1.FM AG and the Price of Owning Network Reliability

Summary

• 1.FM AG is a Swiss online-radio company, not a demonstrated retail internet provider. Its catalogue of more than 65 music channels, free ad-supported access and a low-priced premium tier make attention the product: listeners contribute time or subscription money, while advertisers pay for access to that attention.
• The company has more network substance than a pure brand layered on rented hosting. It is a RIPE NCC member, holds AS60311 and has substantial IPv4 and IPv6 resources. Those records improve portability and routing control, but they do not establish ownership of fibre, data centres or a content-delivery backbone.
• Independence is partial. Public route-observation services show one current upstream for AS60311, NTS Workspace AG. The web service is now presented through the 1cloud.fm domain behind Cloudflare, and a current stream hostname resolves through an outside audio delivery network. Each supplier can add resilience within its own system while leaving 1.FM exposed to contract, integration and concentration risk.
• The financial case cannot be proved from public accounts because revenue, subscriber count, advertising fill, traffic, churn, royalty expense and capital spending are undisclosed. At the App Store's stated $1.99 monthly price, even 10,000 subscribers would generate only $19,900 a month before store commission, tax, rights, bandwidth and labour. Scale and retention therefore matter more than the nominal price.
• A service disruption described by a person presenting himself as 1.FM's technical lead in late 2025 is the most revealing operating evidence. He attributed several hours of website downtime and a degraded iOS service to trouble in a VMware cluster and an unresolved ISP dependency, while directing users to a replacement site. The account is unofficial, but the subsequent domain change and app updates make the episode a serious diligence signal rather than proof of permanent weakness.

The incentive: own the address, rent the scale

An internet radio station can be built almost entirely from rented parts. A company can lease a server, use a third-party streaming host, place a player on a website, submit the station to aggregators and buy advertising insertion. That route keeps capital needs low. It also leaves the station's continuity tied to account credentials, supplier support, hostnames and contracts controlled elsewhere. If the server provider withdraws a machine, the stream host changes a URL or an aggregator loses the feed, the station can still own its music programming while losing the audience.

1.FM chose a more involved position. The company became a member of the RIPE NCC, obtained its own autonomous-system number and maintains Internet address resources. Economically, that creates an option. Addresses announced under the company's routing identity can, in principle, remain associated with 1.FM when equipment or transit arrangements change. Engineers have more control over route announcements, abuse contacts and the way services are reached. The company is less dependent on a single rented address assigned from a hosting provider's block.

That option is not the same as self-sufficiency. An autonomous system still needs other networks to carry traffic to listeners. Servers still need power, storage, switches, security, replacement parts and people. A radio service also needs audio delivery close enough to users that a connection remains stable while they commute, change mobile cells or listen through a smart speaker. Owning network identity can reduce one type of switching cost while increasing the fixed cost of engineering, membership, monitoring and compliance.

The right comparison is therefore not ownership against zero cost. It is selective ownership against three alternatives. One is pure resale, where almost every technical layer is rented and the operator concentrates on channels and promotion. A second is a large public cloud and content-delivery network, where resilience is bought by usage and the supplier owns the network. A third is a hybrid: retain addresses, routing and critical control services, but distribute listener traffic through specialist audio networks and edge providers. The public evidence suggests 1.FM has moved toward that hybrid.

The commercial test is whether the hybrid lowers the total cost of failure by more than it raises normal operating cost. Listeners do not usually reward a radio station for holding an address block. They reward it by staying when the stream starts quickly, by returning to a familiar channel and, occasionally, by paying to remove advertising. Advertisers care about completed impressions and credible audience measurement. The network assets create value only through those outcomes.

Who pays is easier to identify than how much. Free listeners pay with exposure to advertising and with data required to deliver or measure it. Premium listeners pay a monthly fee. Advertisers and intermediaries fund the free service when an impression is filled. 1.FM pays rights holders, infrastructure suppliers, distributors, app stores and staff before retaining anything for equipment refresh or profit. When the chain fails, the immediate cost is asymmetric: a listener switches to another service; the company loses a session, an impression, goodwill and perhaps a subscriber.

A clear legal identity with a narrow operating purpose

The company itself is not ambiguous. Swiss commercial-register summaries identify 1.FM AG as an active company limited by shares in Baar, canton Zug, with UID CHE-287.569.579. It entered the register in January 2011. Its stated purpose covers activities connected with internet media, especially internet radio. Mathias Benedikt Blom has been the sole registered board member with individual signing authority since January 2022. The record also lists the translated names 1.FM SA, 1.FM Ltd. and 1.FM GmbH, but the operative public name is 1.FM AG. (Swiss register summary, business-register record)

That identity matters because the same name could otherwise be mistaken for one of many terrestrial stations using “One FM.” Here, the legal purpose, website, app-store developer and RIPE membership converge on the same Baar company. The Google Play listing names 1.FM AG, gives the Baar address and identifies the product as an online-radio service. Apple's App Store listing names the same developer. RIPE gives the same street address and a network-operations contact.

