Digital Turbine
Software built into Android phones before they are sold, which puts apps and advertisements on the screen, run beside an ordinary advertising exchange.
The scores
Growth3/4 ↑
Revenue up 26.8% and organic after three years of decline, but the record is one year old and the company's own guidance implies it halving.
Business quality2/4 ↓
A real firmware position covers two thirds of revenue; the growing third faces AppLovin, at $5.48B of revenue against $565M.
Financial position2/4 ↑
Digital Turbine still owes $387.8M on loans due 2029, and says replacing them is next year's main cash need.
Execution record2/4 ↑
A results release corrected within a day, an interim finance chief who is also chief accountant, a $336.6M write-off, three years of falling sales.
Per share capture2/4 →
Sales grew 15.2% and sales for each share 5.4%. Money raised from shareholders goes to the lender by contract, as does half of spare cash.
Potential2/4 →
The alternative-distribution prize needs Google, carriers and device volumes all to cooperate, and the largest new agreement states no volume and no term.
Risk profile2/4 →
One repayment date, 2029, that the company says it must replace - settled by lenders in a single step, not by the business.
Valuation2/4 ↓
Buyers pay $26.80 for each dollar of trailing operating profit, and 2.24 times sales against a 1.20 six-year median.
Evidence quality3/4 →
Central claims are filed figures plus IDC and Google's own documentation, but the operating drivers exist only on the call.
The Verdict
We rate Digital Turbine unattractive. Its software sits inside Android phones before they are sold, a spot no advertiser can buy, but it earns less from operations than it pays in interest, and the phone market it takes a share of is shrinking, so the business is weak. The shares have more than tripled on a court ruling opening Android to rival app stores, yet Google is delivering that inside its own store. The price needs operating profit to roughly double and its costly loans to be replaced cheaply, so it is full, not cheap. A lower price would not fix borrowing on a shrinking market.
The long caseCourts are opening phone app stores, but Google is delivering that inside its own store, and fewer phones are sold; Digital Turbine must still earn more on each phone.
The evidence under this - 9 claims
Guidance is a floor a management team chooses, not a forecast it believes, and this one has already been raised once in the same fiscal year - so the honest reading is that the company is guiding conservatively after one strong quarter rather than signalling a slowdown it can see. The comparison base also hardens through the year: the fiscal 2026 quarters the guide is measured against grew 18.2%, 12.4% and 19.6%, against a first quarter that grew nothing measurable because the year-ago figure is not given, so some deceleration in the growth RATE is arithmetic rather than deterioration. What survives both points is that the company's own published range does not extrapolate the June quarter, and the price has.
The chief executive's reading is the opposite and it is not unreasonable: more app stores means more parties needing distribution and targeting, and Digital Turbine sells to them as customers rather than competing with them. A store listed inside Play still has to acquire users, which is what the demand-side platform and SingleTap are for, and the company reports alternative-application partners ramping. There is also no measured evidence either way yet - the program took effect eight weeks before this thesis and nobody has published enrolment numbers, so the direction of the effect on Digital Turbine's revenue per device is an open question rather than a settled one. What is settled is that the remedy is being delivered inside Play, which is not the remedy the equity story assumes.
Two readings push back. On enterprise value against EBITDA the frozen pack puts the shares at 18.8 times against a six-year median of 19.2 - cheaper than their own history, not dearer - and on the guided $150 million of adjusted EBITDA the multiple is about 11 times. And operating profit is depressed by acquisition intangible amortisation: $208,485 thousand of intangibles remained at 30 June 2026 and $16,805 thousand of depreciation and amortisation ran through the June quarter alone, so the operating line carries the cost of businesses bought years ago and already impaired. The counter to the counter is that both of those are the point: EBITDA is the metric that removes the interest, and amortisation of what was acquired is exactly what a serial acquirer's valuation frame must keep in.
The company
Phone networks and handset makers build Digital Turbine's software into Android phones before sale, so it is there the first time the phone is switched on. It places apps in the phone's setup, puts news and weather on the screen, and installs an app in one touch without an app store. That is two thirds of sales.
The other third is an ordinary advertising business: app makers buy advertisements through it, and app owners sell their ad space by automatic auction.
The evidence under this - 2 claims
How it makes money
Advertisers pay for each placement on the phone, and the network or handset maker that installed the software takes a share. What was left in the June 2026 quarter was 45 cents of each dollar. The auction half keeps 80 cents, because it only matches buyers with sellers. No single customer is a tenth of sales.
