadband subscribers at the end of the first quarter of 2026, with FTTH/FTTB at 10.3 million, xDSL at 8.3 million and cable around 1.5 million, according to the sector summary based on BTK data. Türk Telekom's fiber length was reported around 550,000 kilometres, while alternative operators together had about 147,000 kilometres. That is a structural fact, not a marketing claim. Alternative ISPs may compete hard at retail, but the national physical layer remains heavily shaped by the incumbent's network.

A regional operator can be strategically important in its geography and still lack national capex scale. For FIXNET, the implication is simple: competing as if wholesale access were permanently cheap would be reckless; competing where it has local routing, local relationships and enterprise demand is more defensible.

Customer concentration is therefore a two-sided issue. On the retail side, concentration in a few towns can be a strength. Local brand recognition lowers acquisition cost, service technicians know the plant, and customer referrals can matter more than national advertising. The contact page points to Edirne addresses and a Keşan branch, while public registry-style sources also show Edirne roots and branch history. A regional density model can work if truck rolls are short, support can be handled by a compact team and the operator knows which streets, buildings and cabinets produce stable service.

But geographic concentration also concentrates operational shocks. A local outage, a Türk Telekom access problem, a power problem, a routing issue or a support backlog in the region can hit a large share of the customer base at once.

The public complaint signals show the same tension. Şikayetvar contains complaints about outages, slow speeds, support access, modem return costs, refund disputes and billing after cancellation. Those claims are not audited financial data, and they should not be read as a statistical sample of the whole customer base. They are market signals. For a no-commitment ISP, they are especially relevant because they point to the exact places where the model can leak money: cancellation, modem return, service transfer, support wait time, perceived billing unfairness and line reliability.

A complaint about a 934.55 TL modem return shipment, whether fully representative or not, is economically meaningful because it shows how a supposedly flexible relationship can become a dispute at exit. If exit is the product, exit has to be operationally clean.

Forum signals are more balanced. DonanımHaber discussions include users considering FIXNET because it is cheaper or because international latency appears better than some alternatives. One user described a roughly ten-day installation and early satisfaction with support responsiveness. Another compared FIXNET with GIBIRNet and emphasized trace routes, international latency and price changes. Again, this is not audited churn data. It is demand texture. The buyer who discusses trace routes is not the same as a household that only wants streaming and schoolwork.

FIXNET can earn above-commodity loyalty from technically aware customers if its routing and overseas paths are genuinely better. But technically aware customers are also quicker to diagnose congestion, bad peering, high ping or routing regressions. They are loyal to performance, not to the logo.

Support labour is the hidden cost that decides whether customer freedom is profitable. FIXNET's own copy leans into direct call-center communication and experienced staff. LinkedIn shows a small-company profile, roughly in the 11-50 employee range. That size can be an advantage: fewer layers, local knowledge, faster escalation, and the founder/technical team close to the network. It can also be a constraint. A few hundred extra support-heavy customers can overload a small team, especially when wholesale access faults require coordination with Türk Telekom.

Contract-free customers are less tolerant because they have not paid with captivity. If they cannot reach support, the cancellation option becomes real. That is why customer-care productivity is not a soft metric here. It is part of unit economics.

The paid unit is not just speed. FIXNET sells a bundle of speed, latency, freedom, local support and add-ons. The access speed is the easiest part to compare; the rest determines retention. A 35 Mbps VDSL plan at 499 TL is not a high-margin luxury product. It is a defensive household utility. A 100 Mbps fiber plan at 699 TL can work better if the fiber is stable and the modem/support burden is low. Higher-speed fiber at 200 Mbps, 500 Mbps and 1 Gbps gives more revenue per account, but only if the underlying access and backhaul can carry usage without degrading the lower tiers.

Static IP at 150 TL a month is a useful add-on because the incremental cost can be lower than the retail price, but it attracts heavier and more technical users. VoIP can improve account stickiness. Business connectivity can raise ARPU and justify support. The path to a stronger FIXNET is to move customers from commodity access into a broader account without recreating the coercion of a long contract.

