Byline Bancorp Stock: A Record $130 Million — and a Government Guarantee With a Clause

Every shop you have ever walked into sells something you can look at. You can squeeze the avocado, kick the tire, hold the handbag up to the light. A bank is the one shop where the goods are invisible. What Byline Bank sells is a promise written in a spreadsheet: someone in Chicago borrowed money and says they will pay it back. You cannot squeeze that. So what do you do instead? You read the labels. "Record profit." "Government guaranteed." "Well capitalized." And here is the quiet trap — psychologists would file it under processing fluency, but you can call it the label trap: a label that is easy to read feels like a fact you have checked. It is not. A label tells you what a thing is called, not what it is worth. Byline Bancorp (NYSE: BY) wears three very good labels right now, and all three are honestly earned — which is exactly what makes them worth turning over. Almost nobody is looking: our Reddit hype scanner counted 3 mentions in 24 hours (ApeWisdom, as of July 16, 2026). This stock has no forum, no meme, no drum. So let us make a deal: we ignore the labels and read only what Byline told the U.S. securities regulator, the SEC, under penalty of law — the annual report (10-K) for 2025, the quarterly report (10-Q) as of March 31, 2026 and the current report (8-K) of July 23, 2026 carrying the second-quarter results. At the end you will not get a verdict from us. You will get the findings.

Updated on August 2, 2026: This analysis now carries a dedicated chapter on the earnings calls — "What the Earnings Calls Reveal", placed just before the conclusion. We read all ten quarterly call transcripts from 2024-Q1 through 2026-Q2 and checked what management promised one or two quarters ahead against what actually arrived. The result: on the numbers that sit at the centre of every presentation the bank delivers with unusual reliability — net interest income guidance was met or beaten in all nine testable quarters. Alongside that, however, sit several previously stated targets that vanished from the communication between 2024 and 2026 without ever being reconciled, and on the single most important strategic question, crossing ten billion dollars, the stated position reversed within six months. None of the earlier findings in this analysis is weakened by the calls.

What Byline actually does — a corner bank with a federal side door

Byline Bank is a commercial bank in Chicago: 45 branches (44 across the Chicago metropolitan area, one in Wauwatosa, Wisconsin), $9.65 billion of total assets, $7.65 billion of deposits and $7.51 billion of loans as of December 31, 2025. The everyday machine is the oldest one in finance: take deposits cheaply, lend them out dearly, keep the difference. That difference is called the net interest margin, and it is the number to watch — of every $100 of loans and securities on the books, Byline kept $4.22 as margin in 2025. If that sounds unremarkable, it is not: most regional banks live around $3.00, and the gap is the whole story of this company.

Where does the extra margin come from? Through a side door marked Small Business Administration. Byline is one of the most active SBA lenders in the country — an "SBA Preferred Lender", which means it can approve federally guaranteed small-business loans itself instead of queuing for Washington. The mechanics are worth understanding in plain language, because the rest of this analysis hangs on them. A small business borrows, say, $1 million. The federal government guarantees roughly three quarters of it. Byline then sells that guaranteed slice to investors at a premium, keeps the unguaranteed rest on its own books, and keeps servicing the whole loan for a fee. The annual report puts it in one sentence:

"We generally sell the government guaranteed portion of SBA and USDA loans into the secondary market while retaining the non-guaranteed portion of the loan and the servicing rights. This allows us to realize one time gain on sale income along with a recurring servicing and interest revenue stream."

— Byline Bancorp, Inc., SEC annual report 10-K 2025, Item 1 "Business" (Small Business Capital)

Three revenue streams from one loan, in other words. As of December 31, 2025 the bank held $533.0 million of SBA and USDA loans, of which only $97.8 million were still government guaranteed — and it was servicing another $1.6 billion of sold loans for other people. Alongside this sits the rest of an ordinary commercial bank: $1.4 billion of commercial real estate, a leasing arm, and — less ordinary — $805.9 million of "sponsor finance", senior secured loans to private-equity-owned companies with EBITDA between $2 and $10 million. Growth has come as much by takeover as by sweat: five bank acquisitions since 2016, most recently First Security on April 1, 2025. Which brings us to the central tension of this analysis, and it runs through every chapter: the two things that make Byline exceptional — a federally powered lending engine and an appetite for buying other banks — are precisely the two things that make its earnings hardest to read and its returns quietly shrink.