The evidence under this - 2 claims
Where the money comes from, stream by stream
| Stream | Share of sales (fiscal 2026 (year to March 2026)) | How it earns it |
|---|---|---|
| On Device Solutions | 67.7% | Software carriers and phone makers preload on Android phones; paid per app installed and per ad or story shown, sharing revenue with the partner. |
| App Growth Platform | 32.9% | An advertising exchange and network: app makers buy ads and app owners sell their ad space by auction; mostly booked net of payouts. |
Who runs it
William Stone has run Digital Turbine since 2014. He bought three advertising companies in 2021; sales then fell for three years and a large part of what was paid was written off. Sales are growing again now. The finance chief left in May and the head of partnerships leaves in October, so finance is run by an interim. Insiders own about 6%. Bonuses ride on sales and on profit counted before interest, so no pay target is measured after the interest bill.
The evidence under this - 5 claims
The fiscal 2027 performance units add two relative total shareholder return goals to adjusted EBITDA (C29), so half of the newest performance award now rides on the share price, which does carry the interest bill and the dilution.
Why it wins, and who could beat it
Why customers come back
The software lives inside the phone's own system, put there by the company that sells the phone. Reaching that spot takes a deal with each network or handset maker, then testing on every model. That took a decade, and advertising money cannot buy it.
It protects only two thirds of the company. The auction third has no such position and faces AppLovin, which does the same job at a 76% operating margin on 9.7 times the sales.
The evidence under this - 4 claims
The threat
Google. The company's own annual report names Google Play as the main rival for its phone software. Since 22 July 2026 Google has let rival app stores reach Android users from inside Play, so a rival store now installs like any other app, with no network deal and nothing built into the phone. That removes the very difficulty this product exists to remove.
The evidence under this - 3 claims
The chief executive's reading is the opposite and it is not unreasonable: more app stores means more parties needing distribution and targeting, and Digital Turbine sells to them as customers rather than competing with them. A store listed inside Play still has to acquire users, which is what the demand-side platform and SingleTap are for, and the company reports alternative-application partners ramping. There is also no measured evidence either way yet - the program took effect eight weeks before this thesis and nobody has published enrolment numbers, so the direction of the effect on Digital Turbine's revenue per device is an open question rather than a settled one. What is settled is that the remedy is being delivered inside Play, which is not the remedy the equity story assumes.
The rivals: GOOGL, APP, U, and 1 more that is private, in-house or listed abroad.
Who else does this, rival by rival
| Ticker | Rival | How far along | Priced against it |
|---|---|---|---|
| GOOGL | Google, through the Play Catalog Access programPriced against it: Different product | On sale | Different product |
| APP | AppLovin, through MAX and AxonPriced against it: One segment only | On sale | One segment only |
| U | Unity Software, through the ironSource stackPriced against it: One segment only | On sale | One segment only |
| Samsung and Motorola, and the carrier customers generallyPriced against it: Listed abroad | On sale | Listed abroad |
How big it could get
How big this could get
The market is Android phones shipped each year, and it is shrinking. IDC expects just over 1 billion smartphones in 2026, down 16.7%, the steepest fall on record. Android falls 24.3%, and the poorer countries where this company is growing drop over 20%.
The evidence under this - 2 claims
Four things argue against taking these at face value. None has a filed home, so none can be checked against a statement - which is the point of recording them as management claims. The one that carries most weight in the argument, international revenue for each device up more than 40%, is given as a rate with no level beneath it, so it cannot be reconciled to the $109,996 thousand of segment revenue that is filed. The 'nearly 3 billion devices' is cumulative rather than a period figure and is not comparable to IDC's annual shipment count. And this management's numbers had to be corrected once the day after this call. Against that: the figures that CAN be checked reconcile - segment revenue, segment profit, the leverage arithmetic and revenue for each employee all agree with the filings and the headcount series - so the record on the checkable ones is clean.
Who has signed
Almost nothing is committed. The biggest new win, the network Orange in May 2026, gives no number of phones, no minimum payment and no length, and no agreement has been filed.
The evidence under this - 1 claim
The results
The numbers
Sales grew 19% over the twelve months to Jun 2026, about the same pace as the 15% in the year to Mar 2026.
Operating margin is 10%, up from −69% in the year to Mar 2024.
Running costs take 47¢ of each sales dollar, up from 38¢ in the year to Mar 2022.
Shareholders' money lost 20% a year, down from 11% in the year to Mar 2022.
The business brought in $50.9M of cash from operations, 8.5% of sales, before capital spending, which the filer does not tag.
It owes $310M more than the cash it holds, about 6.1 years of the cash its operations bring in.
The share count rose 18% in 4.3 years, and pay in shares costs 2.1% of sales.