The pricing logic has to account for inflation. Turkey's June 2026 CPI release put annual consumer inflation at 32.11 percent, with housing, water, electricity, gas and other fuels rising faster than the headline. That matters to a broadband operator because salaries, rent, electricity, vehicles, fuel, office costs and field work reset in local inflation terms. At the same time, much core network equipment is not priced like a domestic grocery basket. Routers, optics, servers, firewall equipment, software support and many data-center components have foreign-currency exposure.

A no-commitment operator can reprice more freely than an 18-month fixed-price incumbent contract, but price increases can trigger churn. The more FIXNET relies on customers who came primarily for low price, the more dangerous repricing becomes.

This is where contract-free service has a paradoxical advantage. Because the company has not promised a long fixed tariff to every customer, it can adjust retail prices faster than a heavily committed contract base. If costs move sharply, FIXNET can raise prices and let dissatisfied customers leave, rather than carry a loss-making fixed-price cohort for many months. The customer sees freedom; the operator also gets some freedom. But that works only if acquisition cost is not too high. If every customer costs heavily to acquire and install, fast repricing can destroy payback.

A flexible tariff is economically elegant when the operator has low acquisition cost, low installation subsidy, good self-service, clear modem recovery and strong voluntary retention. It is economically ugly when it is paired with expensive physical setup and weak support.

The modem issue deserves more attention than it usually gets. Modem included, modem rented, modem financed, modem returned and modem charged at exit are not customer-experience footnotes. They are balance-sheet mechanics. A national operator can finance millions of devices and spread bad returns across a huge base. A regional ISP must be more careful. FIXNET's public pages show modem-related language in several contexts, including fiber packages that say modem included and application flows that show modem monthly charges under specific campaigns. The right policy is not necessarily to give the modem away.

The right policy is to make ownership, rental, return shipping and residual charge plain before installation. If the customer buys freedom and then meets a surprise device cost at exit, the promise is damaged and support cost rises.

Payment design is another clue. FIXNET's payment-method page lists bank channels and urges automatic payment so service is not paused. Application pages include card details and installment language around installation and modem charges. This is sensible because a small ISP's working capital can be hurt by late payments and collections. The operator that sells to price-sensitive customers without commitment should care intensely about payment friction. A customer who misses a bill may still be expensive to support, especially if suspension and reconnection create calls. Automatic payment reduces that risk.

But it also creates trust obligations: invoices, price changes, one-off fees, refunds and cancellation dates must be clean. A no-commitment brand cannot afford sloppy billing because the customer has already been taught that freedom is the value proposition.

Regulation is not peripheral. BTK consumer material tells customers to check whether an operator is authorized, to understand tariff names and fees, and to use formal cancellation channels. Network security rules impose obligations on authorized operators, with additional duties tied to authorization type and scale. FIXNET's company pages say it holds ISP, infrastructure operator and fixed telephony permissions from BTK. Its RIPE and PeeringDB profiles establish network identity, but authorization and consumer compliance govern the retail relationship. This combination means FIXNET is not operating in a casual internet-access gray zone.

It is inside a regulated telecom market where consumer complaints, cancellation rights, security duties and reference-access rules shape economics.

The geopolitical angle is quieter but real. Edirne sits at Turkey's European edge. FIXNET's story repeatedly points to Europe-facing latency and direct or efficient paths to Sofia and Frankfurt. PeeringDB and TurkIX-related public material support the idea that Bulgarian interconnection is relevant to the company's network. Vivanet's TurkIX page also describes a regional model in which access from Turkey connects to peering that technically takes place in Bulgaria. For customers who care about gaming, cloud access, remote work, enterprise VPNs or cross-border business traffic, this can matter.

It does not make FIXNET immune to Türk Telekom's last-mile dominance, but it gives the company a differentiated network story. A regional ISP with an actual cross-border routing advantage has a better chance than one selling only a smaller invoice.

The risk is overclaiming the network advantage. A 100G peering presence is not the same as a 100G customer experience at every home. Peering reduces some transit costs and can improve paths to networks that participate in the same exchange. It does not fix a bad copper loop, a congested access cabinet, weak Wi-Fi inside the home, a delayed installation, a billing mistake or poor support after midnight. ASN evidence is evidence of operational capability, not proof of retail quality. FIXNET's strongest communication would separate routing advantages from access realities.