Where the stock shows up in our scanner — one hit, and an honest admission

Every day we run about 3,500 stocks through our scanners. Byline has a row in our database — its company profile sits in the stocks section — and as of the July 16, 2026 data cut-off it lit up in exactly one filter: the KCF ranking, which collects companies with positive operating cash flow trading at no more than ten times it. Byline sits at 9.9, near the edge of the gate, one of 1,326 names on a very broad list. That is the entire harvest.

Now the honest part, and we would rather say it than dress it up: our scanner is built for industrial companies, and a bank breaks it. On Byline's row, the price-to-earnings ratio, the price-to-book ratio and the debt-to-equity ratio all come back empty — no value at all. Some of that is a plain data gap on this row (peer banks in our database do carry those numbers). But the deeper problem is that for a bank those metrics are noise even when they are populated. Debt-to-equity is the clearest example: a bank's "debt" is mostly your savings account. Deposits are the raw material, not a warning sign — which is why the same filter reports 0.59 for Wintrust and 4.12 for JPMorgan — two healthy banks, sevenfold apart — and why the price-to-cash-flow that put Byline on our one list reads 6.4 for JPMorgan and 219.2 for Wells Fargo. Same industry, same quarter, numbers that cannot both mean something. A screen that ranks a bank by EV/EBITDA is measuring a fish by how well it climbs.

So we will not pretend the scanner has an opinion here. What it can measure honestly is price behavior, and there the reading is consistent: a relative-strength rating of 73, up about 45 percent over twelve months and 27 percent over six, trading above its 50-day and 200-day averages with an intact power trend, and within about 0.3 percent of its 52-week high (all data as of July 16, 2026). One more clean signal: 4 insider purchases against 2 sales in the past twelve months — the rarer direction, and the opposite of what we usually find. Everything else in this analysis has to come from the filings. Which is where we are going.

The numbers over the years — honestly appraised

Start with what genuinely impresses, because there is a lot of it. Net income has risen every single year: $107.9 million (2023) → $120.8 million (2024) → $130.1 million (2025), the last a record, up 7.7 percent, or $2.89 per diluted share. Revenue reached $446.2 million. The margin went the right way — the net interest margin climbed from 3.97 to 4.22 percent — and the cost machine got tighter: the efficiency ratio (what the bank spends to earn a dollar; below 55 percent is good, below 50 is excellent) improved to 51.83 percent. Credit, the thing that kills banks, simply refused to crack: non-performing loans held at 0.95 percent of the book (2024: 0.90), net charge-offs actually fell to 0.39 percent, and reserves cover 1.45 percent of loans. Return on assets of 1.36 percent is genuinely good — anything above 1.00 percent marks a bank that earns its keep. And 2026 has been better still: net income of $37.6 million in the first quarter, up 33 percent, with the efficiency ratio breaking below 50 for the first time (49.78 percent), followed by $40.2 million in the second.

Now turn the page over. Here is the number nobody puts in a press release:

Return on average tangible common equity — the honest yardstick for a bank, because it asks what the owners' real money earned — has gone the wrong way for two straight years: 16.46 percent (2023) → 14.85 percent (2024) → 13.47 percent (2025). Return on equity slid the same way, 12.50 → 11.61 → 10.86 percent. Read that next to the rising profit and you have the whole puzzle of this stock in two lines. The explanation is not a scandal; it is arithmetic. Byline is stockpiling capital: the CET1 ratio (the regulator's core safety cushion) climbed from 10.35 percent (2023) to 12.33 percent (2025), and tangible common equity grew from 9.06 to 11.29 percent of assets. Profit grew about 21 percent over two years; the equity it sits on grew faster. Remember the mechanism: a bank that hoards capital gets safer and less profitable at the same time — and the second half of that sentence shows up in the ratio long before anyone explains the first half. Why hoard? Because capital is ammunition for the next acquisition. Which is a strategy, not an accident — but you are the one paying for the ammunition while it sits in the magazine.