All the numbers
Filed figures from the company's 10-K and 10-Q, read 2026-10-04. Fiscal years end in March.
| Mar 2022 | Mar 2023 | Mar 2024 | Mar 2025 | Mar 2026 | 12 mo to Jun 2026 | |
|---|---|---|---|---|---|---|
| Size and growth | ||||||
| Sales | $748M | $666M | $544M | $491M | $565M | $600M |
| Sales growth | +138% | −11% | −18% | −9.9% | +15% | +19% |
| Profit | ||||||
| Gross margin | 50% | 54% | 52% | 52% | 57% | 57% |
| Operating margin | 12% | 6.9% | −69% | −11% | 6.0% | 10% |
| Net margin | 4.8% | 2.5% | −77% | −19% | −6.7% | −5.8% |
| What growth costs, per sales dollar | ||||||
| Research and development | 7.1% | 8.5% | 9.9% | 8.0% | 7.2% | 6.8% |
| Sales and marketing | 8.5% | 9.5% | 11% | 13% | 10% | 10.0% |
| General and admin | 19% | 23% | 31% | 35% | 25% | 22% |
| Pay in shares | 2.6% | 4.6% | 6.2% | 6.8% | 2.9% | 2.1% |
| Capital spending | - | - | 4.5% | 5.6% | 5.4% | - |
| Cash | ||||||
| Cash from operations | $84.7M | $113M | $28.7M | $11.9M | $41.8M | $50.9M |
| Free cash flow | - | - | $4.40M | −$15.6M | $11.2M | - |
| Balance sheet | ||||||
| Cash and short-term investments | $127M | $75.1M | $32.9M | $39.4M | $37.7M | $42.9M |
| Debt | $533M | $411M | $383M | $409M | $361M | $353M |
| Net cash | −$407M | −$335M | −$351M | −$369M | −$323M | −$310M |
| Returns and shares | ||||||
| Return on capital | 12% | 3.4% | - | - | - | - |
| Shares (diluted) | 103M | 102M | 101M | 104M | 113M | 121M |
What growth costs
Over the last two years each sales dollar has cost less to win: office and admin costs, product work, marketing and pay in shares all shrank as a share of sales. Much of that came from cutting staff, which does not repeat, and the share paid to phone networks rises with every sale.
The evidence under this - 3 claims
The improvement is from cutting costs as much as from growth: general and administrative dollars fell from $173,647 thousand to $142,124 thousand after two reductions in force (C32), a saving that does not repeat. Against fiscal 2021, before the 2021 acquisitions, every one of these lines is still a larger share of sales.
The last report (4 Aug)
Quarter ended June 2026, filed figures.
On the surface, the first quarter of fiscal 2027, to June 2026, was the turn investors wanted: sales up 26.8%, and $23.1M of operating profit against a loss a year earlier. But 58% of the growth came from the auction third, the half with no protected position.
Below the operating line it still lost money. Interest took $12.9M and the quarter ended $11.3M in the red. The $42.5M profit figure the company leads with gets there by adding back 4.7 times the loss, the interest included.
What it did to the thesisIt showed the turnaround is real at the operating level but not yet for shareholders: interest still pushed the company into a loss, and its headline profit figure adds back far more than that loss. Most growth came from the less protected auction business, supporting the Unattractive view.
The quarter in detail
Sales growth from the previous quarter, filed figures.
Earnings per share $0.19 against $0.14 expected; $-0.09 as filed, the gap being what the company adds back.
The evidence under this - 4 claims
Guidance is a floor a management team chooses, not a forecast it believes, and this one has already been raised once in the same fiscal year - so the honest reading is that the company is guiding conservatively after one strong quarter rather than signalling a slowdown it can see. The comparison base also hardens through the year: the fiscal 2026 quarters the guide is measured against grew 18.2%, 12.4% and 19.6%, against a first quarter that grew nothing measurable because the year-ago figure is not given, so some deceleration in the growth RATE is arithmetic rather than deterioration. What survives both points is that the company's own published range does not extrapolate the June quarter, and the price has.
The mix reading cuts against the thesis as well as for it. App Growth Platform earns 79.6% of its revenue after revenue share against On Device Solutions' 45.3%, so the shift toward it raised the company's non-GAAP gross margin from 47.3% to 49.4% - real operating leverage, not an accounting effect. And the chief executive's account of the brand business, up over 70%, describes demand won from holding companies and named advertisers rather than from adding cheap supply. The concern is what the exchange half is: publisher supply added in Asia Pacific and China is the part of advertising technology with the lowest barriers, and the filing itself names onboarding rather than pricing as the driver.
What management did with the money
It issued 13.5% more shares in fiscal 2026, and the money went to repay the lender rather than into the business. The finance chief resigned in May, and the new bonus targets are measured before interest.