The operating boundary is narrower than the category in which this article appears. A RIPE membership record does not show that 1.FM sells broadband, transit, cloud services or managed networks to small businesses. The commercial register says internet media, not retail connectivity. The apps offer music channels, not access lines. Public routing data shows no downstream networks. The most defensible identity is a digital broadcaster with its own network resources.

That distinction changes the economics. A regional internet provider normally monetises recurring access contracts and can spread backbone cost across households or businesses. 1.FM monetises listening. Network traffic is a cost of delivering the product rather than the product itself. A stream that runs flawlessly but attracts little audience is not valuable; a popular stream that depends on expensive delivery can grow revenue and lose cash at the same time.

The corporate boundary also leaves important questions unanswered. Swiss private-company records do not disclose 1.FM's ownership, current share capital economics, revenue, profit, debt, headcount or related-party transactions in the materials reviewed here. LinkedIn presents a company-size band of 51 to 200 but displays eight associated people; neither figure is audited. An unofficial technical account later described one full-time developer. These measures refer to different things and cannot be reconciled from public evidence. They should not be combined into a confident staff estimate.

There is a similar difference between headquarters and operations. Baar is the registered seat and the location in the RIPE record. The public network footprint includes Swiss and US-associated address ranges; the apps reach global users; current web and stream delivery uses outside infrastructure. Nothing in those facts proves where every server, employee or contractor sits. A listener-facing company can be Swiss in law, international in audience and distributed in operations at the same time.

The product is programmed attention

1.FM's current Google Play description says the service offers more than 65 music channels covering different genres and moods. It advertises Android Auto and Chromecast support, free access with registration, song requests, favourite artists and dedications. The store shows more than 100,000 downloads. The Apple listing describes the same broad model and offers a premium service with advertising removed and higher quality across web and mobile devices.

This is not an on-demand catalogue in the Spotify or Apple Music sense. It is programmed radio. A listener chooses a channel such as trance, chillout, country, blues, baroque or classic rock and accepts a sequence selected by the operator. That limitation is also the product. The listener avoids the work of building a playlist, and a genre channel can create a recognisable mood over many hours. The company's asset is not ownership of the recordings; it is the combination of curation, channel brands, distribution relationships, user habits and reliable delivery.

The revenue model has at least two visible legs. The first is advertising. Google labels the app as containing ads, and historical advertising material reproduced in a 2016 Sonos community discussion described 30- and 60-second audio spots, banners, geographic targeting and campaign reporting. That material also claimed more than 400,000 monthly unique listeners and more than 37 channels. It is useful evidence of the intended model, but it is ten years old, no longer available on the current site and not a safe measure of today's audience.

The second leg is subscription. Apple's current listing states a price of $1.99 a month for ad-free, better-quality listening. That is a low-friction conversion offer: a listener can keep the same channels and remove the main annoyance without paying the price of a full music service. The positioning is clear when compared with current Swiss prices. Apple Music lists CHF13.90 a month for an individual plan, while Spotify lists CHF15.95. Those services provide much broader catalogues, search, personal libraries and offline features, so the prices are not directly comparable. They establish that 1.FM is selling a lighter radio experience at a fraction of the full-service price.

There may be a third leg in branded radio. 1.FM's LinkedIn page lists “Brand Radio” among its specialties. No current rate card, customer list or active product page reviewed for this article substantiates revenue from that activity, so it should be treated as a capability claim rather than a proven segment. The same caution applies to hosts, exclusive shows and premium channels described in older company copy. A catalogue description can show strategy; only contracts and accounts show contribution.

The model has attractive operating leverage. One curated channel can serve many simultaneous listeners. A piece of scheduling work is reused every hour the channel runs. The same app, account system and support site can distribute dozens of stations. An additional listener does not require another presenter or sales employee.

Yet the marginal cost is not zero. Each listening hour consumes delivery capacity and can generate rights expense, advertising-technology fees, payment cost and support load. More channels also mean more playlist management, metadata, monitoring and licensing records. The SUISA guidance says supplementary webcasting stations are settled separately and broadcasters must report the music they use. A 65-channel catalogue is therefore not merely 65 labels on one cheap stream. It broadens choice while multiplying operating surfaces.

The strategic trade is between depth and sprawl. A narrow service with ten strong channels could concentrate audience, ad inventory and engineering attention. A 65-channel network can capture more niche searches and listening moods, but many channels may be too small to monetise well. Public data does not reveal listening by channel, so it is impossible to tell whether the long tail raises retention or dilutes resources.

Revenue is hidden, so unit economics must carry the analysis

1.FM does not publish the figures needed for a conventional financial assessment. There is no public revenue series, gross margin, operating profit, cash balance, subscriber count, monthly active audience, advertising fill rate, average price, churn, listening hours or capital expenditure. Modelled estimates from commercial databases are not a substitute for filed accounts and are not used here.

The absence is not evidence of distress. It is a limit on judgment. A private Swiss company can operate for years without giving listeners or outsiders the economics that a listed media company would disclose. The correct response is to test the visible prices and costs with scenarios rather than invent a valuation.

