IDT: Record Profits, a Cash Conversion Test

The 60-second read

Five things to know

1. Growth is reaching operating profit. Quarterly revenue rose 7% while operating income rose 52%; the higher-margin businesses matter more than their revenue share suggests.
2. The cash headline needs a deposit adjustment. Annual GAAP operating cash was $91.1M; removing the customer-deposit inflow leaves $60.9M, versus $107.8M a year earlier. After capex, that is $37.8M, before owner-specific adjustments.
3. NRS has a changing mix and a changed metric. Merchant services and advertising helped growth; the Rule of 40 calculation now uses total revenue growth instead of recurring revenue growth. Keep the definition change visible.
4. Guidance raises the bar. Management targets FY2027 gross profit of $545–555M and adjusted EBITDA of $176–180M. More accounting profit must eventually produce cash after settlement needs and investment.
5. Not every dollar belongs to the public shareholder. Restricted cash, minority holdings, subsidiary awards and voting control all affect how much economic value reaches IDT Class B shares. Our $60 base uses explicit ownership and cash reserves.

Three scenarios, one probability-weighted number

Three conditional cases, explicit earnings and multiples

Signposts: what would change our mind

Review dates are analytical checkpoints unless an issuer date is explicitly confirmed. A broken thesis overrides a lower-price entry test.

The tape

The market reference is a saved September 28 regular-session quote of $69.61. The separate daily-history series ends at $69.56; it is used for the chart, not substituted for the valuation reference. It is a dated input, not a live quote and not a promise that an investor can transact there. The release itself is dated September 28. We have not independently established its actual public-release clock, so the reporting-day price move is not presented as a clean reaction to this earnings announcement. The company scheduled its earnings call for 5:30 PM Eastern, a separate event from the release.

Price has moved into the upper portion of the saved annual range. That changes the burden of proof: a good quarter can be a good business result without creating an attractive entry point. At this price, the market is paying for continued progress in payments, retail software and cloud communications. Our valuation therefore tests the earnings and owner cash that those businesses can sustain, rather than comparing today with an old share-price high and calling the difference cheap.

The print

This is a stronger operating quarter than the consolidated revenue growth rate suggests. Revenue was $338.979M, while gross profit reached $134.829M. Operating income of $33.205M increased much faster than revenue. We compute these exact quarter-dollar figures by subtracting the filed nine-month statement from the annual release; they reconcile to the rounded quarterly highlights. That method also avoids financial-provider parsing errors in operating expenses and operating income. Per-share figures use the company’s quarterly release, not subtraction of annual and nine-month EPS.

The saved FMP earnings row shows $319M revenue expectations and $0.98 EPS expectations, against $339M revenue and $0.94 adjusted EPS. That suggests a revenue beat and an adjusted-EPS miss, but the provider row does not supply an analyst count or a detailed accounting-basis reconciliation. We treat it as limited context rather than a broad, reliable consensus verdict. There is no reason to recast GAAP EPS of $0.87 as the comparable expectation measure.

For the full year, operating income rose to $121.165M and net income attributable to IDT reached $86.633M. Gross profit grew faster than revenue, which is consistent with the intended shift toward higher-margin businesses. The improvement is real. It is still necessary to separate operating leverage from cash conversion, acquisition contributions, changing non-GAAP definitions and the portion of consolidated profit that belongs to other investors.

The businesses

NRS generated $45.0M of fourth-quarter revenue and $12.0M of operating income. Merchant services and other revenue was $28.5M, advertising and data was $10.1M, and SaaS fees were $4.6M. Advertising benefited from the recent acquisition, so its 49% growth is not automatically an organic growth rate. The business now reports retailer locations and average monthly network gross profit per location, giving a better way to distinguish network expansion from monetization of the existing installed base.

NRS reported 35,400 retailer locations and average monthly gross profit per location of $383, compared with $315 a year earlier. Its Rule of 40 score is now calculated using total revenue growth rather than recurring revenue growth, plus trailing adjusted EBITDA margin. The reported score of 60 is encouraging, but definition consistency matters when comparing it with old presentations or another software business. A payments and advertising network should not receive a pure software multiple merely because the company presents a software-style metric.