Insiders in 2026Q2: 1 purchases and 6 sales, from their Form 4 filings.
The evidence under this - 4 claims
The dilution has stopped for now and the reason it happened is defensible. The At-the-Market programme was terminated, shares outstanding rose only 0.5% in the June quarter, and what the equity bought was a cut in gross debt from $430,000 thousand borrowed to $387,800 thousand outstanding and a fall in net leverage management puts at more than five turns to 2.5 - a genuine transfer of value from lender to shareholder at the cost of ownership. Stock-based compensation is also modest at $2,448 thousand in the quarter, well below the growth rate. The open item is the 10,630,000-share pool increase approved in August and the refinancing the 10-Q names as the principal cash requirement of the coming year, neither of which has yet had its per-share cost set.
The price, and where we disagree
In short. Full. The price pays for a turnaround that has to run much further than Digital Turbine's own forecast, and its lenders are paid before shareholders see anything.
The shares went from $2.88 in March to $11.11 in September. 84% of the last six months' rise was buyers paying more for each dollar of sales, not sales growing.
The test comes on 3 November with the September-quarter results. Even the raised forecast of $650M to $670M for the year implies growth of only 11.4% to 16.0% over the remaining nine months, against 26.8% in June.
The evidence under this - 3 claims
Guidance is a floor a management team chooses, not a forecast it believes, and this one has already been raised once in the same fiscal year - so the honest reading is that the company is guiding conservatively after one strong quarter rather than signalling a slowdown it can see. The comparison base also hardens through the year: the fiscal 2026 quarters the guide is measured against grew 18.2%, 12.4% and 19.6%, against a first quarter that grew nothing measurable because the year-ago figure is not given, so some deceleration in the growth RATE is arithmetic rather than deterioration. What survives both points is that the company's own published range does not extrapolate the June quarter, and the price has.
How we read the multiple
Buyers pay about 27 times a year's operating profit for the whole company, its debt included. Before shareholders see any of that profit, $46M a year goes on interest.
For the price to make sense, operating profit has to roughly double, while the company's own forecast has its growth roughly halving from here.
Its $387.8M of loans, due in 2029 at 11.2% and 11.6% interest, also have to be replaced far more cheaply. They have not been.
Against its own history. Far dearer than this company's own normal for each dollar of sales. The business is better than it was, but it still loses money after interest.
The evidence under this - 4 claims
Two readings push back. On enterprise value against EBITDA the frozen pack puts the shares at 18.8 times against a six-year median of 19.2 - cheaper than their own history, not dearer - and on the guided $150 million of adjusted EBITDA the multiple is about 11 times. And operating profit is depressed by acquisition intangible amortisation: $208,485 thousand of intangibles remained at 30 June 2026 and $16,805 thousand of depreciation and amortisation ran through the June quarter alone, so the operating line carries the cost of businesses bought years ago and already impaired. The counter to the counter is that both of those are the point: EBITDA is the metric that removes the interest, and amortisation of what was acquired is exactly what a serial acquirer's valuation frame must keep in.
Guidance is a floor a management team chooses, not a forecast it believes, and this one has already been raised once in the same fiscal year - so the honest reading is that the company is guiding conservatively after one strong quarter rather than signalling a slowdown it can see. The comparison base also hardens through the year: the fiscal 2026 quarters the guide is measured against grew 18.2%, 12.4% and 19.6%, against a first quarter that grew nothing measurable because the year-ago figure is not given, so some deceleration in the growth RATE is arithmetic rather than deterioration. What survives both points is that the company's own published range does not extrapolate the June quarter, and the price has.
Priced against its rivals
None of the listed rivals above sells the same thing to the same buyers as a large part of what it does. A price comparison needs at least two, so the price is judged against its own history instead.
Where we disagree with the market
The listed way to own Android opening up to rival app stores, with this year's growth proving the turn has arrived.
The court remedy is arriving inside Google Play, which removes the same difficulty this company's product sells against.
The market has bought this story before: in March 2021 the shares reached $102.56 on the idea that app stores would be forced open. Sales then fell three years running, and by November 2024 the shares were $1.18.
The case against us is real. This time the growth is the company's own rather than bought, costs have been cut, and the remedy has actually taken effect. Two more quarters of sales per phone growing above 40% would prove us wrong.
What analysts say, and the evidence
The evidence under this - 5 claims
The chief executive's reading is the opposite and it is not unreasonable: more app stores means more parties needing distribution and targeting, and Digital Turbine sells to them as customers rather than competing with them. A store listed inside Play still has to acquire users, which is what the demand-side platform and SingleTap are for, and the company reports alternative-application partners ramping. There is also no measured evidence either way yet - the program took effect eight weeks before this thesis and nobody has published enrolment numbers, so the direction of the effect on Digital Turbine's revenue per device is an open question rather than a settled one. What is settled is that the remedy is being delivered inside Play, which is not the remedy the equity story assumes.