Start with premium. At $1.99 a month, 10,000 paying accounts would produce $19,900 of gross monthly billings, or $238,800 a year, before tax, refunds and currency effects. If a 15% store commission applied to every account, the company would retain about $16,915 a month before music rights, traffic, engineering and overhead. At 100,000 accounts, those amounts become $199,000 gross and about $169,150 after a 15% commission, still before all other costs.

The 15% is a transparent illustration, not a claim about 1.FM's actual contract. Apple's developer terms state a 30% standard commission, 15% for qualifying programmes and 15% for qualifying subscriptions. Google's published fee table states 15% for automatically renewing subscriptions under the framework applying to the relevant markets. The realised take depends on store, country, programme eligibility, billing route and subscriber tenure.

The calculation exposes both the virtue and weakness of a $1.99 product. It is cheap enough to convert a loyal listener who would reject a CHF14 music subscription. It also produces little cash per account. If the fully loaded contribution after commissions, tax, rights and traffic were $1 a month, 50,000 retained subscribers would contribute $600,000 a year before fixed staff and equipment. If conversion is only a few thousand, subscription is a useful supplement rather than the engine.

Advertising has a different constraint. A free listener creates inventory, not revenue. Revenue appears only when an advertiser buys the impression and the ad can be delivered in the listener's geography. In an October 2025 community exchange, an account presenting itself as 1.FM's technical lead explained that a 30-second trigger could be filled by an outside advertising provider and that 1.FM material played when there was no paid fill. The source is informal and unaudited, but it describes the central economic problem: an unfilled break still interrupts the listener while producing no external advertising payment.

That makes geography and scale critical. A global audience sounds attractive, but an advertiser buying Swiss German listeners may not value a session in Thailand, Brazil or the United States. Different privacy rules, languages, demand levels and sales intermediaries affect fill. The old media material claimed city- and postal-code targeting; the current app listings do not disclose sell-through, price per thousand impressions or advertiser concentration.

Market growth does not guarantee 1.FM's share. IAB Europe's 2025 benchmark puts European digital audio advertising at €1.23 billion in 2025, up 13.9%, with “other audio,” including music streaming and internet radio, at €691 million. Money is moving into the format. It is also moving toward services able to provide scale, targeting, brand safety and measurement. A small broadcaster can benefit from an outside ad network, but that intermediary keeps part of the spend and controls demand access.

TuneIn shows what scale can be worth. In November 2025, Stingray announced an agreement to acquire TuneIn for up to $175 million. It said TuneIn expected $110 million of 2025 revenue and $30 million of adjusted earnings before interest, tax, depreciation and amortisation, served more than 75 million monthly active listeners, offered more than 100,000 stations and was distributed across more than 200 platforms and devices. TuneIn is an aggregator and advertising business far larger than 1.FM, not a direct valuation peer. Its figures show why device distribution and monetisation technology can be more valuable than any individual stream.

For 1.FM, revenue growth would create value only if contribution grows after traffic, rights and support. Buying audience with low-value international impressions can increase listening and worsen cash flow. Adding channels can raise total sessions while lowering average audience per channel. Raising ad frequency can improve short-term inventory and drive listeners away. A credible assessment needs listening hours, paid fill, net advertising revenue per thousand hours, subscriber conversion, churn and contribution by channel. None is public.

One thousand listeners already make a network

The physical unit behind the business is a continuous audio stream. In a technical discussion, the same 1.FM community account stated that the service was broadcasting at 192 kilobits per second and 44.1 kHz. Treating 192 kilobits as the delivered audio rate, one listener consumes about 86.4 megabytes in an hour. An average of 1,000 concurrent listeners over a 730-hour month would require roughly 63 terabytes of audio transfer. At 10,000 average concurrent listeners, the figure is about 631 terabytes. Protocol overhead, duplicated renditions, origin-to-edge transfer, apps, artwork and metadata sit on top.

These are arithmetic scenarios, not disclosed traffic. They explain why a radio operator might own network resources and still use a specialist delivery network. A direct server can handle a modest audience cheaply until a popular channel or geographic concentration saturates a port. A content-delivery supplier pools capacity and locations across customers, turning fixed infrastructure into a usage bill. The operator gives up some control but avoids building an international edge network.

Traffic also has a revenue asymmetry. A subscriber can be highly profitable if the monthly fee covers many listening hours. A free listener in a country with no ad fill can consume hundreds of hours and generate little cash. A short session may trigger a paid pre-roll and cost little bandwidth; a loyal all-day listener may be more expensive even though the relationship is more valuable. The company therefore needs controls that do not punish loyalty: sensible ad pacing, efficient audio formats, geographic delivery and a premium price that reflects heavy use.

The current $1.99 price appears aggressive against the quality promise. If “better quality” means a higher bitrate, premium users consume more delivery capacity while removing advertising. The subscription margin then depends on listening intensity. A light premium user is attractive; an all-day user pays the same and costs more. Tiering by quality, annual billing or a modest price increase could improve economics, but only if churn remains low.

