Buff Technologies 2025: The Second Half Improved, But The Cash Test Is Just Starting

Company Introduction

At first glance, Buff looks like another gaming app with points and rewards. That reading misses the economics. In practice, this is a monetization and advertising company built on top of a gamer audience: the PC platform is still the real revenue engine, while mobile remains in the proof stage. In 2025 revenue rose to $8.497 million, but almost all of that still came from the older platform, not from the new layer that management wants investors to underwrite for 2026 and beyond.

What is working now is clear enough: the campaign channel. Revenue rose 20.8%, average revenue per daily active user climbed to 17.7 cents from 12.7 cents a year earlier, and in the second half management was already talking about a 20 to 22 cent range per active user per day. What is still not clean is cash. The company ended the year with only $351 thousand of cash and cash equivalents, working capital of just $37 thousand, and a mobile business that still did not contribute materially to revenue. So this is not yet a clean growth-company setup. It is a company that improved the economics of the existing PC business while still needing to fund the bridge until mobile turns from product into business.

The practical screen matters too. This is a very small and relatively illiquid stock. On the latest trading day, turnover was only about NIS 170 thousand. This is also not a persistent short story: after a temporary short-float spike to 1.68% on March 6, 2026, the figure returned to 0.00% by March 13, 2026. In other words, the real debate is not technical positioning. It is whether Buff has genuinely turned an economic corner, or simply bought itself more time.

Five points worth holding in mind right from the start:

• The second half looked much better, but it was not clean. Loss from ordinary operations fell sharply, yet profit from continuing operations still leaned on non-cash finance income.
• Part of the improvement ran through the balance sheet. In 2025 the company capitalized $579 thousand of mobile development costs into an intangible asset for the first time.
• The business became more valuable even with less volume. Average daily active users fell from 146 thousand in Q1 to 110 thousand in Q4, while revenue per active user improved materially.
• Liquidity was still tight. Year end rested on a bank line, a term loan, the March 2025 private raise, and the March 2026 public raise.
• Overwolf is far more than a vendor. It sits in the ad infrastructure, the PC software layer, and a material part of the direct-campaign engine.

Events And Triggers

The first trigger: in February 2025 Buff updated its annual arrangements with Overwolf around the campaign channel. Overwolf committed to a minimum 2025 campaign target of $4.1 million and, alongside that, committed to pay Buff about $171 thousand per month even if the full target was not realized. That helped support the 2025 run-rate, but it also made clear how much of the growth in the company’s main channel still depended on one partner.

The second trigger: in August 2025 Buff replaced and expanded its bank financing structure. The company received a credit line of up to $1.5 million against invoices, approved work orders and MRR, plus a $1 million three-year term loan. That matters because it shows the company was no longer living only on equity capital, but it also was not yet funding itself through operations. The bank was financing a bridge, not a free-growth story.

The third trigger: after the balance-sheet date Buff went back to the market. In March 2026 it completed a public equity raise of about $2.3 million gross. In the same month it also disclosed that it had received a bank proposal for up to another $2 million of long-term financing, with draw conditions tied both to additional equity and to a business target related to mobile revenue. That is exactly the difference between a growth story and a proof story: even the next layer of debt is waiting for mobile to start working.

The fourth trigger: in March 2026 the company launched another advanced version of the mobile platform. Management says 2026 should require meaningfully less investment than 2025 and that the focus has now shifted toward KPI improvement and a positive contribution to cash flow. That matters, but the other side matters just as much: in 2025 mobile revenue was still immaterial. So 2026 begins with a more mature product, not with proven commercialization.

The fifth trigger: in early January 2026 Buff signed a non-binding memorandum of understanding to explore a joint AI venture based on its user-data assets. That is interesting optionality, but it is still optionality. The cash story of 2026 will not be decided by that idea. It will be decided by campaigns, mobile, and expense discipline.