If the line rides Türk Telekom VDSL, the customer's last-mile experience is constrained by that plant. If the service is on FIXNET's own Edirne fiber, the claim is stronger. Precision would improve credibility and reduce disappointment.

Substitutes define the ceiling. Türk Telekom can sell the comfort of incumbent scale and an 18-month fixed price. TurkNet can sell national no-commitment simplicity and strong brand recognition among internet-savvy customers, even at a higher current tariff. GIBIRNet can sell a similar challenger proposition with disclosed activation and service fees. PoyrazNet can attack on low no-commitment prices. Mobile broadband and fixed wireless are imperfect substitutes, but for some households they cap the pain of a bad fixed-line experience. Local fiber builders can become sudden threats if they reach a building.

In that market, FIXNET cannot rely on customer ignorance. Its customers can compare prices, read forums, look at traceroutes and switch.

The acquisition math should be read by customer type, not by average package. A household moving from another operator over an existing line can be relatively cheap if the service-number transfer works, the modem is already adequate and the customer understands that the operator is not rebuilding the access line. A new fiber customer can be much more expensive if the order requires hardware, installation coordination and post-install Wi-Fi support. A business customer may be expensive to win but easier to retain if the service becomes part of payment terminals, cameras, cloud accounting, VPN access or branch communication.

A gamer or remote worker may be intolerant of routing changes but willing to recommend the operator if latency is stable. One headline churn number would hide all of this. The company needs customer-level payback discipline.

This is also why the 35 Mbps tier should not be treated as a harmless entry product. At 499 TL, it is a way to keep the brand accessible and to serve households whose infrastructure or budget does not justify more speed. But a low-speed line can consume the same customer-care process as a higher-speed line. It can require the same identity check, address qualification, modem discussion, payment follow-up and cancellation handling. If the subscriber calls repeatedly because a copper pair is noisy, the cost per lira of revenue can be worse than on a 100 Mbps or 200 Mbps customer. The low tier works if the customer is stable and support-light.

It fails if it is a magnet for customers who are both price-sensitive and operationally costly.

The 100 Mbps product is the real battlefield. It is the speed where most households compare offers, where remote work and streaming feel adequately served, and where the difference between 549 TL, 600 TL, 699 TL, 780 TL, 800 TL and 949.90 TL is meaningful but not always decisive. At that tier, FIXNET's message has to be sharper than price. The customer must know whether the line is VDSL or fiber, whether the modem is included or rented, what upload speed is realistic, what happens after a promotional window, and how cancellation works. Price opens the conversation; clarity closes it.

If clarity is missing, the customer will use the no-commitment right at the first disappointment.

Higher-speed fiber creates a different discipline. A 500 Mbps or 1 Gbps customer is not just paying for peak speed; that customer is likely to generate heavier usage, more Wi-Fi expectations, more speed-test disputes and more sensitivity to international routes. If the access is on FIXNET's own fiber, the higher tariff can be a strong margin product because the operator has more control over provisioning and contention. If the service relies on another operator's last-mile conditions, the risk is that FIXNET sells a premium expectation while owning only part of the delivery chain.

High-speed tiers should therefore be marketed with more precise availability and performance language than low-speed utility plans.

The strongest evidence that FIXNET understands this is the enterprise menu. Metro Ethernet, VPN, VoIP and hosting are not cosmetic add-ons. They are ways to sell control rather than just access. A small business that needs a fixed IP, camera upload, branch VPN or a stable voice number is less likely to judge the provider only by the cheapest household tariff. It wants a working service, a reachable support channel and an operator that can talk about routing and faults without a script. These customers can justify higher support effort because the revenue stack is broader.

They also create local density: the same technician, backhaul path and account relationship can support multiple services.

There is a capital-allocation lesson in that mix. FIXNET should be cautious about using residential growth to prove scale if the growth does not improve network density. Adding scattered off-net customers across Turkey may increase the subscriber count but weaken the economics if every fault is remote, every install relies on another party, and brand recognition outside the core geography requires paid marketing. Adding customers in streets, districts and business clusters where FIXNET already has fiber, technicians, enterprise relationships or known routes is different.

It can lower support cost per customer and increase the chance that a customer stays because the operator is locally useful, not merely because the first month was cheap.