The best economic reason to maintain proprietary network capability is not to carry every byte. It is to preserve bargaining and recovery options at the control points. A company-controlled address can host authentication, metadata, monitoring or an origin service while listener-heavy audio goes to multiple delivery providers. Stable domains and well-managed redirects can keep device integrations working while back-end suppliers change. A portable routing identity can support a second facility without requiring every partner to update an address.

That architecture creates value when it is exercised and tested. An unused address block announced through one provider is a fixed cost. A documented failover, independent backup origin and regularly tested alternate stream are operational assets. Public records show the ingredients of control, not the quality of implementation.

What the number resources prove, and what they do not

The RIPE evidence is substantive. 1.FM AG is listed as a Local Internet Registry member at Lindenstrasse 2 in Baar, with Switzerland as the service area. RIPE-derived records identify the company as the organisation behind AS60311, created in August 2013. Public routing summaries associate it with an IPv4 footprint of roughly 2,560 addresses and a large IPv6 allocation, including 2a04:5a40::/29. (RIPE member record, AS60311 record, route summary)

The footprint is more complicated than one Swiss block. IPLocate's current summary lists 185.33.20.0/22 under 1.FM and several US-associated prefixes registered to EGI Hosting, alongside IPv6 ranges. Aggregate and more-specific announcements overlap, so a prefix count should not be read as a simple address total. Registration names and routing origin also describe different rights: a block can be registered to one organisation and announced by another under contract.

These records prove that 1.FM has an established routing identity and resource-management responsibilities. They support the conclusion that network operations are relevant to the business. They do not prove that 1.FM owns the routers announcing every range, the buildings containing them, the fibre connecting them or the servers producing each stream. They also do not prove that all listed ranges are currently used for 1.FM listeners.

The difference is important because number resources are often mistaken for physical scale. IPv6 illustrates it. A /29 contains an enormous address space, but the size says almost nothing about traffic, hardware or audience. IPv6 allocation policy provides room to assign networks without scarcity at the individual-address level. It is an operational capability, not a balance-sheet valuation.

IPv4 can have scarcity value, but the article cannot infer ownership economics from a routing table. Some addresses may be provider-associated or contractually constrained. A clean transfer would depend on registry status, agreements and need. The useful value for a broadcaster is continuity and addressing flexibility, not a speculative price per address.

Maintaining the capability has visible overhead. RIPE's 2026 fee schedule sets the annual contribution at €1,800 per LIR account, with charges for certain independent resources and autonomous-system assignments. That bill is small beside staff or transit but represents the floor, not the total. Engineering time, route security, registry records, monitoring, abuse handling, equipment and upstream service cost much more.

The public routing picture is concentrated. IPinfo, IPLocate and CIDR Report each show one observed upstream adjacency: AS15576, NTS Workspace AG. The RIPE database policy record also contains an import and export statement for AS198326, Ortcloud GmbH, as well as NTS. That policy entity was last modified in 2018 and is not proof that both paths carry current production traffic. Observed routing and intended policy can diverge.

One visible upstream is not equivalent to one cable or one upstream supplier behind the entire service. NTS describes itself as operating a Swiss fibre backbone and carrier-neutral data centres in Bern and Zurich. Its PeeringDB entry lists connections at SwissIX, CIXP, AMS-IX and other exchanges, facilities in several Swiss cities, and a selective peering policy. Public summaries show several upstreams for NTS. 1.FM can therefore inherit meaningful route diversity beyond its first commercial hop.

But inherited diversity does not remove first-hop concentration. If 1.FM's contract, access circuit, router session or support relationship with NTS fails, NTS's many peers may be irrelevant. True independence would require a second active path with separate physical routing, tested failover and sufficient capacity. The public table cannot confirm those conditions.

No public route summary reviewed here shows downstream networks for AS60311 or direct peering. That is consistent with a content company operating its own network rather than an internet provider selling connectivity. It also means the autonomous system is not visibly collecting transit revenue or using a broad peering fabric to lower traffic cost directly.

The current service is already a hybrid

The user-facing architecture gives more context. The old www.1.fm address now redirects visitors to radio.1cloud.fm. Public domain observations identify the new site as being served through Cloudflare. A current stream hostname used in the company's own 2025 community guidance resolves through the audiocdn.com and cdnstream1.com domains, while another legacy icebox host remains under 1.fm. Cloudflare classifies cdnstream1.com as audio streaming. (current site observation, audio-delivery domain record, 1.FM service discussion)

This is a rational separation of duties. Cloudflare can absorb web demand, improve security and serve users from distributed locations. A specialist audio network can carry the high-volume stream. 1.FM can retain channel programming, accounts, metadata, domains and some origin infrastructure. Outsourcing the edge does not erase network ownership; it places owned control inside a larger supplier chain.

The chain has at least five technical failure domains. There is the playlist and automation layer that decides what plays. There is the origin or encoder that produces the stream. There is the content-delivery layer that replicates it. There is the website or app that tells users where to connect. There are aggregators and devices such as TuneIn, Alexa, Android Auto and Chromecast. A channel can be audibly healthy at the origin and unavailable through a particular device because metadata or a URL is stale.

