Research Solutions, Inc

Item 7. Management’s Discussion and Analysis of Financial Condition and Results of Operations

Cautionary Notice Regarding Forward-Looking Statements

The following discussion and analysis of our financial condition and results of operations for the years ended June 30, 2026 and 2025 should be read in conjunction with our consolidated financial statements and related notes to those financial statements that are included elsewhere in this report. Our discussion includes forward-looking statements based upon current expectations that involve risks and uncertainties, such as our plans, objectives, expectations and intentions. Actual results and the timing of events could differ materially from those anticipated in these forward-looking statements as a result of a number of factors, including those set forth under “Risk Factors” and elsewhere in this report.

We use words such as “anticipate,” “estimate,” “plan,” “project,” “continuing,” “ongoing,” “expect,” “believe,” “intend,” “may,” “will,” “should,” “could,” and similar expressions to identify forward-looking statements. All forward-looking statements included in this report are based on information available to us on the date hereof and, except as required by law, we assume no obligation to update any such forward-looking statements.

Overview

Research Solutions was incorporated in the State of Nevada on November 2, 2006, and is a publicly traded holding company with five wholly owned subsidiaries as of June 30, 2026: Reprints Desk, Inc., a Delaware corporation, including its wholly owned subsidiary Resolute Innovation, Inc., a Delaware corporation, Scite, LLC, a Delaware limited liability company, Reprints Desk Latin America S. de R.L. de C.V., an entity organized under the laws of Mexico, and RESSOL LA, S. DE R.L. DE C.V., an entity organized under the laws of Mexico.

We are a vertical software-as-a-service (“SaaS”) and artificial intelligence (“AI”) company providing software and related services to help research-intensive organizations simplify the research process, save time and money. We offer various software platforms (“Platform” or “Platforms”) that are typically sold to corporate, academic, government and individual researchers as cloud-based SaaS via auto-renewing license agreements. Corporate, academic, and government customers typically sign up under annual or multi-year agreements paid annually in advance. Individual researchers can sign up under an annual or a month-to-month agreement and are typically billed monthly. Our Platforms also facilitate the sale of published scientific, technical, and medical (“STM”) content sold as individual articles (“Transactions”) either stand alone or via one or more of the research Platform solutions we provide. When one or more of the Platform solutions are used to purchase Transactions, customers pay for those transactions through monthly billing or via credit card for individual researchers. In addition, our Platforms facilitate rights and permissions for customers to re-use content, ensuring copyright compliance for research, regulatory and marketing use cases as well as the utilization of content with AI applications and for the training of AI models. Our Platforms enable life science and other research-intensive organizations to simplify their research and development activities through our advanced search (i.e. Discovery Tools), tools to access and buy STM articles required to support their research (i.e. Access), as well as tools that manage that content across the enterprise and on an individual basis (i.e. Manage). The Platforms also include advanced AI (“Generative AI”) based assistants to help researchers understand the quality of the articles they are reviewing, speed up the review process, and to more fully understand how various research papers relate to each other.  In addition to STM content, the Platforms provide additional context to the research process by including the ability to search and assimilate a variety of other types of data such as patent, clinical trial, regulatory and competitive intelligence data. They also typically deliver a return on investment to the customer by reducing the amount of time it takes a research organization to find, acquire and manage content, in addition to also driving down the ultimate cost per article and overall research costs over time.

Platforms

Our cloud-based SaaS Platforms consist of proprietary software and Internet-based interfaces sold to customers through an annual or monthly subscription fee. Legacy functionality falls into three areas.

Discovery Tools – Our Scite.ai and Resolute.ai solutions facilitate search (discovery) across virtually all STM articles available. These solutions include basic search solutions and advanced search tools. These tools allow for searching and identifying relevant research and then purchasing that research through one of our other

24

Table of Contents

solutions. In addition, these tools increasingly enable users to find insights in other datasets adjacent to STM content, such as clinical trial, patent, life science & medtech regulatory information, competitor and technology landscape insights, in addition to searching the customer’s internal datasets. Scite.ai includes full text search capability on most of the world’s STM content providing better search results and citation information as supporting or contrasting evidence. This powers our AI assistant and literature search engine and gives researchers better insights into any topic. The advanced search solutions are sold through a seat, enterprise, or individual license. These Platforms are deployed as a single, multi-tenant system across our entire customer base. Customers securely access the Platforms through online web interfaces and via web service APIs that enable customers to leverage Platform features and functionality from within in-house and third-party software systems. The Platforms can also be configured to satisfy a customer’s individual preferences. We leverage our Platforms’ efficiencies in scalability, stability and development costs to fuel rapid innovation and competitive advantage.