Four things argue against taking these at face value. None has a filed home, so none can be checked against a statement - which is the point of recording them as management claims. The one that carries most weight in the argument, international revenue for each device up more than 40%, is given as a rate with no level beneath it, so it cannot be reconciled to the $109,996 thousand of segment revenue that is filed. The 'nearly 3 billion devices' is cumulative rather than a period figure and is not comparable to IDC's annual shipment count. And this management's numbers had to be corrected once the day after this call. Against that: the figures that CAN be checked reconcile - segment revenue, segment profit, the leverage arithmetic and revenue for each employee all agree with the filings and the headcount series - so the record on the checkable ones is clean.
What could go wrong, and how we would know
The risks
- The lender comes first. Interest ran 1.72 times operating profit last year, and from this year half of any spare cash goes to the lender by contract.
- A shrinking phone market. With Android shipments falling a quarter, every new phone this company reaches has to be won from someone else.
- Networks building their own. Its annual report says the bigger competition is networks and handset makers writing this software themselves.
- Shaky reporting. The June-quarter loss was announced at $3.2M and corrected to $11.3M the next day, with an interim finance chief.
The evidence under this - 7 claims
Share taken from a shrinking base is still revenue, and there is a real mechanism behind it: the Orange agreement, the Motorola and Samsung volumes management named, and the chief executive's point that a higher average selling price is what funds the advertising inventory in the first place. Rising device prices can raise revenue per device even as units fall, which is exactly what the company reports. Two things blunt the counter. IDC has iOS gaining share as Android falls, and iOS is a base Digital Turbine cannot enter at all. And share gains are finite and lumpy in a way a growing market is not: there is a fixed number of large carriers and handset makers, each is won once, and the company does not disclose how many it has or how many are left.
Three things argue the other way. The chief financial officer flagged the adjustment on the call itself rather than leaving it to be discovered, so the disclosure was voluntary and same-day. The whole item is non-cash and touched neither the adjusted EBITDA of $42.5 million the market actually trades this company on nor, the 8-K/A states, the reported cash flows or compliance with debt covenants. And the correction the amendment describes - an $8,173 thousand liability wrongly assumed in a prior business combination - makes the balance sheet better rather than worse, since it reduced liabilities and raised stockholders' equity. What it does not explain away is that the GAAP headline the market saw on print day was understated by a factor of 3.6 and that no analyst on the call asked about it.
What we still do not know
- What one phone earns the company in a year. It reports how fast that grows, never the amount.
- How many rival app stores have joined Google's program since July. Nobody has published it.
- When, and on what terms, the 2029 loans are replaced.
The evidence under this - 3 claims
Four things argue against taking these at face value. None has a filed home, so none can be checked against a statement - which is the point of recording them as management claims. The one that carries most weight in the argument, international revenue for each device up more than 40%, is given as a rate with no level beneath it, so it cannot be reconciled to the $109,996 thousand of segment revenue that is filed. The 'nearly 3 billion devices' is cumulative rather than a period figure and is not comparable to IDC's annual shipment count. And this management's numbers had to be corrected once the day after this call. Against that: the figures that CAN be checked reconcile - segment revenue, segment profit, the leverage arithmetic and revenue for each employee all agree with the filings and the headcount series - so the record on the checkable ones is clean.
What would change our view
Leverage into a shrinking base is the combination with no self-correcting mechanism. The preload franchise is real, but its moat is distribution through carriers and handset makers - a position its own annual report says faces its more material competition from carriers and manufacturers building that software themselves (C34). A refinancing would help the arithmetic and would not touch the unit base, and the unit base is what the franchise is a share of. This was the most marginal of the three archived and is the one most likely to be reversed.
- Reopens the researchThe debt goes and the unit base stops shrinking
It needs all of:
- ○The 2029 loans are repaid or replaced at a much lower interest cost, so operating profit reaches shareholders.watched as it happens
- ○Sales from the phone software grow year on year for two quarters running, while global phone shipments stop falling.checked at the next results, due by 3 Nov
We would also be wrong if
- Sales per phone keep growing above 40% for two more quarters, faster than Android shipments fall.
- It starts reporting sales from rival app stores as a line of its own, and that line grows.
- Sales growth for the rest of the year to March 2027 stays above the 16.0% the top of the forecast implies.
What is coming
- 3 November 2026 Q2 fiscal 2027 results: Whether revenue for each device keeps growing near 40%, whether the full-year range is raised again, and whether either term loan is refinanced cheaper.
