Repay Holdings: a working payments business, but too little protection for common equity

REPAY looks cheap at about 6x EBITDA. After the KUBRA acquisition, however, the question for common shareholders is not how fast EBITDA grows. It is how much cash is left after reinvestment, interest, taxes, TRA payments and compensation.

At $3.84 on September 22, 2026, the shares trade 15.8% below my present Base value of $4.56. That is a real discount, but far short of the 55% margin of safety this leverage requires. A functioning payments business can still be an unattractive new purchase when leverage, reinvestment and senior cash claims leave too little protection for common equity.

Route: SF-01 Standard Value, subtype leveraged transaction-based integrated payments and bill presentment platform. Economic classification: Standard Company (4/12). Primary method: STANDARD equity multiple on normalized Year 3 owner earnings, discounted to today. Secondary method: 10-year enterprise DCF. All values are present values per economic share, not 2029 price targets.

The investment thesis: cash conversion, not headline growth

REPAY provides integrated payment processing and bill payment solutions. Consumer Payments serves consumer lending, automotive, receivables and recurring bills. Business Payments focuses on accounts payable automation and supplier payments. REPAY earns payment processing and service economics. It does not lend its own balance sheet, so bank, utility rate base and SaaS ARR valuation methods do not apply.

KUBRA, acquired on June 1, 2026, adds bill presentment and customer communications for utilities, government and insurance. The closing announcement cited about $372 million in cash. The subsequent 10-Q reports about $354.1 million of aggregate consideration and $348.2 million of cash paid net of acquired cash. This analysis uses the statutory figures.

KUBRA's FY2025 revenue was about $239 million, with adjusted EBITDA of $49 million and a gross margin around 48%. Management's combined FY2025 reference, before synergies, is about $548 million of revenue and $178 million of adjusted EBITDA.

The enlarged business could create considerably more cash. But the common equity claim is leveraged to reinvestment, refinancing and compensation costs. An apparently low enterprise multiple is not sufficient evidence of a large margin of safety.

Routing and economic metric

The valuation anchor is normalized owner earnings after required reinvestment, interest, entity-level cash taxes, TRA payments and an economic compensation charge. A STANDARD equity multiple is applied only after this bridge.

Headline adjusted EBITDA and company-defined free cash flow are comparison metrics only. Q2's 75% company FCF conversion is not multiplied by four.

Current financial performance

FY2026 guidance includes seven months of KUBRA. KUBRA contributed about $21 million of revenue in June alone, so the 33% headline growth is not an organic run rate. The Business Payments split is more useful than the acquisition-inflated consolidated figure. New distribution relationships, including NxtEdge/DiningEdge and Visa Platform Connect, are potential channels, but no separate cash flow is added for them.

Debt, TRA and share count

The revolver is borrowing capacity, not cash or equity. Restricted cash of $43.8 million is excluded from excess cash, and $20 million of cash is treated as an operating reserve, leaving $63.7 million of excess cash. Debt is counted at principal, not at the lower amortized book value.

The converts mature on July 15, 2029, with an initial conversion price of about $13.02. The term loan has a springing maturity provision that generally moves its maturity to 91 days before the notes mature unless specified exceptions are met. Every scenario refinances the existing principal without assuming an equity raise.

Existing options (0.80 million at a $6.13 weighted exercise price) are added, together with their exercise proceeds, only when the modeled value exceeds the strike. Future compensation is charged as a recurring cash-equivalent cost rather than through a second automatic share growth schedule, so it is not counted twice. The award schedule does not reconcile perfectly to the EPS disclosure, and the difference is carried as a conservative reserve.

Cash flow quality

The $22.5 million purchase of a distribution partner's economic interests bought an economic asset. It is not evidence of a recurring $45 million annual maintenance bill, so it is neither ignored in history nor annualized into the forecast. A separate normalized renewal allowance is used instead. The negative H1 residual is a diagnostic, not a claim that the business permanently burns cash.