Then came 2026. The first quarter was the first evidence the drag might be ending; the second confirmed it. In the first quarter Byline earned $37.6 million (up 33 percent), the efficiency ratio broke below 50 for the first time (49.78 percent) and ROTCE climbed back to 13.77 percent from 12.92 percent a year earlier. On July 23, 2026 the bank filed a current report (8-K) with the second quarter: net income of $40.2 million, $0.90 per diluted share, revenue of $117.7 million, an efficiency ratio of 46.93 percent (adjusted 46.51 percent — the best since the company went public) and a return on average tangible common equity of 14.47 percent. The board raised the quarterly dividend on July 21, 2026 by 16.7 percent to $0.14 per share, payable August 18, 2026; the company also repurchased 274,528 of its own shares during the quarter, for a total payout ratio of 35.9 percent. So the return has turned — and yet the capital cushion still grew, to a CET1 ratio of 12.92 percent. The magazine is getting fuller, not emptier.

What the filings say — the uncomfortable truths

Uncomfortable truth no. 1: the government guarantee has a clause — and Byline holds no reserve against it

"Government guaranteed" is the most reassuring label in this whole business, and it is the one most worth turning over. The guarantee is real: when an SBA borrower defaults, Washington covers its share. But the guarantee is conditional on Byline's own paperwork — and if the SBA later decides the loan was written or serviced sloppily, it can come back for the money. The annual report says so in the flattest possible language, and then adds a sentence that should make you sit up:

"If the SBA establishes that a loss on an SBA guaranteed loan is attributable to significant technical deficiencies in the manner in which the loan was originated, funded, or serviced by us, the SBA may seek recovery of the principal loss related to the deficiency from us. Generally, we do not maintain reserves or loss allowances for such potential claims and any such claims could materially adversely affect our business, financial condition, or results of operations."

— Byline Bancorp, Inc., SEC annual report 10-K 2025, Item 1A "Risk Factors" (Guaranteed Loans Risks)

Sit with the second sentence. Not "we reserve conservatively for this"; not "we estimate the exposure at X". No reserve at all. That is a legitimate choice — such claims are rare, hard to estimate, and accounting rules do not force a number onto a risk you cannot measure. But it means the exposure is carried at zero on a balance sheet that services $1.6 billion of sold loans for others and holds $533.0 million more. And the dependency runs deeper than a clawback. The same risk factor spells out that Byline's whole SBA franchise sits on a permission slip: lose the Preferred Lender status and "we may lose some or all of our customers to lenders who are SBA Preferred Lenders"; and "any default by the U.S. government on its obligations or any prolonged government shutdown could impede our ability to originate SBA loans or other government guaranteed loans or sell such loans in the secondary market". Translated: Byline's best business has a silent partner it did not choose and cannot fire — one that writes the rules, sets the volume caps through Congress, and occasionally shuts down. What a dependency on a single counterparty does to an otherwise healthy business is something we took apart at EverQuote, where one insurance carrier pays 40 percent of revenue. Byline's version is less concentrated but harder to negotiate with.

Uncomfortable truth no. 2: the famous SBA premium is not actually where the money is

Now for the pleasant surprise — pleasant, but with a sting. Byline is known as an SBA machine, and the intuitive worry follows: those one-off gains from selling guaranteed loans must be propping up the profit, and they are cyclical, so the profit is fragile. It is a good hypothesis. The filings do not support it. Here is the actual split:

Of $446.2 million of 2025 revenue, $385.3 million — 86.3 percent — was net interest income: plain, recurring, spread. The famous gains on selling government guaranteed loans came to $22.7 million, about 5.1 percent of revenue. And they went down:

"Net gains on sales of loans were $22.7 million for the year ended December 31, 2025 compared to $24.5 million for the year ended December 31, 2024, a decrease of $1.8 million, or 7.4%. The decrease in net gains on sales was primarily driven by lower market premiums for U.S. government guaranteed loans."