Every dated event ahead
- 3 November 2026 Q2 fiscal 2027 results: Whether revenue for each device keeps growing near 40%, whether the full-year range is raised again, and whether either term loan is refinanced cheaper.
- December 2026 Orange rollout in Europe begins: Whether the largest new distribution agreement gains a disclosed device count, volume or revenue, or stays a narrative without terms.
- 31 December 2026 Minimum-liquidity covenant relief and fee waiver window end: Whether the company prepays the specified tranche to avoid duration fees and meets the $20M liquidity floor again from January, or needs another paid amendment.
- February 2027 Q3 fiscal 2027 results, the seasonally strongest quarter: Whether the second half grows faster than the 11.4% to 16.0% guidance implies, and whether a permanent finance chief has been named.
Since we wrote this
- 5 OctThesis reviewThe long case now sits under the verdictAfter the five-name pilot, the owner asked for the long case on every thesis.
- 4 OctThesis reviewOwner's checklist: revenue streams, cost of growth, management, catalystsThe owner's checklist found the method silent on who runs the company, what growth costs, where the money comes from and what is dated ahead; this pass fills those four blocks from the proxy and filings.
- 26 SepThesis reviewWhat would prove us wrong now points up. View heldEvery test of the view was a way the bull case fails, which could only confirm a negative view. They now name what would show it wrong: faster sales per phone, a cheaper loan, profit that doubles.
- 16 SepThesis writtenUnattractive, medium convictionAfter the shares more than tripled in six months on the Google app-store ruling, with interest still larger than operating profit.
Also checked 1 time with nothing new to act on.
Sources 19
Sources - 19
Every claim above cites one of these, or states the reasoning that derived it. The quote is frozen at the date it was read: a link can rot, and an agent fetch returns a reading of a page rather than the page. Backs counts the whole claim ledger, so it can name a claim this page does not surface.
- S1primarycompany filing2026-05-26Item 1: the two reportable segments and what sits in each - Application Media and Content Media inside On Device Solutions, Advertising Solutions and Ad Monetization inside App Growth Platform - the Competition section naming Google Play, Facebook, Snapchat, Unity Software, InMobi, Magnite, AppLovin, The Trade Desk, Adobe and Liftoff, and $40,476 thousand of product development cost. Item 1A: the transformation programme completed in the fourth quarter of fiscal 2025, the carrier-concentration risk factor, the indebtedness risk factors and the $336,640 thousand goodwill impairment recorded in fiscal 2024. Item 7: the fiscal 2026 income statement line by line - net revenue $565,251, revenue share $243,638, income from operations $34,042, interest expense net $58,580, loss on extinguishment of debt $9,795, loss before income taxes $31,340, income tax expense $6,392 - unrestricted cash of $37,719, and the Financing Agreement terms. Item 8: the balance sheet with 120,315,203 shares outstanding against 105,977,642 a year earlier, the revenue-recognition principal-versus-agent policy, and the statement that no single supply partner or customer exceeded 10% of net revenue in fiscal 2026, 2025 or 2024. Item 9B: the resignation of Chief Financial Officer Steve Lasher and the appointment of Chief Accounting Officer Josh Kinsell as interim Chief Financial Officer.backsC01C04C07C08C11C17C20C21C22C23C24C25C28C30C32C33C34C35C36C37C38C41
- S10primaryindustry data2026-07-22The attacker's own account of what the Epic injunction actually requires it to do: third-party United States Android app stores could sign up for the Play Catalog Access program from 22 June 2026, onboarding opened 15 July 2026, and the program took effect on 22 July 2026 - so a rival store is now distributed from inside Google Play rather than around it. Also the 1 October 2026 date from which developers in the external-offers and alternative-billing programmes must report transactions and pay service fees.
On June 22, 2026, we notified developers that unless they are opted-out by July 22, 2026, Google Play will begin to provide their app listing(s) to third-party US Android app stores. … Additionally, third-party US Android app stores can now sign-up for the Play Catalog Access program. … On July 15, 2026, interested third-party US Android app distribution platforms and app stores can begin onboarding into the Third-party App Store on Play Program. … As of July 22, 2026, this program is now in effect.
backsC16C17 - S11primarycompany release2026-05-19The largest new distribution agreement in the story management tells, and what it does not say: Orange will use the SingleTap technology and alternative app distribution platform across Europe, with rollout from the second half of 2026. No contract volume, no term, no revenue commitment and no device count is disclosed.