Access – Our Article Galaxy® (“AG”) and Article Galaxy Scholar (Academic Library version) (“AGS”) solutions allow for research organizations to load their entitlements (subscriptions, discount or token packages, and their existing content library of articles) and AG/AGS manages those entitlements in the background enabling the researchers to focus on acquiring articles they need quickly and efficiently at the lowest possible cost. When used in conjunction with our Discovery Tools Platforms, customers can initiate orders, route orders based on the lowest cost to acquire, obtain spend and usage reporting, automate authentication, and connect seamlessly to in-house and third-party software systems. In addition, Article Galaxy facilitates rights and permissions for various re-use cases, including the utilization in AI applications and training of AI applications, ensuring copyright compliance for our customers.

Manage – Our References solution offers a comprehensive reference management solution with built-in document delivery capabilities specifically designed to meet the collaboration and security needs of research- intensive organizations.  This user-friendly Platform enables researchers to seamlessly organize their literature, collaborate with team members, and access a vast collection of scientific content. By integrating organization tools with instant access to millions of scholarly articles, our References solution streamlines the research workflow and enhances productivity for scientific professionals.

AI models are integral to powering the unique insights our platforms provide as well as the user experience customers enjoy. Natural language processing (“NLP”) and AI models are used to enhance metadata, define connections between topics and content items as well as to generate data and metrics employed to enable users to rapidly identify and understand the value of content they need for their research. We also use state of the art AI models, such as large language models (“LLMs”) to include generative AI “assistants” in several parts of the research workflow today and will continually add capability as we move forward. Today we employ generative AI technologies as a basis for our recommendation engine in our Discovery Tools, Access, and Manage Platform solutions. In addition, generative AI based “assistants” in some of our solutions allow the researcher to ask questions about articles, groups of articles (folders), and more. We also have the capability to provide near full text search on STM content in the Scite.ai solution where the publisher gives us the rights to do so. The ability to not only mine an article’s full text but also show snippets of full text is unique to our Company and allows our generative AI assistants to provide highly accurate results with a very low incidence of hallucinations as part of a retrieval augmented generation framework focused just on STM content. We intend to continue investing in our platforms and in our integrations with third-party AI applications, through new product enhancements and expanded dataset coverage.

Our Platforms are generally deployed as a single, multi-tenant system across our entire customer base. Customers securely access the Platforms through online web interfaces and via web service APIs that enable customers to leverage Platform features and functionality from within in-house and third-party software systems. Our Platforms can also be configured to satisfy a customer’s individual preferences. We leverage our Platforms efficiencies in scalability, stability and development costs to fuel rapid innovation and to gain a competitive advantage.

Transactions

We provide our researchers with a single source to the universe of published STM content that includes over 200 million existing STM journal articles for instant download, 50 million journal articles for rent, 10 million online book chapters, and 45 million only in print journal articles. In addition, we add between 2 to 4 million newly published STM

25

Table of Contents

articles each year. STM content is rented or sold to our customers on a per transaction basis. Researchers and knowledge workers in life science and other research-intensive organizations generally require single copies of published STM journal articles for use in their research activities. These individuals are our primary users and while they typically purchase the articles via one of our Platform solutions, we do have some customers that just order articles from us on behalf of end-users in their organizations.

Core to many of our Platform solutions is providing our customers with ways to find and download digital versions of STM articles that are critical to their research. Customers submit orders for the articles they need which we source and electronically deliver to them generally in under an hour, in most cases in seconds. This service is generally known in the industry as single article delivery or document delivery. We also obtain the necessary permission licenses from the content publisher or other rights holder so that our customer’s use complies with applicable copyright laws and we are expanding these services to include the use of content in AI applications and for the training of AI models. We have arrangements with hundreds of content publishers that allow us to distribute their content. The majority of these publishers provide us with electronic access to their content, which allows us to electronically deliver single articles to our customers often in a matter of seconds. While a vast majority of the articles are available in electronic form, the Company also has workflows to deliver older paper-based articles through relationships we have built with libraries around the world.

Inflation Risk

We do not believe that inflation has had a material effect on its operations to date, other than its impact on the general economy. However, there is a risk that our operating costs could become subject to inflationary and interest rate pressures in the future, which would have the effect of increasing our operating costs, and which would put additional stress on our working capital resources.

Critical Accounting Policies and Estimates

The preparation of our consolidated financial statements in conformity with accounting principles generally accepted in the United States, or GAAP, requires us to make estimates and assumptions that affect the reported amounts of assets, liabilities, revenue and expenses, and related disclosure of contingent assets and liabilities. When making these estimates and assumptions, we consider our historical experience, our knowledge of economic and market factors and various other factors that we believe to be reasonable under the circumstances. Actual results may differ under different estimates and assumptions.

The accounting estimates and assumptions discussed in this section are those that we consider to be the most critical to an understanding of our consolidated financial statements because they inherently involve significant judgments and uncertainties.

Software Costs

Based on its nature, our software development costs are expensed as incurred. The finalization of our project development process precipitates the rapid commercialization and deployment of new products and enhancements. We continuously review our projects, processes and the nature of our software development costs to determine if there are changes that would meet the requirements for capitalization under Accounting Standards Codification (“ASC”) 350-40, Internal-Use Software.