Wholesale dependence also changes negotiation posture. A regional ISP cannot force the incumbent access network to behave like its own plant. It can select products, monitor faults, escalate tickets, manage customer expectations and choose where to sell aggressively. But if the underlying access provider has a provisioning delay, a cabinet issue or a maintenance window, the retail customer often blames the retail brand. That is why the no-commitment promise needs a fault-communication system. Customers forgive constraints more readily when the operator tells them what layer has failed and what is being done.

They leave when they feel trapped in silence. In this model, transparency is not a virtue signal. It is churn management.

The looking-glass and abuse-contact evidence matters for the same reason. Running a visible AS with public reachability tools means the company is participating in the operational internet, not only buying a consumer access feed. But with that status comes cost. Abuse mailboxes, route hygiene, RPKI validity, IRR records, blacklisted customer machines, bot traffic, spam, DDoS complaints and law-enforcement requests all consume labour. A network that serves both eyeball users and hosting or static-IP customers must police reputation carefully.

Cheap broadband customers can create expensive abuse work if compromised devices sit behind the network. Static-IP and hosting customers can create higher-value revenue, but they also raise the reputational stakes.

RPKI and IRR are not marketing details. They reduce routing ambiguity and help other networks decide whether FIXNET's originated prefixes should be trusted. Public data showing valid ROAs for significant netblocks is useful evidence that the operator has at least some routing discipline. The AS-FIXNET-TR set matters because peers and transit providers use route-set information to build filters. But these controls are hygiene, not a moat. Many competent networks can do them.

The moat, if there is one, comes from combining them with geography: Edirne density, Bulgaria-facing interconnection, business customers and support that understands the local plant.

The company's Europe-facing story should be monetized through the customers who notice it. A normal household may not care whether traffic to a game server or cloud application leaves through Istanbul, Sofia or another route, as long as video works. A gamer, software shop, call center, remote worker, logistics company or multi-branch business might care. Those customers can become profitable if FIXNET sells measured performance, not vague speed. The article's judgment would be stronger if FIXNET published latency ranges, route examples, availability metrics or business SLA terms in a clean commercial format.

Without those, the network story is plausible but not fully bankable.

One practical weakness is information architecture. The public site has useful material, but it asks the customer to assemble the economics from homepage cards, package pages, application forms, business terms and campaign paths. That is not unusual for a regional ISP, but it is costly. Every unclear difference between "connection fee", "activation fee", "modem included", "modem rental", "static IP", "line freeze", "current campaign" and "annual prepaid" can become a call. Calls are labour. Labour is margin.

A no-commitment company should be unusually good at showing the total first invoice, the normal monthly invoice, the cancellation process and the device-return rule before the customer presses submit.

The same clarity would protect the brand from inflation. In a market where prices move often, customers become suspicious of every invoice change. If FIXNET explains which campaigns are fixed for a defined period and which no-commitment tariffs can change, it can reprice without appearing arbitrary. If it relies on scattered pages and call-center explanations, every price increase becomes a trust event. Larger operators can survive resentment through inertia. A contract-free operator cannot. It has to make the customer believe that the lack of a contract is a fair exchange, not a cover for surprise repricing.

The firm should also be careful with "no penalty" language. It is a strong selling point, but it should not be allowed to imply that every exit cost disappears. A fair exit can still require payment for used service, return of rented equipment, settlement of a modem installment, or a lawful one-off fee that was disclosed at signup. The distinction is between a punitive lock-in charge and a transparent settlement of real costs. If the distinction is not plain, the operator will win orders with one sentence and lose reputation with the final invoice.

In this business, the last invoice is part of customer acquisition because former customers write the market's memory.

There is a second reason to avoid hidden subsidy: it attracts the wrong customer. If the upfront cost is artificially low, customers with uncertain tenure are more likely to try the service casually. Some will become loyal. Others will leave quickly, taking installation and support cost with them. A modest, transparent setup charge can improve the pool by making the customer think about tenure without trapping them. That is a better filter than a long lock-in. It preserves the freedom promise while reducing pure option-taking by customers who have no intention of staying if another discount appears next month.