— Byline Bancorp, Inc., SEC annual report 10-K 2025, Item 7 MD&A "Non-interest Income"

The premium shrank 7.4 percent and net income still rose 7.7 percent. So the cyclical-one-off worry is, on these numbers, largely wrong — and the share of income that comes from fees keeps falling, not rising: non-interest income was 14.55 percent of revenue in 2023, 13.65 percent in 2025, and just 11.15 percent in the first quarter of 2026 (it rebounded to 14.34 percent in the second quarter, when fair-value marks and a stronger sales quarter coincided — $6.1 million of gains on $78.1 million of guaranteed loans sold). But now the sting, and this is the part the SBA story gets right for the wrong reason: the SBA does not show up in the fee line, it shows up in the margin. Byline's loan book yielded 7.07 percent in 2025 — a startling number when a typical regional bank earns somewhere near 5.5 to 6 percent — because SBA 7(a) loans float against the prime rate at a fat spread. That is why the margin is 4.22 percent instead of 3.00. So the dependency is real, and it is bigger than the fee line suggests, not smaller — it is just wearing different clothes. Remember the image: you were looking for the government's money in the tip jar; it is in the wages.

Uncomfortable truth no. 3: five takeovers have left $181.9 million of goodwill — 14.3 percent of the equity is air

Byline grows by buying: Ridgestone (2016), First Evanston (2018), Oak Park River Forest (2019), Inland Bancorp (2023) and First Security on April 1, 2025. Every one of those deals was paid at a premium over the net assets acquired, and the premium has a name and a home on the balance sheet:

"As of December 31, 2025, we had goodwill of $181.9 million, or 14.3% of our total stockholders' equity."

— Byline Bancorp, Inc., SEC annual report 10-K 2025, Item 1A "Risk Factors" (goodwill impairment)

Goodwill is not a fraud and not a fiction — it is what you paid for a business above the sum of its parts, and often it is worth every cent. But it is the one asset that cannot absorb a loss, pay a depositor, or be sold on a Tuesday. It is a receipt, not a resource. Which is why banks are judged on tangible book value, and why the gap matters here: book value per share is $27.84, but tangible book value per share is only $23.44 (December 31, 2025). Roughly every sixth dollar of book value is a receipt from an old deal. Counting all intangibles, $200.5 million of the $1.27 billion of equity is untouchable in a crisis. There is a second, subtler cost: goodwill is tested for impairment every year, and the test leans on the company's own market value and discounted cash flows — so a long slump in the share price can force a non-cash writedown, in the middle of a downturn, precisely when nobody needs more bad news. What that looks like when it goes wrong we watched at Grocery Outlet, where $307 million of goodwill was written off in one stroke. Byline has shown no sign of it — credit is clean, the deals have been small and local, and management has integrated four of them without visible damage. But the honest framing is this: each acquisition converts hard capital into a promise that the acquired bank keeps performing. Five promises are now on the books, and the capital hoard from truth no. 2 is being built to make a sixth.