AUSTIN, Texas, May 19, 2026 /PRNewswire/ … With rollout expected to begin in H2 2026 and gradually expand across devices and countries in Europe, Orange subscribers across Europe will benefit from customized user journeys on their mobile devices, enabling new app discovery opportunities tailored to their needs and preferences.
backsC18 - S12primarycompany filing2026-02-19AppLovin Corporation Form 10-K for the fiscal year ended December 31, 2025 AppLovin CorporationThe scale of the company Digital Turbine's own annual report names as a competitor in both segments, read in that company's filing rather than in Digital Turbine's: MAX, the in-app bidding product that competes directly for the publisher inventory the DT Exchange is winning, and the Axon recommendation engine behind it. Searched for any mention of on-device, preload or carrier distribution and there is none in the document, which answers whether the largest named rival competes in On Device Solutions.
MAX is our in-app bidding solution that optimizes the value of publishers' advertising inventory by running a single unbiased, real-time competitive auction … Revenue | $ | 5,480,717 … Income from operations | 4,151,914 | 1,910,956 | 772,411
backsC20 - S13primaryvendor data2026-09-16Frozen research pack, content/APPS/research-pack.jsoncontent/APPS/research-pack.jsonPrice, market value of $1,343,598,960 and enterprise value of $1,653,255,960 at 16 September 2026; the 20-day average daily range of 5.71%; the decomposition of the price move into the part the business did and the part the market did; the six-year medians of every valuation ratio; the FINRA short-interest settlement; the analyst-consensus series and its disagreement with the snapshot; and the vendor-against-filings cross-check, which agrees to 0.0% on both revenue and operating income.backsC23C24C25C26C27C32C35
- S14secondaryvendor data2026-09-16The peer trajectory the pressure test needed: Unity's revenue rising from $1.391 billion in 2022 to $2.187 billion in 2023 on the ironSource merger and then falling to $1.813 billion in 2024 and $1.850 billion in 2025, against operating losses of $832.8 million, $755.1 million and $422.6 million. AppLovin's revenue of $5.481 billion and operating income of $4.152 billion in 2025. Magnite's revenue of $713.9 million and operating income of $97.8 million in 2025.read, but no claim rests on it
- S15secondaryvendor data2026-09-16The record of the last time this market held this view: a high of $102.56 on 2 March 2021, a low of $1.18 on 19 November 2024, a March 2026 fiscal year-end close of $2.88 and $11.11 at 16 September 2026. The 1,432 sessions the frozen pack's 759 bars cannot reach.backsC22C23
- S16primarycompany filing2026-02-11Unity Software Inc. Form 10-K for the fiscal year ended December 31, 2025 Unity Software Inc.The filed revenue of the peer that ran this cycle first, read in its own annual report rather than in a vendor re-key: total revenue of $1,849,648 thousand in 2025, $1,813,255 thousand in 2024 and $2,187,317 thousand in 2023. It also splits the advertising half out - Grow Solutions at $1,228,239 thousand in 2025 against $1,328,143 thousand in 2023 - so the part of Unity that matches Digital Turbine's App Growth Platform is measurably smaller two years after the merger that built it.
Total revenue | $ | 1,849,648 | $ | 1,813,255 | $ | 2,187,317
backsC21 - S17primarycompany filing2026-07-13Beneficial ownership at the 1 July 2026 record date, the executive officers and their start dates, the director biography of William G. Stone III, the fiscal 2026 annual incentive design and payouts, the long-term incentive design and the pay-versus-performance table.backsC38C39C40C41
- S18primarycompany filing2026-09-24Michael Akkerman, Chief Business Officer, resigned to take a chief executive role elsewhere, effective 30 October 2026; no successor named and no agreement filed.backsC38
- S19primarycompany filing2026-10-01A universal shelf registration on Form S-3ASR filed 1 October 2026, and a resale prospectus supplement covering 1,222,418 shares issuable on exercise of the warrants issued to the lenders under the Financing Agreement of 29 August 2025; the company receives no proceeds from the resale.backsC41
- S2primarycompany filing2026-08-05The filed first-quarter statements: net revenue $165,983, income from operations $23,061, interest expense net $12,890, unrealized loss on derivatives $10,799, other expense net $9,094, net loss $11,329 and diluted loss per share of $0.09. Note 9 Debt: Term loan A $285,000 at an 11.2% effective rate and Term loan B $102,800 at 11.6%, both due August 2029, the April 2026 amendment cutting the liquidity covenant from $20,000 to $15,000 for a $5,000 fee, and the mandatory prepayment provisions. Note 1: the revision of prior period financial statements for an immaterial error of $8,173 in liabilities assumed in a prior business combination. The segment note giving ODS revenue of $109,996 and segment profit of $49,865 against AGP revenue of $56,596 and segment profit of $45,070. The cash flow statement: $17,859 of operating cash flow, $11,585 of interest paid, $7,506 of income taxes paid, $16,805 of depreciation and amortisation and $2,448 of stock-based compensation. Unrestricted cash of $42,930, and the statement that the principal cash requirements for the next twelve months consist primarily of refinancing certain loan tranches.backsC02C03C05C06C07C08C09C10C11C13C15C23C24C25C31C33C34C35C37