BOSS Money’s digital channel processed 6.6M transactions, while the retail channel processed 0.9M. Digital represented 88.1% of volume. Digital send volume grew 38%, faster than transaction growth, while average BOSS Money revenue per transaction declined to $5.68. More dollars sent by customers are not IDT revenue. The economically useful question is the margin earned on that activity after customer acquisition, funding, settlement and payout costs. Fintech gross margin improved, but selling expense also increased as the company invested in growth.

net2phone subscription revenue reached $24.5M and seats reached 447,000. Management expects the annual recurring revenue run rate to surpass $100M in the current quarter. That is a management target for a run-rate metric, not $100M of revenue already earned in the quarter. The new AI tools and integration layer may improve retention and sales, but the release does not quantify a separate AI revenue line. We therefore value the disclosed communications economics and treat AI as a product capability whose financial effect still needs measurement.

Traditional Communications still supplies substantial profit. Fourth-quarter operating income was $16.6M, and annual adjusted EBITDA was $77.3M. Within that segment, digital payments and wholesale activity offset pressure in BOSS Revolution calling. Annual gross profit nonetheless declined. A stable consolidated EBITDA figure can conceal offsetting businesses, so the model assigns traditional operations a lower multiple than the growth businesses and does not assume legacy calling suddenly becomes a growth asset.

Who receives the profit

The quarter’s $33.205M operating profit becomes $36.485M pretax income after interest and other income. Taxes consume $12.058M, leaving $24.427M consolidated net income. Another $2.717M belongs to noncontrolling interests, leaving $21.710M for IDT. This is why consolidating a subsidiary’s revenue and operating profit does not mean public IDT shareholders own all of the resulting earnings.

Non-GAAP EPS is $0.94 versus GAAP EPS of $0.87. The annual comparison is $3.82 versus $3.46. The reconciliation removes stock compensation, severance and other operating items, with tax effects. Those adjustments can clarify recurring trends but they do not all disappear economically. Employee awards transfer ownership, severance can recur, and subsidiary compensation can change the fraction of value ultimately owned by IDT. Our earnings multiple route uses GAAP EPS to avoid treating every add-back as free.

Historical adjusted EBITDA also needs a consistent basis. The current release presents FY2025 adjusted EBITDA of $131.7M, while older materials contain earlier definitions and values. The raw historical rows are retained in the dataset, and the current year-over-year comparisons use the current release’s comparable basis. We do not splice together the largest number from each version to create a smoother growth chart. GAAP operating income provides a useful second anchor when definitions change.

Earnings quality and cash conversion

The cash-flow statement is the most important cross-check in this packet. GAAP operating cash fell from $127.061M to $91.069M even as operating profit rose. Customer-deposit inflows supported both years and are not available to fund ordinary corporate operations. Removing those flows leaves $107.826M in FY2025 and $60.925M in FY2026. These calculations agree with the company’s rounded non-GAAP cash disclosure. The decline is not an accusation of bad accounting; it is an economic question about the funding required to support the growth.

The annual cash-flow statement shows $75.087M absorbed by settlement assets and disbursement prefunding. Management attributes the decline principally to the timing of these working-capital movements. That explanation is plausible for a growing remittance business, but it is a hypothesis to test over subsequent reporting dates. A temporary funding swing can reverse; a structurally larger prefunding requirement consumes capital as the business grows. One quarter cannot distinguish those outcomes reliably.

The company’s current receivable balance is $60.750M, up from the comparative $44.932M at July 2025. Using quarter-end receivables divided by quarterly revenue times 91 gives a rough collection-days proxy, not a billing-cohort study. Acquisitions and mix can affect the comparison. Inventory days are not manufactured for a service company without a separately disclosed material inventory line. A blank or inapplicable ratio is more useful than a precise number with no valid denominator.

The balance sheet and control

The release calls the balance sheet debt-free at quarter end. Unrestricted cash, debt securities and current equity investments total $271.940M. That does not include $138.650M of restricted cash, which must stay separate from shareholder surplus. Cash plus debt securities alone is $259.298M; current equities can fluctuate. Our operating valuation recognizes only $150M as excess cash, reserving the remaining $109.298M of that conservative liquid-asset measure for operating, regulatory and transaction needs. That reserve is an analyst assumption, not a company declaration that exactly that amount is legally restricted.

The April 2026 quarterly report discloses fully diluted ownership of 80.2% in NRS and 89.9% in net2phone. Those are dated ownership observations, not verified September percentages. The sum-of-parts route uses them as assumptions and separately reserves value for the acquired OnCore minority. The current release does not provide a replacement ownership table. We will update the calculation when the annual filing supplies it, rather than silently assuming that 100% of every growth business belongs to IDT.