Uncomfortable truth no. 4: the record margin was partly rented from the Fed — and the funding is getting tighter

The margin went from 3.97 to 4.22 percent in a single year, and that leap deserves an explanation rather than applause. It did not come from charging borrowers more — the loan yield actually fell, from 7.40 to 7.07 percent. It came from paying depositors less: the average cost of deposits dropped from 2.61 to 2.17 percent after the Federal Reserve cut rates by 75 basis points over 2025. In plain terms: the central bank cut what Byline pays for its raw material faster than it cut what Byline charges for the finished product, and the bank pocketed the difference. That is skillful, and it is also borrowed. The report is unusually candid about how narrow the ridge is:

"Accordingly, we continue to cautiously manage our deposit pricing strategies to seek to maintain our net interest margin. As the competition for funding among banks remains high, and customers continue to seek higher yields, we have adjusted our deposit pricing accordingly. To the extent we offer higher interest rates on targeted interest-bearing deposit products to maintain current customers or attract new customers, our interest expense may increase, perhaps materially. Furthermore, if we fail to offer interest rates at a sufficient level to keep these demand deposits, our core deposits may be reduced, which would require us to obtain funding in other ways or risk slowing our future asset growth."

— Byline Bancorp, Inc., SEC annual report 10-K 2025, Item 1A "Risk Factors" (deposit pricing and interest rate risk)

Pay more, and the margin goes. Pay less, and the deposits go. Byline has been threading that needle well, but three numbers show the thread pulling tight. First, the cheapest money is draining away: non-interest-bearing deposits — accounts that cost nothing — have fallen from 26.56 percent of all deposits (2023) to 23.78 percent (2025), and to 23.21 percent by June 30, 2026. Second, the bank is fully lent: loans as a share of deposits climbed from 92.64 to 98.37 percent in a year (96.10 percent on June 30, 2026), meaning almost every deposited dollar is already out the door and further loan growth needs new funding rather than spare cash. Third, and worth knowing without panicking about it: estimated uninsured deposits were $2.7 billion at the end of 2025, up from $2.2 billion ($2.6 billion on March 31, 2026) — roughly 35 percent of all deposits sit above the FDIC's $250,000 guarantee. That is not unusual for a business bank, and 2023 taught everyone that uninsured money is the money that moves first. And the vise has begun to close a little: in the second quarter of 2026 the cost of deposits rose for the first time, from 1.91 to 1.92 percent, and the margin eased from 4.33 to 4.28 percent. Remember the yardstick: a margin built on cheap deposits is only as durable as the depositors' patience.

One truth we went looking for and did not find, because it belongs in the record: the office-tower disaster. Chicago commercial real estate has been a horror story for three years, and Byline is a Chicago CRE lender with $1.4 billion outstanding. But the exposure is smaller than the headline suggests: "CRE office represents 9.2% of our total CRE portfolio as of December 31, 2025, compared to 10.5% as of December 31, 2024" — a shrinking sliver of a diversified book (9.1 percent on March 31, 2026), roughly 1.7 percent of total loans. Non-performing loans across the whole bank were 0.95 percent at year-end and 0.92 percent on June 30, 2026, and net charge-offs fell further, to 0.24 percent in the second quarter. On the evidence in these filings, the office bomb is not ticking here. We looked; it is fair to say so.

Valuation: about $1.7 billion for a bank that earns 14 percent on its tangible money

In mid-July 2026 Byline stock cost about $37.48, for a market value of roughly $1.7 billion (price as of July 15, 2026; all valuation figures data as of July 16, 2026). Against 2025 earnings of $2.89 per diluted share, that is a price-to-earnings ratio around 13 — on the trailing twelve months through March 31, 2026, about 12.2. Bank investors usually reach for a different ruler, and it tells the more useful story: against tangible book value of $24.48 per share (June 30, 2026), the stock trades at about 1.5 times tangible book; against plain book value of $28.86, about 1.3 times. Is that expensive? The honest answer is: it is the price of the returns. A bank earning 13 to 14 percent on tangible equity conventionally deserves to trade above tangible book — roughly, a bank should be worth book value multiplied by how much better its returns are than what shareholders demand. At 1.5 times, the market is paying for those returns to hold, and for the capital hoard to eventually be put to work. If ROTCE slid back toward 12, that multiple would be generous. If the first two quarters of 2026 (13.77 and 14.47 percent, efficiency below 50) mark the turn, it is not.