Efficiency, Profitability And Competition

The business improved through quality, not through user growth

The most interesting part of Buff’s 2025 is not user expansion. It is better monetization of the users it already has. Average daily active users fell from 146 thousand in Q1 to 110 thousand in Q4, yet ARPDAU rose 39% to 17.7 cents. In the second half, management was already pointing to a 20 to 22 cent range per user per day. That is not a contradiction. The company says explicitly that it has been moving toward a more value-driven model, in which each active user generates more money.

The path there ran through a sharp mix shift. Campaign revenue rose 82% versus 2024, while ongoing advertising revenue fell. Management explains that through four forces acting together: ongoing advertising requires heavier continuous investment, its revenue is recognized over a longer user life cycle, direct campaigns generate much higher yield per ad slot, and pricing in ordinary advertising has also been eroding. That matters because Buff did not merely “grow in campaigns.” It made a conscious decision to give up part of the older mix in order to improve the economics per user.

The company also supports that with more selective marketing. About 80% of marketing spend was directed toward Tier 1 markets such as North America, Australia and New Zealand, with another 20% going to Western Europe and other higher-value markets. So even at the user-acquisition layer, Buff preferred quality over volume.

The second half improved sharply, but a large part of it was not pure operating performance

The headline management pushed is understandable: in the second half revenue rose to $4.872 million, loss from ordinary operations narrowed to just $610 thousand, and comprehensive profit turned positive at $267 thousand. On the surface that almost looks like breakeven. That reading is too generous.

First, the core business still lost money in the second half. Profit from continuing operations did turn positive at $330 thousand, but only after net finance income of $940 thousand. The finance note breaks that down: $1.213 million came from the revaluation of investor warrants, and another $245 thousand came from the discounting of the founders’ compensation liability. Those are real accounting items. They are not recurring operating engines.

The picture gets even clearer when the separate option-valuation attachment is brought in. At the end of December 2025 the share price used for measurement was NIS 10.12, while the warrants’ exercise price was NIS 19. The fair value of each warrant fell to about NIS 0.003, and by March 3, 2026 the warrants had already expired. Put simply, a meaningful part of the finance-line improvement came from investor warrants becoming nearly worthless. That is valid accounting. It is not evidence that the underlying business had already become self-funding.

The same caution is needed in R&D. In 2025, and effectively starting in the second half, the company began recognizing part of mobile development costs as an intangible asset. The amount capitalized in 2025 was $579 thousand. That helps explain why R&D expense fell from $1.328 million in the first half to $738 thousand in the second half. Again, that does not mean the cash burden disappeared. It means part of the cost of building mobile moved from the income statement to the balance sheet and investing cash flow.

That chart tells the story more honestly than the headline. Buff added $1.462 million of revenue, but cost of revenue jumped by $1.414 million. Almost all of the incremental revenue was absorbed along the way. That is not automatically bad, because campaign costs are recognized in cost of revenue even when part of them economically resembles selling expense. But it does mean that the mix improvement in 2025 had not yet translated into a clean full operating improvement.

Competition and concentration still matter

From a competition perspective, Buff is not operating from a position of market strength yet. The company itself says its market share is not material. In the broader field it competes with in-game loyalty programs and with players such as Sony Rewards, GameStop Power Up Rewards and MyNintendo. On mobile it explicitly mentions names such as FreeCash, Gamelight and Mistplay. So mobile opens a larger opportunity set, but it also pushes Buff into a segment where much bigger and more proven players already exist.

On top of that, commercial concentration around Overwolf is larger than a superficial reading suggests:

The company did work during 2025 to add more agencies and reduce this dependence, and that matters. But at year end 2025 and into early 2026, anyone buying the Buff story is still, to a degree, buying the stability of the Overwolf relationship as well.

There is one more limitation that matters. Buff is not a backlog business. The company says explicitly that its activity is not characterized by an order backlog, except for short-term backlog that can arise around direct campaigns signed close to the report date. Mobile developer agreements are also typically valid only through the end of the calendar year and can be cancelled on short notice. So even when the growth looks strong, its visibility is still relatively short.