Governance is another reason to demand an explicit valuation discount. The latest saved proxy gives Howard Jonas 70.5% of combined voting power as of October 17, 2025. Class A has three votes per share and Class B has one-tenth of a vote. Equal dividend and liquidation rights do not produce equal control. The audited FY2025 financial statements and internal-control opinions were unqualified; that is valuable evidence about reporting, but it does not remove capital-allocation risk or make related-party subsidiary awards economically irrelevant.

The August 2026 credit amendment extends the revolving facility to July 2029 and increases seasonal availability to $50M in May and December, versus $25M in other months. It is secured and carries subsidiary guarantees. Unused borrowing capacity is flexibility, not cash already owned by shareholders. An older NRS investment also carries a redemption window beginning around its fifth anniversary in September 2026. We preserve that disclosed contractual risk without asserting that redemption has actually occurred.

Three routes to value

Our base fair value is $60, with high uncertainty and a $40–85 scenario range. This is a model judgment, not a company target. The three routes cluster around the high fifties and low sixties, even though they examine different economic claims. At $69.61, the shares already require stronger conversion or more optimistic assumptions. For a new purchase under this high-uncertainty assessment, a 25% margin of safety would put the price checkpoint near $45, and only if the operating thesis remains intact.

The normalized $75M owner-cash starting point is deliberately above the current deposit-adjusted FCF result. It assumes that part of the settlement funding drag unwinds, while leaving room for capital investment, subsidiary claims and the economic cost of compensation. If cash fails to recover, this base is too generous. Conversely, consistent conversion of management’s higher EBITDA guidance could justify more. The terminal-value share makes discount rates and sustainable growth especially important; neither variable should be treated as a harmless spreadsheet default.

The reverse test keeps those model mechanics constant and solves for the normalized starting owner cash that supports the market price. The result is $91.1M, compared with our $75M assumption. This is not a forecast attributed to the market or management. It shows how much economic work the current price requires under one transparent set of assumptions. The model’s equity value at the market price is $1,740.2M; subtracting our $150M cash allowance leaves $1,590.2M assigned to operations.

The scenario values use separate illustrative earnings and multiple combinations. Bear: $2.50 EPS at 16 times. Base: $3.75 at 16 times. Bull: $4.25 at 20 times. With probabilities of 25%, 50% and 25%, respectively, the weighted value is $61.25. These are conditional cases, not confidence intervals. A regulatory failure, larger funding requirement or poor acquisition could produce an outcome below the bear case; a successful monetization could exceed the bull case. The range is a decision aid rather than a promise of limits.

Wall Street and management

The saved provider coverage is thin. The ratings summary contains one buy and one hold, while the available dated changes include Freedom Broker maintaining Buy on October 8, 2025 and a much older Chardan downgrade. No usable price-target-news or target-consensus rows were returned. We therefore do not invent a fresh Street target, average an empty dataset, or describe an old opinion as a response to this release. The absence of a target does not mean that no analyst has ever valued the business; it describes the evidence available in this research snapshot.

Management’s new ranges are the cleaner forward anchor than the thin provider estimates. The midpoint EBITDA target is $178M, about 15% above FY2026. Gross-profit midpoint is $550M, about 11% higher. Those comparisons are computed, and the ranges remain forecasts. Our hold call does not depend on claiming management is wrong. It depends on the stock already asking for a substantial portion of that progress while the owner-cash evidence remains uneven.

Management scorecard

Management has built meaningful profit streams outside legacy calling, and this quarter provides evidence that the mix shift is working. NRS monetization, digital remittance mix and net2phone operating leverage all improved. Credit is due for that execution. The next standard should be stronger, because the business now produces material earnings: management must show that higher activity can fund itself, that acquisition contributions earn an adequate return, and that subsidiary incentives do not consume the value created for the parent.

1. How much of the $75.087M annual settlement and prefunding outflow should reverse, and what permanent funding is required per dollar of remittance growth?
2. How much of NRS advertising growth came from OnCore, and what margin and minority share accompany that acquired contribution?
3. What portion of the $271.9M unrestricted liquidity is genuinely surplus after regulatory, settlement and operating requirements?
4. What are the updated fully diluted ownership percentages in NRS and net2phone after all awards and exchanges?
5. What owner-cash conversion should investors expect if the $176–180M FY2027 adjusted EBITDA range is achieved?