The five-year average company FCF of about $54.9 million is not the valuation denominator, because perimeter, working capital, financing, SBC and TRA differ across years. For illustration, 2025 company FCF less $16.3 million of TRA and $18.3 million of SBC leaves $14.5 million. The analogous 2024 figure is $80.3 million, helped by favorable working capital. Neither year alone is a clean normal.

Normalization: the Base model

Years 1, 2 and 3 are rolling annual periods ending around September 2027, 2028 and 2029, each with a full year of KUBRA. Every forward row is an estimate, not company guidance.

EBITDA reconciles as $178 million plus operating improvement plus cost synergies. The program's roughly $15 million of cost synergies is kept separate from about $5 million of capex savings, which sit in the investment line and are not added to EBITDA again. Base investment includes a $5 million annual intangible renewal allowance. The assumption that investment intensity falls from 11.0% to 9.0% of revenue is necessary for this Base case, not an achieved result.

Year 1 owner earnings are $31.8 million against $203 million of adjusted EBITDA, a 15.7% conversion. By Year 3 the model requires about $63.4 million on $236 million, or 26.9%.

Primary valuation: STANDARD equity multiple

Owner earnings equal adjusted EBITDA, less required investment, cash interest, entity cash taxes, recurring compensation, other recurring costs, working capital and TRA. At least $5 million of term principal is repaid each year, and positive owner earnings otherwise go to debt reduction, which lowers later interest. That cash is not counted again as a dividend or a terminal cash pile.

Base 9x sits inside the framework's 8x to 10x range for risky companies and assumes a functioning, moderately growing business after meaningful deleveraging. Bear 6x is an explicit distressed override. Bull 11x requires demonstrated cash conversion and lower risk, and at a 15% required return it implies roughly 5.4% long-run cash growth, so it is not a SaaS premium.

Base: Year 3 owner earnings of $63.4 million times 9, less the remaining finance lease, discounted three years at 15%, plus $63.7 million of excess cash, divided by 100.6 million shares, gives $4.35. Because interest and TRA payments already reduce owner earnings, debt and the TRA liability are not subtracted again.

Secondary valuation: 10-year enterprise DCF

The DCF values unlevered operating cash instead of choosing an equity multiple. It shares operating evidence with the primary method, so the two are not statistically independent. Economic equity equals enterprise value less $787.5 million of debt and $1.5 million of finance leases, plus $63.7 million of excess cash, less the $194.3 million TRA fair value. The TRA is deducted once here because FCFF is before TRA payments.

Base WACC of 10.5% combines a 15% equity hurdle and about 9% cost of debt with a 20% tax shield on a 40/60 equity to debt mix, rounded up from 10.3%. After Year 3, Base cash flow growth fades from 5% to 2%. In Bear, enterprise value falls below the financial and TRA claims, so common equity is floored at zero. That is a stress result, not a claim that default is certain.

Historical cross-check

A cheap-looking company FCF multiple can coexist with a much higher owner cash flow multiple. A clean historical multiple distribution on today's post-KUBRA denominator does not exist, so no historical target price is used. Historical share prices are context, not a floor.

Valuation triangulation

The primary method receives 75% and the DCF 25% in every scenario. Owner earnings are the most direct metric for a highly levered equity, while the DCF is more exposed to tax, TRA duration and long-run fade assumptions.

Base convergence passes the framework's 20% test. The tails do not: the Bear DCF is zero and the Bull divergence exceeds 30%. That uncertainty is not solved by averaging. It is reflected in the Confidence Score.

What is the market already pricing in?

Using the same cash, share count, horizon and capital structure as the forward model:

The market does not price flawless execution. It prices a materially smaller cash flow achievement than the Base case, which is the strongest valuation argument against an overly bearish view. Upside can come from about $8.7 million more Year 3 owner earnings, a multiple rising from about 7.8x to 9x, or a combination. But the expected value advantage is modest once substantial probability is given to a severe equity downside, and it is much smaller than the required margin of safety.