That last point was visible in the 2025 service discussion. Users reported that streams worked in TuneIn's mobile app but failed through Amazon's Alexa integration. The technical account initially thought the stream incident was separate, then observed that TuneIn might depend on old website services. This is not a verified incident report, but it captures a common operational reality: distribution dependencies can turn a website change into a smart-speaker outage even when audio servers remain up.

The hybrid can be stronger than full ownership if responsibilities are clear. A specialist supplier may have more capacity, more routes and 24-hour staff than a small broadcaster can afford. It can be weaker if 1.FM lacks visibility, contractual response or a tested exit. The public evidence does not disclose service-level agreements, supplier concentration, egress terms, backup locations or the time needed to move streams.

The decision should be judged by recovery time, not vendor count. Ten suppliers can create ten support queues. Two well-integrated delivery paths with known failover may be better than a long list of logos. 1.FM's own addresses are valuable if they shorten a move between those paths. They are less useful when listeners and devices depend on hostnames embedded in outside systems that cannot be changed quickly.

The 2025 disruption was a test of capital allocation

The most informative public account of 1.FM's reliability came from an unofficial venue. In November 2025, a Reddit user named bri_1fm wrote that 1.FM's VMware cluster had technical difficulties and that the ISP had not provided a solution. The post said the website had been down for several hours, the legacy iOS app was degraded, the new beta site had been promoted earlier than planned and the company intended to migrate to a more stable hardware platform. The writer asked listeners to report missing streams.

In a separate introduction, the same account described himself as the company's technical lead, sole full-time developer and system administrator. Subsequent replies discussed backup playlists, a secondary playlist server, direct stream mounts and the risk of switching a backup service. These statements are not audited company disclosures. The identity and completeness of the account have not been independently verified. They deserve less weight than a filed report or official status page.

They still matter for three reasons. First, the account gave technically specific explanations in a dedicated 1.FM community and supplied working service addresses. Second, the public website did move to the 1cloud.fm address described in the discussion. Third, app-store records show major updates in late 2025 and early 2026 after a long gap in the iOS version history. The surrounding facts align with a real migration, even if every detail cannot be confirmed.

Economically, the incident shows what underinvestment can cost. A virtualisation cluster is meant to separate workloads from individual machines, but it does not eliminate shared storage, management, licensing, network or facility failures. If one cluster hosts the website, account services, metadata and playlist functions, redundancy inside that cluster may not protect the service from a cluster-level fault. A backup playlist reduces total silence but can create repetition and stale metadata, lowering product quality even when an audio connection remains open.

The incident also shows the limits of supplier accountability. Saying the ISP had not produced a solution identifies a dependency, not a transfer of business risk. The listener's relationship is with 1.FM. If the upstream, hosting company or virtualisation supplier is slow, 1.FM still loses the session. Contracts can provide credits, but a service credit rarely replaces lost loyalty or a cancelled $1.99 subscription.

The company's response had real strengths. It used a replacement web service, exposed direct stream mounts, communicated with listeners and worked through app updates. That is practical recovery. The new site behind Cloudflare and the use of a specialist audio-delivery domain suggest that the company did not merely restore the old arrangement.

The weaknesses are equally clear. A new site was apparently pushed into the primary role before a planned launch. Some channels, metadata and device paths were missing or stale. A proposed switch to a backup playlist server was described as capable of taking everything down if it failed. Those are signs that failover had not been fully rehearsed or isolated.

Reliability spending is often hard to justify before a failure. A second cluster, separate storage, another upstream and continuous testing all look like idle expense when the service is healthy. For a free radio network with uncertain ad fill, the temptation to defer them is strong. Yet the outage put the cost on the revenue-generating surface: website access, app usability, channel discovery and aggregators. The avoided loss from redundancy can be larger than the hardware bill even when no single incident is catastrophic.

This is where owning network capability should change behaviour. A company that pays for an LIR, runs an autonomous system and maintains addresses has already accepted some fixed engineering cost. The incremental strategy should be to make that capability useful during failure: two active paths, independent name service, outside monitoring, a tested origin switch and device integrations that follow stable URLs. Otherwise the resources become technical credentials rather than insurance.

The cost base extends far beyond transit

Network delivery is only one large line in an internet-radio cost stack. Music rights can scale with revenue, cost, channels and geography. SUISA's current guidance says professional web radios and stations capable of more than 6,000 simultaneous connections are subject to the ordinary broadcaster tariff rather than the simple hobby-radio amount. For a non-commercial web radio with no more than 6,000 simultaneous connections, the general fee is CHF120 plus VAT per programme per month. 1.FM is visibly commercial and therefore should not be assumed to qualify for that simplified treatment.

The 2026-2028 Common Tariff S makes the broader burden clearer. It bases compensation on relevant broadcaster revenue or cost, varies authors' rights with the proportion of protected music and adds neighbouring rights. The tariff and SUISA guidance also impose music-reporting duties. A service built almost entirely from commercial recordings therefore carries both cash and administrative cost.