- S3primarycompany release2026-08-04The figures the market traded on the day of the print: a headline reading "First Quarter GAAP Net Loss of $3.2 Million and GAAP EPS of $(0.03)", an income statement showing other expense, net of $(921), loss before income taxes of $(868) and a net loss of $(3,156), and the raised fiscal 2027 outlook of $650 million to $670 million of revenue and $145 million to $155 million of non-GAAP adjusted EBITDA.backsC09C12C13
- S4primarycompany release2026-08-05The explanatory note stating that the sole purpose of the amendment is to correct the original earnings release, and the amended income statement restating other expense, net to $(9,094), loss before income taxes to $(9,041) and the net loss to $(11,329) or $(0.09) a share. The adjusted EBITDA reconciliation, unchanged at $42.5 million, showing the full add-back list: stock-based compensation $2,448, depreciation and amortisation $16,805, interest expense $12,890, other expense $9,094, unrealized loss on derivatives $10,799, income tax expense $2,288, severance $156 and foreign exchange $(681).backsC09C10
- S5primarycompany release2026-05-26Fiscal 2026 full-year figures: revenue $565.3 million, GAAP net loss $37.7 million or $(0.33) a share, non-GAAP adjusted net income $64.9 million or $0.56 a share, and non-GAAP adjusted EBITDA of $122.5 million against $72.3 million in fiscal 2025. The fourth-quarter split - On Device Solutions $91.0 million up 5%, App Growth Platform $52.1 million up 57% - and the first fiscal 2027 outlook, of $630 million to $650 million of revenue and $135 million to $145 million of adjusted EBITDA.backsC12C13
- S6primarytranscript2026-08-04The operating metrics that appear in no filing: international On Device Solutions revenue up nearly 80% year over year, global devices up double digits and revenue per device up more than 40%, direct brand advertising up over 70% and the DT Exchange up over 40%, net leverage of 2.5 turns against more than 5 a year earlier, 80% of DT Exchange advertiser spend from non-gaming partners, nearly 3 billion devices and more than 80,000 apps carrying the technology, and run-rate revenue per employee above $1 million against about $800,000 a year earlier. Also the chief financial officer's statement of a GAAP net loss of $3.2 million and his warning that a non-cash adjustment was still being finalised, the chief executive's reading of the Epic-Google outcome as a tailwind, non-GAAP gross margin of 49.4% against 47.3%, and the whole question section: two analysts, eight turns, no question on the debt, the interest bill, the warrant liability or the pending adjustment.backsC03C09C15C17C19C27C30
- S7primarycompany filing2026-08-25Stockholder approval of a fourth amendment to the 2020 Equity Incentive Plan increasing the authorised pool by 10,630,000 shares from 20,560,000 to 31,190,000, and the say-on-pay vote of 37,927,832 for against 26,468,367 against with 337,470 abstaining. Also 120,936,038 common shares outstanding at the 1 July 2026 record date.backsC08C28
- S8primarycompany filing2026-09-08The 2 September 2026 grants to the four named executive officers, half in performance stock units and half in time-based units, and the performance conditions: four equally weighted goals, two of them adjusted EBITDA for fiscal 2028 and fiscal 2029 and two of them relative total shareholder return against the S&P Software & Services Select Industry Index.backsC29C40
- S9primaryindustry data2026-08-26IDC, "Smartphone Shipments Set for Record 16.7% Drop in 2026, as the Memory Crisis Hits Full Force" International Data CorporationThe measured size and direction of the device base On Device Solutions is sold onto, from the firm that counts the shipments rather than from the company that monetises them: a 16.7% decline in 2026 to just over 1 billion units, with Android bearing almost the entire decline at 24.3%, emerging markets expected to drop over 20%, iOS share rising almost four points to a record 23.6%, and the average selling price up 27.6% to $581.
worldwide smartphone shipments will fall 16.7% in 2026 to just over 1 billion units … Android bears almost the entire decline, falling 24.3% in 2026 … iOS share moves in the opposite direction, increasing almost four percentage points from last year to a record-high 23.6% share … The mass market does not hold up, especially in emerging markets, which are expected to drop over 20% this year … The average selling price of a smartphone will reach $581 in 2026, up 27.6% in a single year
backsC14C15