We would rather see management answer those questions with reconciled numbers than announce another product label. WhatsApp transfers, wallets and AI integrations can be useful, but customer economics decide their value. Capital allocation also needs a per-share test: a buyback is attractive when the shares are purchased below a defensible intrinsic value and the operating businesses still have enough funding. It is not automatically accretive simply because the common-share count declines.

Risks and the competing cases

The strongest bull case is that settlement timing masks a much stronger cash engine. NRS can monetize a larger retailer network, digital remittances can improve margins, and net2phone can sell more services into an expanding seat base. Traditional Communications still generates cash that can help fund those businesses. If FY2027 guidance arrives with sustained owner-cash conversion, today’s earnings multiple may prove reasonable and our normalization could be too cautious.

The strongest bear case is that growth requires more prefunding, selling expense and subsidiary incentives than the headline adjusted EBITDA suggests. In that case, a higher accounting margin does not produce equivalent parent cash. Traditional gross-profit erosion could remove the internal funding source just as the newer businesses demand more capital. A controlled voting structure limits the public shareholder’s ability to change that allocation, and paying a growth multiple leaves little tolerance for disappointment.

We land at hold with three-out-of-five conviction. The operating progress prevents a simple negative reading, while the price and conversion gap prevent a buy. This is not a claim that IDT is a weak business. It is an assessment of the return offered at the dated market price under our assumptions. Better conversion can raise fair value; a lower price can improve prospective returns; worsening economics can lower the base even if the shares have already fallen.

Dated checkpoints

The review dates are our monitoring checkpoints, not asserted company publication deadlines. We will replace them with verified issuer dates when available. For the next packet, every signpost above should be graded against the new filing before the narrative is rewritten. If settlement funding reverses, the thesis must acknowledge it. If it persists, calling it timing again without new evidence is not an explanation. The same discipline applies to the earnings guidance and attributable ownership.

There is no prior Charged Alpha packet for this exact IDT quarter in the production records checked for this run. These signposts establish the starting record. Future updates should preserve this $60 base, the $69.61 price reference and the stated assumptions so readers can see whether the analysis improved, rather than replacing yesterday’s assessment without a trace.

Data, sources and method

Sources and arithmetic follow a strict hierarchy. The current SEC earnings exhibit supplies the new print and forward guidance. Filed quarterly and annual financial statements supply historical accounts, ownership and risks. The SEC companyfacts dataset is used only with exact periods and retained accession provenance. FMP supplies dated market prices, estimates and coverage context. Where provider financial parsing disagrees with the statutory statement, the statutory statement wins; the rejected provider snapshot remains available for audit.

Quarterly dollar amounts for fourth quarters are computed by subtracting the filed nine-month cumulative value from the annual value. Weighted-average shares and EPS are not additive, so reported quarterly figures are used instead. Revenue, gross profit, operating income, pretax income, tax, noncontrolling income and parent income are checked through their bridges. Historical non-GAAP definitions can differ, and their original source rows remain retained. A comparison is not described as organic, constant currency or cash available to shareholders unless the source supports that interpretation.

Free cash flow here means operating cash less capital expenditure. Deposit-adjusted free cash flow additionally removes customer-deposit changes; it remains a consolidated analytical measure before all owner-specific adjustments. DCF means discounted cash flow. NCI means noncontrolling interests. ARR is annual recurring revenue run rate, which is not the same as earned annual revenue. SBC is stock-based compensation. A margin of safety is a discount to our uncertain model value, not a guarantee against permanent loss. Probability weights are analytical judgments and can be wrong.

• ex_950508.htm — saved primary source; SHA-256 38d1f0b965e1ed0c…
• idt20260430_10q.htm — saved primary source; SHA-256 38e339d46aa345fb…
• formdef14a.htm — saved primary source; SHA-256 87a3a757afb43d01…
• form10-k.htm — saved primary source; SHA-256 3ee4ccb54e2bc002…
• CIK0001005731.json — saved primary source; SHA-256 be48dc34c2936415…
• idt20260814_8k.htm — saved primary source; SHA-256 bb137d53674829cd…
• 0001437749-26-028388.txt — saved primary source; SHA-256 f23ed0d7396c8763…

All full source URLs, capture times and complete SHA-256 digests are retained with the episode dataset. Market snapshots were captured on September 28, 2026. The packet is educational research and is not individualized investment advice. Our models omit no known debt merely because the company calls itself debt-free; operating leases, customer obligations, subsidiary claims and working-capital needs are considered separately. We have not inferred an earnings-release time from an EDGAR acceptance timestamp.