Territory adds another layer. SUISA's internet-radio licensing note distinguishes stations confined to Switzerland from stations receivable abroad and points to additional rights handling through Audion. The older English tariff text states that SUISA and SWISSPERFORM do not grant all relevant rights outside Switzerland. 1.FM describes a global audience, so a Swiss licence should not be read as proof of complete worldwide clearance. Actual rights depend on content, interactivity, territory and agreements that are not public.

Labour is another fixed cost. Someone must schedule music, ingest files, manage loudness, maintain metadata, respond to rights reports, build apps, handle accounts, support listeners, sell advertising, monitor systems and resolve abuse. Automation lowers the number of people required but increases key-person and software risk. The unofficial claim of one full-time developer, if accurate, would make knowledge concentration a major issue. The register's single board member creates a separate governance concentration.

Equipment refresh is easy to understate in a hybrid model. Even when audio delivery is outsourced, origins and control services need compute, storage and backup. Drives fail, firmware expires, security updates stop and virtualisation platforms change commercial terms. Spare capacity must be bought before it produces revenue. The 2025 discussion of migration to a more stable hardware platform implies a real refresh decision, but no budget or completion report is public.

Third-party software and app compatibility create recurring cost without appearing as hardware. Apple's version history shows a major gap between an iOS release in January 2020 and updates in November 2025 and January 2026. During that period operating systems, privacy rules, device interfaces and store requirements changed. Keeping an app merely available is different from maintaining it as a dependable route to the service.

Distribution intermediaries take both money and control. App stores can retain a share of subscriptions. An advertising provider retains part of media spend. A delivery network charges for traffic or capacity. An aggregator can alter URLs, ranking, ad insertion or device support. Cloudflare can reduce attack and bandwidth risk while becoming another contract that must be paid and configured. Each relationship can be economically rational; the combined take determines whether audience growth produces profit.

Regulatory overhead belongs in the cost base too. Swiss radio law, music reporting, privacy, consumer billing and international data rules all require attention. None is likely to dominate the economics alone. Together they make a one-person operating model fragile.

Customer concentration is hidden in two markets

1.FM serves two customer markets. The listener receives music and may pay a subscription. The advertiser buys access to listeners. The company needs both markets to remain balanced: too much advertising damages listening, while too little paid demand leaves free traffic unfunded.

Listener concentration can occur by channel, geography, platform and device. A service with 65 channels may still depend on five popular stations. A global brand may depend on US advertising. A direct website may account for little listening if TuneIn, smart speakers or car systems dominate. The Google app's 100,000-plus downloads do not reveal active users, listening hours or revenue. An installation ten years ago and an all-day subscriber both count as one download.

Platform concentration can be more dangerous than customer concentration. If a large share of sessions starts through TuneIn or Alexa, a feed-mapping problem can remove access without any failure in 1.FM's own network. The 2025 user reports show that this failure mode is plausible. Android Auto and Chromecast broaden reach but add compatibility work. Apple's small number of visible Swiss reviews does not measure the global iOS base.

Advertising concentration is equally opaque. A single programmatic partner can simplify sales across countries and supply targeting that a small publisher cannot build. It can also determine fill, pricing, reporting and creative quality. The community account's statement that 1.FM did not choose certain ads suggests some control sat with the provider. If the largest demand partner changes terms, 1.FM may have no direct advertiser relationship to protect revenue.

The old 400,000-listener claim cannot solve this gap. “Unique listeners” can be measured by device, account, cookie or IP over different periods. It says nothing about hours, geography, valid traffic or monetisation. Triton Digital's current methodology emphasises measured sessions, filtering and known technical issues, illustrating why an audience claim needs a date and method. (Triton methodology)

A resilient company would disclose or at least manage four concentration measures: share of listening from the top five channels, top five countries, top three platforms and top advertising partner. It would also track premium churn by acquisition route. Without those facts, an outside judgment should assume that concentration may be material and unpriced.

Competition begins with the stop button

The closest substitute is not another company with an autonomous system. It is any audio service already installed on the listener's phone, car or speaker. Switching costs are almost zero for a free listener. Spotify, Apple Music, YouTube, TuneIn, DI.FM, terrestrial radio streams, podcasts and personal music libraries all compete for the same hour.

Full music subscriptions win on control. A user can search a vast catalogue, create playlists, download music and move between devices. They cost much more than 1.FM Premium in Switzerland, but a household already paying for one sees the marginal cost of leaving 1.FM as zero. Algorithmic playlists can imitate a genre channel while allowing skips and saved tracks.

Internet-radio aggregators win on breadth. TuneIn offers access to more than 100,000 stations and has deep device distribution. A listener can find 1.FM inside an aggregator and then move to another station without visiting 1.FM's app. That makes TuneIn both a channel partner and a competitor. Distribution raises audience while weakening the direct customer relationship.

Genre specialists compete on curation. DI.FM, for example, sells commercial-free access to a network of electronic-music channels and more than 200 additional channels across related services. 1.FM can counter with its mix of moods and eras, but it must make each channel distinctive. Generic playlists have little defence against a larger service.

Free broadcast and community radio compete on human presence and local information. A listener in the 1.FM community explicitly asked for presenters and real-world updates. Adding live hosts could deepen loyalty and sponsorship value, but it raises labour cost and turns a low-cost automated channel into a scheduled media operation. The company should not add that expense across 65 stations without evidence that it raises listening or ad yield.