Cash Flow, Debt And Capital Structure

The cash frame needs to be explicit here. In Buff’s case, the right frame is all-in cash flexibility. The question is not how much the business might generate in a theoretical steady state before growth investment. The question is how much money was actually left after all real cash uses.

On that basis, the picture is sharp. In 2025 operating cash flow was negative $3.461 million. Investing cash flow was negative $836 thousand, including $579 thousand of capitalized mobile development and $250 thousand placed into deposits. Financing cash flow was positive $3.659 million, mainly due to the bank line, the March 2025 equity raise, and the non-tradable warrants issued to investors. Even after all that, cash fell by $615 thousand and ended the year at just $351 thousand.

That is the heart of the story. 2025 was not the year in which the operating model funded itself. It was the year in which the operating picture improved while financing still carried part of the weight. Even if short-term deposits of $253 thousand and restricted deposits of $148 thousand are added, the liquidity layer remains thin relative to the burn the company experienced over the year.

There is another nuance worth surfacing: the equity deficit. At the end of 2025 equity was negative $771 thousand. At the same time, the board explains that this amount includes a $1.213 million founders’ compensation liability, and that excluding this item equity would actually have been positive by $442 thousand. That does not make the issue disappear, but it does show that the balance sheet is somewhat less distressed than the headline deficit alone might imply. Management is also evaluating ways to convert that liability in order to reduce the deficit, though no binding resolution had been taken by the report date.

On the debt side, Buff is not currently in a classic covenant crunch, but it is living inside a tightly managed financing relationship. The credit line carries Term SOFR plus 4.75%, the term loan carries Term SOFR plus 6.5%, and the bank also required a variable security cushion, a single-customer exposure limit within the credit framework, customary acceleration clauses, and a range of collateral undertakings. One detail matters especially: the single-customer exposure limit does not apply to Overwolf. So precisely where the company has commercial dependence, the financing structure is also willing to accept concentration.

The company also does not hedge its currency exposure. That matters because most of its revenue and expenses are in U.S. dollars, but it also bears shekel-denominated expenses such as payroll, rent and advisors. In 2025 that already showed up in $239 thousand of FX expense. It is not the main problem at Buff, but it is another reason the thesis is still not clean.

Outlook

Four non-obvious points to hold before talking about 2026:

• 2025 did not prove operating profitability. It proved sharp improvement, but the full year still ended with a $3.054 million operating loss.
• The second half proved there is an engine, not that the engine is already self-funding.
• Mobile is still not the base case. It is a future engine that the bank is already using as a condition for future financing.
• The campaign channel is both the solution and the risk. It earns more per user, but it also comes with volatility, concentration and short commercial visibility.

The right way to describe 2026 is a proof year. Not a clean breakout year. Not a rescue year. A proof year. Management’s own presentation lays out the markers clearly: 50% to 60% growth in direct campaigns during 2026, expansion into territories such as Germany, France and the UK, an IOS mobile launch, monetization of data assets, and stronger focus on cash and operational profitability in the second half of 2026.

The critical point is that those targets sit on a base that still has not been proven. Mobile, for example, was already presented with an ambition for daily revenue in the tens of thousands of dollars, yet in 2025 it was still not material. So 2026 should not be read through the presentation. It should be read through the transition from promises to reported numbers: how many games are actually onboarded, how many partners are added, what happens to acquisition-cost versus monetization, and when the revenue contribution becomes visible in the income statement itself.

The PC layer has its own test. Shifting from ordinary ad inventory toward direct campaigns clearly improved revenue quality per user, but it also made the company more dependent on campaign budgets, intermediaries and the ability to keep renewing short-cycle deals. The company itself describes this revenue stream as volatile. So the real 2026 question is not simply whether campaigns keep growing. It is whether they keep growing without re-creating heavy dependence on a single agency or partner.

From a funding perspective, the market will watch two things in parallel. First, did the March 2026 public raise really buy enough operating time. Second, can the company show the bank enough mobile progress to justify the next layer of debt. The fact that the second loan in the March 2026 proposal is explicitly tied to a business target related to mobile revenue says a lot: the bank is no longer willing to fund the story on vision alone.