On the supplier side, the company should assume that wholesale access will not save weak retail economics. Regulation can create a predictable access framework, but it does not guarantee a fat spread. The incumbent's own retail offers, national challengers and local competitors all discipline price. The operator that owns less plant has to be better at selecting customers, lowering support cost and adding services. The operator that owns more plant has to be better at filling that plant with customers who value it. FIXNET lives in both worlds. The strategic danger is managing them with one tariff psychology.

The company's small scale should not be treated only as weakness. A regional operator can sometimes make decisions faster than a national one. It can fix bad copy, call back customers, adjust routes, recognize repeated building faults and tailor business services without a committee. That agility is worth money if it reduces churn and increases trust. But agility has to be institutional, not heroic. If the model depends on a few people remembering every customer and solving every escalation manually, it will break when volume rises. FIXNET's next economic upgrade is likely process, not another slogan.

The article's conclusion therefore rests on a hard line between freedom and subsidy. Customer freedom is a product feature. Subsidized uncertainty is a financial leak. FIXNET can sell the first if it makes the second small. That means segmenting the base, publishing cleaner fees, recovering modems, pricing wholesale lines according to payback, investing in on-net density, and using its network evidence to win customers who care about more than the cheapest monthly number. The more it does those things, the more a contract-free offer becomes an advantage.

The less it does them, the more the company becomes a lender to customers who can leave before repaying the loan.

The positive scenario is not complicated. FIXNET keeps household pricing competitive but stops treating every household as equally valuable. It pushes Edirne on-net fiber where it has better control. It uses wholesale VDSL and FTTx to maintain regional reach, but it prices installation and modem recovery honestly enough that churn does not destroy payback. It turns technically aware households into advocates through stable routing and transparent outage communication. It sells more static IP, VoIP, Metro Ethernet, VPN, hosting and business access into accounts that need support and will pay for it.

It uses open peering and regional exchange presence to keep traffic costs and latency under control. It improves public tariff clarity so support does not waste labour explaining fees after the fact.

The negative scenario is equally direct. FIXNET leans too hard on low headline prices, attracts customers who are expensive to install and quick to leave, absorbs support costs on wholesale lines it cannot fully control, and then has to reprice into an angry base. Complaints about outages, support access, modem return and refunds multiply. Local goodwill erodes. The company still has a network, but the retail economics weaken because the no-commitment promise has become a churn amplifier rather than a trust engine. A regional ISP can survive thin margins; it cannot survive thin margins plus messy exits.

Facts that would reverse this judgment are specific. If FIXNET disclosed a high share of on-net fiber customers, low monthly churn, clean payback inside six months, strong automatic-payment adoption, low bad debt, transparent modem recovery and rising enterprise revenue, the company would look materially stronger. If public data showed that most customers are off-net wholesale lines with high complaint rates, weak support response, repeated billing disputes and a need to raise prices above national challengers, the company would look weaker. If Türk Telekom wholesale terms became sharply less favorable, off-net margin would compress.

If FIXNET gained more direct local fiber and stronger enterprise density, the model would improve. If a national no-commitment competitor chose to attack Edirne with local marketing and comparable routing, FIXNET's customer-acquisition cost would rise.

The final judgment is that FIXNET has a real but narrow economic proposition. Customer freedom can be profitable when it is sold by an operator with local density, controlled costs, clean fee recovery and genuine network differentiation. FIXNET has pieces of that: Edirne identity, public AS47288 evidence, 100G exchange presence, open peering, Türk Telekom wholesale access, own-fiber claims in central Edirne and a services menu beyond residential broadband.

It also has the classic fragilities of the model: inconsistent public tariff presentation, dependence on incumbent access outside its footprint, small-team support exposure, high-inflation cost pressure and complaint signals around exactly the moments when no-commitment service is tested.

FIXNET should not try to become a miniature Türk Telekom. The better strategy is to be the regional operator that makes freedom financially clean. Charge clearly for setup. Make modem economics explicit. Keep the low-price promise where the access cost supports it. Sell business and technical customers the network advantages they can actually feel. Do not hide behind the ASN when the last-mile fault is elsewhere. Do not let cancellation become a dispute. If the customer pays for the right to leave, the operator earns the right to keep that customer only by making staying easier than switching.

That is a demanding business, but it is not a bad one.

Sources