For advertisers, substitutes are broader still. Social video, retail media, search, podcasts and large music platforms offer larger audiences and established measurement. IAB Europe's latest figures show video and retail media growing faster than the overall digital market. 1.FM's defence is not scale. It is a specific listening context, long sessions, genre affinity and geographic targeting. Those advantages need independent measurement to command a price.

For the company, the infrastructure substitutes are also real. It could shut its autonomous system and put everything behind cloud and audio-delivery suppliers. That would remove registry and routing work but increase exit dependence. It could build a second physical network path and retain more control, raising fixed cost. Or it could keep the current resources as a portability layer while outsourcing listener traffic to two independent delivery providers. The last option appears most proportionate for a small global broadcaster.

The alternative that should be rejected is vague “ownership” without resource allocation. A press claim of independence does not repair a cluster, buy a second circuit or staff an incident. Strategy must be visible in redundant capacity, tested recovery and supplier terms.

Regulation is both licence to operate and margin pressure

Swiss broadcasting rules apply beyond terrestrial frequencies. The Federal Office of Communications says the Radio and Television Act is technology-neutral and that internet programmes can be subject to notification unless they are of minor journalistic significance, including a technical threshold of fewer than 1,000 simultaneous devices. It also says failure to notify correctly can lead to an administrative sanction of up to CHF10,000. Whether every 1.FM channel falls within a particular exception depends on audience and editorial characteristics that are not public.

Advertising rules also affect product design. OFCOM's guidance says Swiss radio advertising and sponsorship provisions can cover streamed programmes and require advertising to be distinguishable from programme material. That is not merely legal formality. A clean break helps the listener understand why the music stopped, while badly timed insertion can damage the experience and create complaints.

Music licensing is the largest visible sector-specific burden. Rights payments protect the creators and performers whose recordings make the service possible. They also create a floor under costs that a pirate or unlicensed service avoids. Compliance can therefore be a modest barrier to entry, but it does not give 1.FM exclusive content or pricing power. Every legitimate competitor faces a rights stack appropriate to its service.

Data protection cuts across the advertising model. Google Play says the app may collect personal information, app performance data and device or other identifiers, while saying data is encrypted in transit and users can request deletion. These are developer-supplied disclosures, not an independent privacy audit. A service offering accounts and location-targeted ads needs a clear map of which company controls account, device, listening and advertising data.

The Swiss data-protection authority's tracking guidance warns that clicks, viewing and other behaviour can be combined into profiles and recommends minimisation, clear information and meaningful user choice. Its 2026 factsheet explicitly includes advertising identifiers and app tracking among technologies that can support targeted advertising.

European reach can add the GDPR. The European Commission says the rules apply to a company outside the EU when it offers paid or free services to people in the EU or monitors their behaviour there. Switzerland is outside the EU, but 1.FM describes listeners in Europe and distributes through global apps. Compliance scope depends on targeting and processing, not simply the Baar address.

These obligations do not invalidate targeted advertising. They change its economics. Consent choices can reduce addressable inventory. Data-processing agreements and user requests consume staff time. A breach can damage trust. Contextual advertising by channel and country may be less precise than behavioural targeting but easier to explain and more aligned with a genre-radio service.

Unofficial signals deserve questions, not a verdict

The public app feedback is mixed. Google Play displays thousands of reviews and more than 100,000 downloads. Reviews posted after the 2025 update include complaints about streams hanging, lost favourites and background behaviour, while other users praise the app's simplicity and reliability. Apple's smaller review set includes a 2024 complaint that several channels had stopped working and a late-2025 comment that the new version was better.

None of that is a representative satisfaction survey. Store ratings vary by country and device, unhappy users may be more likely to post, and an old complaint may describe a version no longer distributed. The comments are useful because they identify failure modes that align with the migration discussion: missing channels, repeated interruptions, app state and stream continuity.

The dedicated 1.FM Reddit community is more detailed and less independent. A person presenting himself as the technical lead answers questions, explains repairs and accepts fault. That candour is a positive operating signal. It is also evidence of reliance on an informal channel without the controls of an official status page. The account's self-description as sole full-time developer raises key-person risk if accurate, but it remains unverified.

The most encouraging signal is persistence. The company has existed since 2011, its network identity dates to 2013, the Android app was updated in October 2025, the iOS app received updates after the outage and the channels continue to appear in third-party radio directories. A service that survives a difficult migration has some operational capability and listener loyalty.

The most concerning signal is that the migration exposed multiple coupled dependencies at once. Website, iOS access, metadata, playlist state and TuneIn or Alexa paths were all discussed around the same period. That can indicate accumulated technical debt. The only way to distinguish a completed repair from a temporary workaround is current uptime, incident and architecture evidence, which is not public.

Where the next franc should go

1.FM's scarce capital should not be spent making the company look like a large carrier. It should be spent at the points where a failure loses audience and where a second option can be exercised quickly.