The professionals' view is politely unexcited: six analysts cover the stock — two strong buys, one buy, three holds, no sells — with an average price target of $38.60, roughly three percent above the mid-July price. Read that honestly: the analyst consensus is not a catalyst, it is a shrug. Their estimates see $3.28 per share for 2026 and $3.39 for 2027, which puts the forward multiple near 11.4. That 2026 estimate already looks tight: Byline earned $1.73 per diluted share in the first half alone ($0.83 plus $0.90), an annualized $3.46. The dividend is real and small: after the 16.7 percent increase to $0.14 a quarter (declared July 21, 2026, payable August 18, 2026) it runs at $0.56 a year, a yield around 1.5 percent on the July price — this bank is not paying you to wait; it is keeping the money. And it keeps issuing shares alongside the buybacks: on June 3, 2026 it registered 1,697,594 shares for the 2026 omnibus incentive plan and another 200,000 for the employee stock purchase plan, together about 4.2 percent of the 45,191,493 shares outstanding on June 30, 2026, against 274,528 shares repurchased in the second quarter. Two closing notes for the complete picture: directors and executive officers together held 29.09 percent of the shares as of April 8, 2026, dominated by a single 26.15 percent holder (see our side finds), and the stock trades within about 0.3 percent of its 52-week high after a 45 percent twelve-month run. You can find more metrics in the Byline company profile of our scanner.

Opportunities and risks at a glance

What speaks for Byline Bancorp:

• A margin most regional banks cannot reach: net interest margin of 4.22 percent in 2025 (Q1 2026: 4.33 percent) on a loan yield of 7.07 percent, powered by a genuine specialty — one of the country's most active SBA lending franchises, with $1.6 billion of loans serviced for others (annual report 10-K 2025).
• Record earnings with real quality: net income of $130.1 million (+7.7 percent), return on assets of 1.36 percent, efficiency ratio of 51.83 percent — and 86.3 percent of revenue is recurring net interest income, not one-off gains.
• Credit is clean in a market where it was supposed to break: non-performing loans of 0.95 percent, net charge-offs falling to 0.39 percent, reserves at 1.45 percent of loans, and CRE office at just 9.2 percent of the CRE book and shrinking (December 31, 2025).
• A fortress balance sheet and ammunition: CET1 of 12.92 percent (June 30, 2026; 2023: 10.35 percent), tangible common equity of 11.36 percent of assets, tangible book value per share up from $17.98 (2023) to $24.48 (June 30, 2026) — plus a proven integration record of five bank acquisitions since 2016.
• The first half of 2026 turned the key ratio: net income of $37.6 million and $40.2 million, efficiency below 50 percent for the first time and then down to 46.93, ROTCE back up to 14.47 percent, the dividend raised 16.7 percent — and 4 insider purchases against 2 sales in twelve months (data as of July 16, 2026).

What speaks against it:

• Returns fell for two straight years while profit rose: ROTCE from 16.46 percent (2023) via 14.85 to 13.47 percent (2025), return on equity from 12.50 to 10.86 percent — the capital piled up faster than the earnings on it, and it is still waiting on a deal that has not happened yet: CET1 rose again in the second quarter of 2026, to 12.92 percent.
• The best business has a silent partner: the SBA can reclaim losses from "significant technical deficiencies", against which the bank keeps, in its own words, no reserves; the Preferred Lender status, the program rules and the funding caps all sit with the federal government, and a shutdown can stop originations and sales outright (10-K 2025, Item 1A).
• 14.3 percent of the equity is goodwill ($181.9 million): of $27.84 of book value per share, only $23.44 is tangible — the accumulated receipts of five takeovers, testable for impairment every year against the bank's own market value.
• The margin leap was rented: the cost of deposits fell from 2.61 to 2.17 percent only because the Fed cut 75 basis points, while the loan yield fell too (7.40 → 7.07 percent, and 6.84 percent in the second quarter of 2026); meanwhile free deposits shrank from 26.56 to 23.21 percent of the total (June 30, 2026), loans reached 98.37 percent of deposits in 2025, and uninsured deposits grew to $2.7 billion.
• Concentration and quiet complexity: essentially one metropolitan market (44 of 45 branches in Chicagoland, 67.6 percent of CRE loans in Illinois), plus $805.9 million of sponsor finance — leveraged loans to private-equity-owned companies with $2 to $10 million of EBITDA — inside what looks like a community bank; our scanner finds it in only one list and cannot rate it on classic fundamentals (data as of July 16, 2026).

