Spontaneous curiosity, October 20, 2026. The trigger was an archival investigation on Habr into “1C killers.” The subject turned out to run deeper than “bad platforms failed to take off.”
There is a genre more alive than any press release: obituaries for would-be killers. In October 2026, Habr ran an investigation into platforms that, at various times, had been declared replacements for 1C — Ananas, 2C, TURBO, OneBase, Jmix, lsFusion, Odoo. The outcome was predictable: almost all are either dead or huddled in a narrow niche. The author methodically went through their websites and saw what usually only investigators see: dates of the last updates, 2007–2012; comments saying “it works, but nobody needs it”; forums with three people in them.
But among the comments beneath that article was one remark worth an entire study. Its meaning: none of the contenders ever built a coherent army of franchisees ready to implement the system in any village.
That’s not about code. It’s about the logistics of people.
ERP is a strange product. The license price is an admission ticket, not a purchase. Accounting is not a photograph of a business but its legal shadow: it exists only in the forms the tax authorities accept, and only in the habits of a particular chief accountant. Buying a box and “installing” it is like buying a scalpel and deciding the operation has been performed.
This leads to a non-obvious conclusion: an ERP vendor has no way to sell its product directly unless it is prepared to maintain thousands of service branches in cities where it has neither an office nor employees. And this is where the fork in the road begins — one that 1C took differently from everyone else.
In 2000, Computerworld Russia reconstructed the network’s origins and included a direct quote from Boris Nuraliev:
At the end of 1993, it became clear that a promising sector of the market would be individual customization and support for users of accounting software. The simple solution for 1C would have been to create service branches and offices. But we believed it was right to preserve and develop our partners’ businesses, so as a possible way out of the anticipated situation, we proposed putting the provision of increasingly expensive economic-software implementation services onto an industrial footing.
Read that again, because the entire trick is hidden here. The vendor refused vertical integration into services — what today would be called an “in-house competence center.” It handed the last mile over to other people and, more importantly, handed them the margin.
The model built by 1996 was proclaimed “the first franchising network in the Russian computer business.” Its growth is impressive even by the standards of a modern venture-capital slide: in January 1996, around 80 franchise agreements; in December, 800.
The price of entry was laughable. Under the terms of the late 1990s: submit an application, pay a fee (in 1999 — $100 per quarter), train and certify an implementation specialist within the quarter — and do not sell competitors’ accounting products. In return: the right to buy software at a 55% discount, operate under the brand, receive support, and get 20% of local advertising costs reimbursed. Buy more than 50 products in a quarter, and the fee was waived.
Notice what is being sold here. Not software — the right to be part of a system that makes your software sellable.
A 55% discount — a figure that has survived thirty years unchanged. In October 2024, RBC quoted a 1C representative on the price increase: “our implementation partners will receive the larger part of this increase (the 1C:Franchisee discount is 55% of the retail price). The salaries of specialists and other expenses of our partners are also rising, and we must care about the profitability of their work.”
It is a passage remarkable for its frankness. The vendor publicly explains a price increase not by its own needs, but by the channel’s profitability. Compare that with the usual logic: a manufacturer raises prices to earn more. Here — so the partner does not die somewhere between rising developer salaries and a fixed discount.
And this is not a one-off act of generosity. Look at the network’s structure in 2000: out of 2300 permanent partners, around 1500 worked under the franchise model, with the average firm employing more than 23 people. Even then, this was not a “dealer network,” but a distributed service industry with an economy of its own.
What is happening to that economy now is best shown by TAdviser’s market figures. Estimates of the Russian market for 1C solutions vary wildly — from 80 billion rubles to 150–250 billion. But one estimate contains a surgically precise detail: of 150 billion rubles, “30–40 billion is partner and franchisee revenue, while the remainder comes from the vendor and captive companies.”
In other words, the partner layer earns an amount comparable to the vendor — and by some estimates, more. The army is not impoverished. The army is fed.
Now, about “any village” — that very wording from the Habr comment. It is not a metaphor, but a description of a business model that 1C implemented literally.
The official franchise terms contain a clause few people read: for small towns and remote territories, the requirements for the number of certified specialists are reduced; for sole proprietors — one-person operations — there is no initial fee at all. The minimum requirements for a location: one employee and 10 square meters.
Ten square meters. One person. This is not a corporate standard; it is a standard of viability. The vendor deliberately admitted micro-businesses into its network that would never have passed SAP’s selection process — because for a town of 12 thousand people, the alternative is not “the best integrator,” but “nobody at all.”
There is an opposite pole as well. The network’s official pages list around two dozen statuses: competence centers for ERP, CORP, CRM, document workflow, personnel management, budget accounting, healthcare, education; real automation centers; 1C:Consulting; 1C:Accounting Services. The average firm has 30 employees, but the spread runs “from solo specialists to firms with several thousand employees.”
This is not a network. It is a fractal structure, where at every scale there is a player that fits.
The closest conceptual analogue is Odoo. The Belgian company built an open ERP with a partner network, and according to partner reports as of October 2025, there are around 9900 partners worldwide, including 4440 in Europe. That is more than 1C’s figure: “more than 8000 organizations” holding 1C:Franchisee status across Russia and the CIS.