The first priority is an independent recovery path. The company needs a second origin and playlist service outside the primary virtualisation and storage failure domain, with automatic or rehearsed switching. The backup must preserve enough metadata and channel mapping that apps and aggregators do not become separate repair projects.

The second is connectivity diversity. A second upstream is valuable only if it uses separate access, power and routing equipment and is tested under load. NTS appears to provide strong onward connectivity, so replacing it is not the objective. The objective is to avoid one commercial or physical first hop deciding the availability of company-controlled services.

The third is stable distribution. Device partners and third-party apps should point to durable company-controlled names that can be moved between delivery providers. 1.FM should retain more than one audio-delivery option for its largest channels, even if the long tail remains on one provider. The most valuable failover covers the stations producing the most listening and revenue.

The fourth is people. Documentation, access control, on-call coverage and a second engineer are forms of redundancy. If one person really holds most development and system knowledge, an additional circuit cannot repair the organisational failure domain. A small company can use a retainer with an outside operations firm rather than immediately building a large staff, but the second party must practise recovery.

The fifth is measurement. Advertising inventory without audited listening and fill data will sell at a discount. Subscription investment without conversion and churn data is guesswork. A monthly operating view should connect channel listening hours to delivery cost, rights cost, ad revenue, premium conversion and incidents. That would identify which channels deserve refresh and which are being subsidised without strategic benefit.

The sixth is pricing. The $1.99 premium tier is an effective entry offer but may be too low for heavy global listening if store, rights and delivery costs absorb most of it. An annual plan could improve retention and reduce payment churn. A higher tier might offer better quality or more control. Any increase should follow evidence that premium users value continuity and curation, not simply an attempt to repair margins.

Acquisitions or channel expansion should come last. Adding another radio brand produces little value if the existing control plane remains fragile. Buying audience before measuring contribution can turn growth into a larger bandwidth and licensing bill. The company should first make reliability observable and monetisation repeatable.

What would change the judgment

Several disclosures would materially strengthen the case that 1.FM's network ownership creates economic value.

The first is a current revenue bridge separating advertising, subscriptions, branded services and other income. It should include paid accounts, churn, average realised subscription price, ad fill, net advertising revenue per thousand listening hours and the share retained after intermediaries. Growth in listener count without this bridge is not enough.

The second is channel and platform concentration. Listening hours for the top ten stations, countries and distribution partners would show whether the 65-channel catalogue is diversified or mostly decorative. No single aggregator or advertising provider should be able to remove a majority of revenue without a practical alternative.

The third is reliability evidence. Monthly availability by listening path, severity-one incidents, recovery time, failed stream starts and service credits would reveal whether the 2025 migration improved outcomes. A diagram is less valuable than a record of tested failover between independent origins, delivery providers and upstream paths.

The fourth is a capital and supplier plan. Hardware age, virtualisation platform, storage redundancy, annual refresh spending, reserved delivery capacity and contract expiry dates would show whether the company is investing or merely extending old equipment. A second active upstream with physical-path evidence would improve the network judgment.

The fifth is rights and regulatory scope. Confirmation of current Swiss notification status, channel-level music reporting, international licensing arrangements and data-controller responsibilities would reduce the risk that global reach creates an unfunded obligation.

The sixth is organisational resilience. Verified staffing, named incident cover, documented recovery ownership and evidence that more than one person can operate the service would reduce key-person risk. The single-member board is legal, but stronger governance would help when technical and commercial decisions concentrate in a small group.

The negative triggers are the mirror image. More prolonged stream interruptions after the migration, loss of major app or aggregator distribution, continued dependence on one untested playlist path, rising ad load without paid fill, premium churn after a price change, or a decline in active channels would weaken the case. So would evidence that number resources remain announced but no longer support meaningful company-controlled services.

Judgment: useful control, unfinished reliability

1.FM AG owns enough network identity to be more than a playlist brand renting a single server. Its RIPE membership, autonomous system and address resources create real operational options. Its long life, broad channel catalogue, active apps and successful move to a new web surface show persistence. The low premium price gives loyal listeners a simple way to fund the service.

The evidence does not show a reliable network owner in the stronger economic sense. Public routing observations show one upstream for AS60311. Current delivery depends on outside web, audio, app and aggregation platforms. The clearest service episode exposed coupled infrastructure, supplier and staffing risks. Financial disclosure is too thin to know whether advertising and subscriptions cover rights, delivery, labour and refresh while leaving a return.

The company can make ownership pay if it uses the resources to preserve choice: stable names, portable services, two tested delivery paths, independent recovery and measurable customer outcomes. Listeners benefit from uninterrupted music, advertisers benefit from completed and credible impressions, and 1.FM earns the spread after paying every supplier in between. If those options are not funded, the company carries the complexity of ownership while still accepting the failure risk of resale.

That is the price of reliability for a small digital broadcaster. It is not the annual RIPE invoice or the cost of one more server. It is spare capacity, disciplined switching, rights administration, supplier leverage and enough people to recover when the first plan fails. 1.FM has assembled several of the necessary assets. Whether they form a durable business depends on evidence the company has not yet made public.