What the Earnings Calls Reveal

We read all ten Byline Bancorp earnings call transcripts from 2024-Q1 through 2026-Q2 in chronological order and checked what management promised one or two quarters ahead against what actually arrived. The picture splits in two. On the numbers that sit at the centre of every presentation, the bank delivers with unusual reliability: net interest income guidance was met or beaten in all nine testable quarters, and pre-tax pre-provision return has now stayed above two percent for 15 consecutive quarters. Alongside that, however, sit several previously stated targets that have vanished from the reporting without ever being reached or formally withdrawn.

The clearest case is the ten billion dollar asset threshold, above which tighter rules and interchange fee caps apply to US banks. In 2024-Q1 the chairman said crossing it organically could be seen "right in 2025", adding that the mark did not consume the firm's thinking. By 2024-Q3 that had become a range of second half 2025 to first half 2026, expressly extendable in the same breath to full year 2026, and by 2024-Q4 the range reached into the third quarter of 2026. Only in 2025-Q3 did a firm date appear: the first quarter of 2026. That single date was missed and left uncommented: at the end of 2026-Q2 total assets are unchanged at 9.9 billion dollars. The reversal weighs heavier for us. In January 2026 the CFO conceded securities had been sold at year end to stay below the mark, while the president assured investors that going forward "you will not really see any type of management activity to try to keep us below a certain level". By July 2026 the message was that if the ability to stay below ten billion existed in early December, the company "would probably take you up on that, and we would just manage the balance sheet accordingly".

Three further points are worth tracking:

• The CFO explicitly named a sub-90 percent loan-to-deposit target in 2024-Q2, and in 2024-Q3 the decline to 92 percent was celebrated as good progress. The metric then disappeared from prepared remarks and moved the other way: 2025 loan growth was 8.9 percent against deposit growth of 2.5 percent, and in 2026-Q2 the ratio stands at 96 percent. The target carried no deadline, so it cannot be dated and missed — but it has not been reached either. No explanation was offered, and no analyst asked.
• The eight to nine percent tangible common equity target range was a fixed part of the 2024 commentary. Since 2024-Q3 the bank has run above it, most recently at 11.4 percent, roughly 250 basis points above its own stated band. In 2025-Q4 an analyst asked directly whether anything should be read into the target disappearing from the decks; the answer cited concrete steps — a dividend increase, buybacks executed in the quarter, a new authorisation covering five percent of shares — and the wish to keep room for acquisitions, but no new target range.
• Expense guidance was mostly met but clearly missed in 2025-Q3: 56 to 58 million dollars was guided, 60.5 million was reported. Conversely the 2026 full year guide, most recently 59 to 60 million per quarter, sits above the actual run rate of 57 and 56.5 million in the first two quarters of 2026 — and was called unchanged in 2026-Q2 even though the lower bound had moved up from 58 to 59 million.

On the other side of the ledger, where the bank puts a number on the table it is highly dependable. Guidance of 83 to 85 million dollars of net interest income for 2024-Q2 produced 86.5 million; 95 to 97 million for 2025-Q3 produced 99.9 million; 97 to 99 million for 2025-Q4 produced 101 million. An analyst raised the pattern openly in 2025-Q4, and management pointed to hard-to-forecast loan payoffs and a deliberately cautious deposit cost assumption it keeps beating internally. The First Security acquisition was also delivered faster than promised: guided for the first half of 2025, closed on April 1, with the systems conversion completed the same month. What was not delivered were the purchase marks, book value dilution and earnings accretion for that deal, deferred twice to the next call in 2025-Q1 and then absent from the 2025-Q2 call, with no follow-up question.