The data and AI layer is optionality, not a base-case forecast. The January 2026 memorandum of understanding may eventually turn into something interesting, but it should not sit at the center of the 2026 underwriting case today. Anyone buying Buff mainly on the data story is getting ahead of the filing.

Over the next 2 to 4 quarters, the thesis strengthens if four things happen together: campaigns keep growing through a broader agency set, mobile moves from KPI language to visible revenue in the statements, cash burn falls without requiring another fast equity round, and the gap between operating improvement and accounting improvement narrows. The thesis weakens if the opposite happens: campaign volatility stays too high, mobile commercialization slips again, and the company needs more debt or dilution simply to buy time.

Risks

The first risk is mobile commercialization risk. In 2025 the company invested in mobile, capitalized part of development, and talked about positive KPI thresholds as the next milestone. But real revenue was still immaterial. If 2026 does not show actual commercialization, a large part of the story simply gets pushed out again.

The second risk is commercial and platform concentration. Overwolf is not just a technical vendor. It is a partner in the ad stack, a material player in direct campaigns, and an entity that can remove the app from its store or discontinue access to its ad platform. Until Buff builds a broader partner base, that remains a meaningful risk.

The third risk is funding and dilution. The company raised about $2.7 million at the start of 2025 and another roughly $2.3 million in March 2026. The bank is willing in principle to provide more funding, but not without conditions. That means shareholders are still living in a setup where both the capital markets and the bank are being asked repeatedly to fund the bridge to proof.

The fourth risk is revenue visibility. The company states plainly that it is not a backlog business, and that campaigns can create strong quarters alongside much weaker ones. Mobile-developer agreements are also short-dated and cancellable on short notice. So any attempt to read 2025 as the start of a clean linear upward path would be too simplistic.

The fifth risk is currency, privacy and platform exposure. Buff operates across global users, does not hedge its exposures, faces privacy and information-security risk, and also depends on Google, Apple and other distribution platforms. These are not the decisive risks in the story today, but in a small company with a narrow liquidity cushion even a secondary event can become meaningful quickly.

Conclusions

Buff at the end of 2025 looks better than Buff at the beginning of 2025, but it is still not clean. The company showed that the legacy PC platform can extract more value from each user, that campaigns can move the second half to a much more credible level, and that the cost-response has already started. On the other hand, year-end cash was very thin, mobile still had not shown material revenue, and a meaningful part of the bottom-line improvement came from financing and accounting, not only from operations.

Current thesis in one line: Buff proved in 2025 that the improved PC core is more valuable than it looked before, but it still has not proved that it can fund the leap into mobile without leaning again on equity, debt and key partners.

What changed versus the older read of the company is not whether there is a real product at all. It is the way the core should be read. The question is no longer “is there a business here.” The question is “is the improved PC core strong enough to carry the cost of the mobile transition.” The strongest counter-thesis is that the second half was driven mainly by warrant revaluation, capitalized development and tighter spending, so once the company has to invest again for growth, the relief fades. What can change the market’s reading over the short to medium term is a mix of continued campaign growth through more than one agency, clear early signs of mobile revenue, and a real decline in dependence on external funding.

That matters because Buff is sitting exactly on the line between a small product company with genuine improvement and a company still buying time until proof arrives. If 2026 lets mobile start contributing and shows that the second half of 2025 was a base rather than a one-off, the market read changes. If not, 2025 will be remembered as a partial improvement year, not as a full turning point.

Over the next 2 to 4 quarters Buff has to pass three very clear tests: show that campaign growth can continue without deeper one-partner dependence, move mobile out of the “immaterial” category and into reported revenue, and reduce cash burn so that financing stops being the main thesis. The thesis strengthens if operating losses stay low without accounting help and mobile revenue finally appears. It weakens if the company again needs urgent capital or if the second-half improvement turns out to have been mainly an accounting shape rather than a durable operating step.