Revenue Recognition

We account for revenue in accordance with Financial Accounting Standards Board (“FASB”) Accounting Standards Update (“ASU”) 2014-09, Revenue from Contracts with Customers (Topic 606), (“ASC 606”). The underlying principle of ASC 606 is to recognize revenue to depict the transfer of goods or services to customers at the amount expected to be collected.

Revenues are recognized when control of the promised goods or services are transferred to a customer, in an amount that reflects the consideration that we expect to receive in exchange for those goods or services. We derive our

26

Table of Contents

revenues from two sources: annual licenses that allow customers to access and utilize certain premium features of our cloud-based SaaS research intelligence platforms and the transactional sale of STM content managed, sourced and delivered through the Platform.

We apply the following five steps in order to determine the appropriate amount of revenue to be recognized as we fulfill our obligations under each of our agreements:

Platforms

We charge a subscription fee that allows customers to access and utilize certain premium features of our Platforms. Revenue is recognized ratably over the term of the subscription agreement, which is typically one year, provided all other revenue recognition criteria have been met. Billings or payments received in advance of revenue recognition are recorded as deferred revenue.

Transactions

We charge a transactional service fee for the electronic delivery of single articles, and a corresponding copyright fee for the permitted use of the content. We recognize revenue from single article delivery services upon delivery to the customer provided all other revenue recognition criteria have been met.

Stock-Based Compensation

We periodically issue stock options, warrants and restricted stock to employees and non-employees for services, in capital raising transactions, and for financing costs. We account for share-based payments under the guidance as set forth in the Share-Based Payment Topic 718 of the Financial Accounting Standards Board (“FASB”) Accounting Standards Codification, which requires the measurement and recognition of compensation expense for all share-based payment awards made to employees, officers, directors, and consultants, including employee stock options, based on estimated fair values. We estimate the fair value of stock option and warrant awards to employees and directors on the date of grant using an option-pricing model. Depending on the type of restricted stock award, the fair value of our restricted stock is estimated based on the market price of our common stock on the date of grant or with the assistance of a valuation specialist, using the Monte Carlo simulations on a binomial model with a derived service period. We recognize compensation expense on the straight-line basis over the requisite service period for awards subject to time vesting conditions and the graded tranche basis for awards subject to market vesting conditions. Forfeitures are accounted for as

27

Table of Contents

they occur. We recognize stock-based compensation within the consolidated statements of operations and comprehensive income with classification depending on the nature of the services rendered.

Under ASC 718, repurchase or cancellation of equity awards, the amount of cash or other assets transferred (or liabilities incurred) to repurchase an equity award shall be charged to equity, to the extent that the amount paid does not exceed the fair value of the equity instruments repurchased at the repurchase date. Any excess of the repurchase price over the fair value of the instruments repurchased shall be recognized as additional compensation cost.

Allowance for Credit Losses

Our trade accounts receivable are recorded at amounts billed to customers and presented on the consolidated balance sheets net of the allowance for estimated credit losses, and typically due within 30 days. We evaluate the collectability of our trade accounts receivable based on a number of factors. In circumstances where we become aware of a specific customer’s inability to meet its financial obligations to us, we estimate and record a specific reserve for bad debts, which reduces the recognized receivable to the estimated amount we believe will ultimately be collected. In addition to specific customer identification of potential bad debts, bad debt charges are recorded based on our historical losses, our forecast and an overall assessment of trade accounts receivable outstanding. We established an allowance for doubtful accounts of $103,217 and $182,324 as of June 30, 2026 and 2025, respectively. We recorded a credit to bad debt expense of approximately $23,000 in the year ended June 30, 2026 and a bad debt expense of approximately $141,000 in the year ended June 30, 2025. We had write-offs of approximately $56,000 and $49,000 in the years ended June 30, 2026 and 2025, respectively, which reduced our allowance for doubtful accounts.

Foreign Currency

The accompanying consolidated financial statements are presented in United States dollars, the functional currency of our company. Capital accounts of foreign subsidiaries are translated into US dollars from foreign currencies at their historical exchange rates when the capital transactions occurred. Assets and liabilities are translated at the exchange rate as of the balance sheet date. Income and expenditures are translated at the average exchange rate of the period. Although the majority of our revenue and costs are in US dollars, the costs of Reprints Desk Latin America and ResSol LA are in Mexican Pesos. As a result, currency exchange fluctuations may impact our revenue and the costs of our operations. We currently do not engage in any currency hedging activities.

​

​

​

​

28

Table of Contents

Comparison of the Years Ended June 30, 2026 and 2025

Results of Operations

​

​

Revenue

​

​

29

Table of Contents

Total revenue decreased $751,024 or 1.5%, for the year ended June 30, 2026 compared to the prior year, due to the following:

​

​

Cost of Revenue

​