Our conclusion: operationally Byline is a well run bank with genuinely robust guidance quality on net interest income, a growing book value and a rising dividend. Investors should nonetheless separate what the bank delivers from what it lets itself be measured against. Several self-set targets quietly left the communication between 2024 and 2026 without ever being reconciled, and on the single most important strategic question, crossing ten billion dollars, the stated position reversed within six months. The next checkpoint is the 2026 full year report.

A human conclusion

Back to the labels from the opening — because with a bank, that is not a metaphor, it is the only method you have. Three labels, three findings. "Record profit" is true, and it is the least informative thing on the box. Byline earned $130.1 million, more than ever, and the number that actually tells you what the owners' money did — return on tangible equity — fell two years running, from 16.46 to 13.47 percent. Not because the bank got worse, but because it is quietly stacking capital for a purchase it has not made. That is a defensible strategy and a real cost, and you are paying it now for a benefit that arrives later, if it arrives. At least the first half of 2026 showed the return can turn — 13.77 percent, then 14.47. "Government guaranteed" is true, and it is conditional. Washington stands behind the loans — as long as Byline's paperwork holds, its Preferred Lender badge holds, Congress funds the program and the government stays open; against the clawback risk there is no reserve at all. And "SBA gain-on-sale story", the label the market has stuck on this bank, is simply wrong — those gains are 5.1 percent of revenue and falling, while 86.3 percent is plain lending spread. That is the good news; the sting is that the government's fingerprints are not in the tip jar but in the wages, in a 7.07 percent loan yield that no ordinary Chicago bank could earn.

So what is actually on the shelf? A well-run, unfashionable, genuinely profitable bank with clean credit, a rented margin, a fortress balance sheet, a quarter of its stock in one pair of hands and a strategy that requires it to keep buying rivals to justify the capital it is holding back from you. At about 1.5 times tangible book, the price says the market believes the returns will hold. The filings say the returns fell for two years and only just started rising again. Both of those are true at once, and the next few quarterly reports will tell you which one was the label and which one was the merchandise. What you make of it is your decision. And that is exactly as it should be.

Sources

All original documents used in this analysis — to read for yourself:

• Byline Bancorp, Inc. — SEC annual report 10-K for 2025 (filed February 27, 2026)
• Byline Bancorp, Inc. — SEC quarterly report 10-Q as of March 31, 2026 (filed May 1, 2026)
• Byline Bancorp, Inc. — SEC current report 8-K of July 23, 2026, Exhibit 99.1: second-quarter results as of June 30, 2026
• Byline Bancorp, Inc. — SEC proxy statement DEF 14A 2026 (filed April 20, 2026; stock ownership as of April 8, 2026)
• Byline Bancorp, Inc. — SEC current report 8-K of June 3, 2026: results of the annual meeting held June 2, 2026
• Byline Bancorp, Inc. — SEC registration statement S-8 of June 3, 2026 (1,697,594 shares for the 2026 omnibus plan plus 200,000 for the employee stock purchase plan)
• Byline Bancorp, Inc.'s complete SEC filing history: EDGAR overview (sec.gov)
• Earnings call transcripts 2024-Q1 through 2026-Q2 (public transcripts, 10 calls) — review as of August 2, 2026
• Fundamental data (metrics, quarterly series, valuation; data as of July 16, 2026), reconciled with the SEC filings.
• Screener and rating data: in-house stock scanner (data as of July 16, 2026); Reddit mentions: ApeWisdom (as of July 16, 2026).

Transparency & disclaimer: This analysis is a journalistic contextualization of publicly available information and is not investment advice, not a financial analysis in the regulatory sense, and not a solicitation to buy or sell securities. Stock investments carry substantial risks up to total loss. All information without guarantee; the data cut-off is noted in the text in each case. The author holds no position in Byline Bancorp stock at the time of publication.