Analyst targets averaged $6.85 (median $6.00, range $4.25 to $10.00, five analysts). Targets are forward-looking and differently dated, so they do not replace the Base case.

For an optical check, 95.15 million shares at $3.84 give equity of about $365.4 million. Adding face debt and finance leases and subtracting all unrestricted cash gives an enterprise value of about $1,070.7 million before TRA. That is about 6.2x the FY2026 EBITDA guidance midpoint of $172.25 million, or 5.3x the model's full-perimeter Year 1 EBITDA of $203 million (6.2x if the TRA claim is added). Neither is the owner earnings valuation.

Bear, base and bull scenarios

Each cell shows Year 1 / Year 2 / Year 3 in $ million. All forecasts are estimates, and interest and debt are calculated from them.

Bear assumes slow growth, only $2 million, $4 million and $5 million of annual cost synergies, persistent reinvestment and a costly refinancing. Revenue need not collapse for the residual equity claim to be severely impaired. Base lifts EBITDA from $178 million to $236 million through operating improvement and $15 million of synergies, with cash discipline and no further acquisition or takeover premium. Bull uses the same $15 million synergy ceiling and reaches $272 million of EBITDA through stronger volumes, monetization and operating leverage, while reinvestment and compensation remain real costs.

The 7.1% gap between the probability-weighted value and the price compares present values. It is not an annual return or a three-year target. The probabilities are subjective, not market-implied, and a restructuring tail can still reach zero even though the triangulated Bear value is positive.

Sensitivity

The DCF is sensitive to the cost of capital. That is a reason for a low robustness score, not permission to pick the most favorable cell.

Twelve-month price touch: conditional illustration only

Calibrated probabilities are not available. The table assumes zero-drift random price movement with an assumed, not observed, annual volatility. It is a reproducible sensitivity, not a price forecast, and the events do not add up to 100%.

Where the stock becomes attractive

Applied to the $4.56 Base value, the maximum buy price is $2.05. At 60%, the conservative edge is $1.82, giving a conditional buy band of $1.82 to $2.05, valid only with an intact thesis. Below it, check whether the decline reflects new fundamental damage.

At $3.84, the shares trade 87.2% above the maximum buy price. Reaching it would take a 46.6% decline with unchanged fundamentals. That is not a prediction that such a decline will happen.

REPAY pays no regular common dividend, and the valuation assumes zero dividend income.

Reasons to own or follow RPAY

These are reasons to own RPAY at an adequate price, not a conclusion that today's price meets the required margin of safety.

• Embedded workflows. Replacing payment and billing integrations takes real effort, so a moderate moat can support repeat cash generation without a dominant network effect.
• Broader recurring billing. KUBRA's utility and government relationships add a different demand mix from consumer credit repayment flows.
• Business Payments growth. Normalized organic growth of 19% outpaced Consumer Payments, though the model does not apply that rate to the whole company.
• Debt reduction. In Base, financial debt falls from $789.0 million to about $639.0 million by Year 3, helping absorb the higher refinancing coupon on the converts.
• Measurable commitments. Synergy and leverage targets let investors judge capital allocation in dollars of cash and debt.

What could go wrong?

• Cash conversion. Software, equipment and relationship renewal may consume more cash than the EBITDA presentation suggests. An extra $10 million of annual reinvestment cuts the Base value from $4.56 to about $3.74.
• Refinancing. Cheap 2029 converts may be replaced by much more expensive debt, and the springing term loan maturity reduces flexibility. A 200 basis point financing shock cuts the primary value to about $3.41.
• Operations. Security incidents, payment network dependence, client losses or monetization pressure can harm repeat economics.
• Capital allocation. Further aggressive M&A, dilution or large TRA and reinvestment outflows can prevent per-share value creation even while revenue rises.

The chief risk is a combination: slow cash improvement, expensive refinancing and weak principal reduction. Restricted cash and book goodwill do not give common shareholders a dependable liquidation floor.