But quantity is the least interesting part of this comparison. The construction is what matters.
Odoo gives partners a commission of up to 20%, a coupon, visibility on its website, marketing support, a sandbox. And the key point: “Odoo will not take % from partner services” — the vendor takes no percentage of the partner’s service revenue. Services are declared “your main source of revenue”; the partner sets its own prices. This is not a franchise with royalties and price controls, but certification plus a free market.
With 1C, it is the other way around: there is a regular payment (a royalty of 3000 rubles per quarter), there are network rules, and among them are two clauses that would cause a scandal in the libertarian world of open-source software:
The second clause is channel exclusivity. A partner cannot sell “1C and Odoo on the side.” They choose a side.
And now the question people usually do not ask: if a franchise restricts freedom, why do tens of thousands join it?
The answer, it seems, is that 1C sells not freedom, but predictable demand. A partner in Kaluga does not have to guess whether there will be an ERP market tomorrow. They know that accountants file reports, that forms change every year, that updates must be installed, that people pay for this. The restriction of freedom is offset by a guarantee of work.
An army is not free. It requires management, and 1C managed it with mixed success.
In 1999, Kompaniya recorded a quality crisis with almost indecent bluntness. Customer complaints, and a quote from the vendor itself: “At present, roughly half of our franchisees do not have a sufficient number of qualified implementation specialists.” Add bad faith — inflated timelines, unfinished work followed by extra charges to fix it.
The response was bureaucratic, and that is a compliment: the “principle of industrial quality,” standard sets of implementation-technology recommendations, “service books” with marks for each stage of an order, a control system. A partner survey found support for the idea from only 60% — meaning 40% of the army saw oversight as an encroachment.
A similar story unfolded around dumping. In a city of more than a million people, there were on average 10 franchisees; some Moscow partners cut prices by as much as 50% below the list price. In June 1999, 1C prohibited discounts greater than 30–35% from the recommended price — as a separate clause in the contract.
And the third, most unpleasant bill came due in the 2020s. The labor market. According to figures cited by Delovoy Peterburg from hh.ru: 37.7 thousand 1C vacancies versus 7.3 thousand job seekers. By comparison, on the same date: Python — 11.9 thousand vacancies (including Data Science), JavaScript — 6.8 thousand, Java — 5.3 thousand, PHP — 3.2 thousand.
There are more 1C vacancies than Python, JS, Java, and PHP vacancies combined. And that is not a sign of technological triumph — it is a sign that demand has outrun the education pipeline.
This produced something nobody could have predicted in 1999: franchisees do not compete with the vendor, or with Odoo. They compete with one another for people. The director of Rekssoft’s Business Applications department put it without illusions: “The number of 1C specialists is quite large, but they are all already employed.”
And SoftBalance, in a TAdviser review, added a detail that makes the picture decidedly unceremonial: since February 2025, the market has seen “a certain number of underutilized teams periodically emerge among integrators.” At the same time — a shortage of people and idle teams. This is not a contradiction; it is a structural imbalance: the army is large, but its specialization and geography do not match where the money is now.
The tempting answer is: because it is difficult. But “difficult” is not an explanation; it is an excuse. Copying a network in 2026 is cheaper than it was in 1996: there is CRM, there are marketplaces, remote work, online courses.
The real reason, it seems, lies in the nature of the asset. A network of franchisees is not the vendor’s property. It is loyalty that must be reaffirmed every year. A partner stays with you as long as your brand brings them a stream of orders, and leaves in the quarter that stream dries up. No contract can hold a firm that has no work.
Hence what I would call the platform vendor’s paradox: to have an army, you have to give it most of the money and most of the control. 1C gave up both: 55% license discounts, free pricing for services, limited oversight. A company that wants to manage its channel imperatively will get a channel that knows only how to sell boxes.
This is precisely where, in my view, the “1C killers” stumbled. They arrived with better architecture, a modern stack, and a beautiful landing page — and an offer to partners: sell our product. And the partner heard: “learn everything again, lose your clients, lose your team, start from scratch” — and found no economic sense in it. No technology wins if there is nobody to carry it to the last accounting office in the district town.
There is a temptation to read this story as one about monopoly. 1C won, supposedly, because it was first to claim the clearing and build a fence around it. There is some truth in that reading, but it is dull and, it seems, incomplete.
A more interesting version is this: 1C did not build a monopoly on software, but a monopoly on the last mile. Software can be copied, a platform rewritten, a license made free. But you cannot reproduce in a single year 8000 firms that know what proper accounting looks like in a particular region, a particular industry, and a particular accounting office.
The economics here works almost like nature. An ecosystem of small organisms is more resilient than one large organism: it survives the loss of any node, fills every empty niche, and does not require central management for every action. 1C does not build branches — it builds an environment in which branches grow on their own.
And this model has a hard condition for survival: the vendor must remain affordable and predictable. Once the license becomes expensive enough that leaving makes more sense for the partner, the army begins to melt away — not at once, but at the edges, as an ecosystem should. Price increases of 10–18% in 2025 are not critical yet. The question is at what number the erosion will begin — the kind management will not notice immediately.
Because in this model, the most dangerous threat is not a competitor with better code. It is its own greed.