Management execution

In 2025, REPAY spent $38.5 million on buybacks and repaid $72.0 million of debt, but the subsequent large debt-funded acquisition means buybacks alone do not establish an excellent capital allocation record. Repeated acquisition impairments, cash metrics that need many exclusions and possible reliance on EBITDA growth rather than nominal debt repayment are red flags. Detailed public reconciliations and measurable commitments are positives.

The board rejected Forager Capital's revised non-binding $5.25 per share proposal on July 13, 2026 as undervaluing the company. That is the board's opinion, not an intrinsic value fact, and the offer is neither a floor nor part of the Base case.

Management credibility is assessed as mixed: 9/20.

Stock Analyza scorecard

The low return on capital score means there is insufficient evidence of durable superior returns, not that a reported ROIC is negative.

Undervalued versus Base and a limited-margin valuation score are not contradictory. The first uses only the gap to Base, while the score also weighs Bear protection, convergence and data risk.

Confidence score

A C grade means the central estimate is usable only with a wide safety margin and explicit monitoring. If financing ever becomes a near-term binary survival question, the framework's D rating and no-buy override would replace this model.

What I would watch from here

The next quarterly report will be the first with a full quarter of KUBRA. Management has also indicated an investor day in December 2026.

The three most important thesis breakers

1. Persistently inadequate owner cash after required investment and TRA.
2. Inability to solve the 2029 refinancing without major harm to common equity.
3. Renewed capital allocation that reverses deleveraging.

Adversarial review

Two internal review passes were performed: an arithmetic and capital structure recomputation, and an adversarial economic review. A separate checker recomputed the scenario bridges, both valuations, the reverse valuation, the sensitivity grids, the scores and the margin of safety; 184 numerical checks passed.

The review used debt principal instead of book value, a full economic share bridge, and a cash-equivalent compensation charge. It removed double deductions of TRA and capex, did not annualize the one-off partner purchase as maintenance, isolated financing shocks from operating taxes, and replaced unsupported price-touch point estimates with a labeled volatility illustration. Corrections moved values in both directions.

The strongest counter-thesis is that KUBRA creates more durable billing economics, investment normalizes faster, synergies arrive and Business Payments keeps growing. The market-implied Year 3 owner earnings of about $54.7 million are below the $63.4 million Base endpoint.

The verdict survives the moderate one-factor and consensus-centered tests, but not the strongest optimistic combination. That is why the evidence needed to upgrade matters. The publication gate status is PASS_WITH_WARNING because the share award reconciliation, cash tax and TRA duration, and post-acquisition normalization remain material estimates.

Final verdict: hold at $3.84, watchlist for new money

The payment business is not worthless. The problem is that the common equity claim offers too little protection against owner cash flow disappointment and refinancing costs. The probability-weighted value is only $4.11, and a large headline Bull value does not remove the downside.

Verdict: WATCHLIST for new money at $3.84, conditional HOLD for existing holders. Triangulated Base value $4.56, upside 18.7%, required margin of safety 55%, maximum buy price $2.05.

For existing holders, holding is conditional on evidence of cash conversion and refinancing progress, not on recovering the purchase price or another takeover proposal. A price below $2.05 is a conditional re-entry trigger, not an unconditional order.

The strongest thesis breaker is persistent failure to generate owner cash after required investment and TRA, while the debt cannot be refinanced without substantially impairing the common equity claim.

Sources

Primary sources include REPAY's June 2026 Form 10-Q, the Q2 2026 earnings release and earnings supplement, the KUBRA closing announcement and closing presentation, the August 2026 investor presentation, the FY2025 results and the rejection of the revised Forager proposal.

Management guidance history comes from REPAY's full-year results for 2021, 2022, 2023 and 2024. Distribution context: NxtEdge/DiningEdge and Visa Platform Connect announcements. Analyst targets, consensus estimates and historical year-end prices are from a third-party market data aggregator and serve only as secondary cross-checks.

The reference market price is $3.84 at 2:35 pm EDT on September 22, 2026, regular session, not a closing price. Balance sheet, debt and rates are as of June 30, 2026.