← David Hay hub  ·  Research hub  ·  Research library

David Hay — Portfolio Update, Double Edition: QNST + J

A short Monday double-header on two names Haymaker already owns, and a study in how differently he treats a winner that is accelerating versus one that has merely started working. QuinStreet (QNST), recommended Feb 13 at $11.25, reported Q4 FY2026 after the close on Aug 6 and "the stock jumped 36% in the after-hours session" — revenue $373.9M, +42.7% YoY, a 4.1% beat; adj. EPS $0.50, a 13.6% beat; and Q1 FY2027 guided to $375M against a $358.8M consensus. At ~$20.70 it is +84% in under six months with a fresh 52-week high, and the sequential tape is "the cleanest data presentation in the story" — $287.8M → $346.1M → $373.9M → $375M guided, "four consecutive quarterly records… with no sign of deceleration." The stock is now above the $19 average target, which Haymaker discounts on principle: "analyst targets are almost always backward-looking relative to earnings inflections of this magnitude." Medium-term case unchanged — 12-15% EBITDA margins on $1.5B revenue → $180-225M EBITDA → a $2.7-4.1B cap at 15-18×, "more than a double… in excess of a triple" — but "what has changed is the timeline." Verdict: "Hold the core position," with a "rational basis for trimming 25 to 33%" of an outsized weight. Credit is handed off: "the main credit for this big winner goes to Sy Jacobs… a stock-picking superstar," whose late-2023 Sprott (SII) call "was a four-bagger… and it's still nearly a three-bagger." Jacobs Solutions (J), re-recommended May 15 at $128.30 and now ~$145 (+13%), is conceded to have been "a 'meh' performer" since the April-2024 first highlight at the same upper-$120s price — but the Aug 4 Q3 print is "a clean thesis confirmation": adj. EPS $1.84 +14%, adj. EBITDA $367M +17% at a 15.2% margin (+109bp), and "the headline that matters most: backlog reached a record $29 billion, up 27%… with a trailing 12-month book-to-bill of 1.4 times." Guidance was raised for a third consecutive quarter to $7.20-7.30 (~19% growth), direct AI infrastructure is 11% of adjusted net revenue, life sciences/advanced manufacturing grew 24% (a company record), and net leverage hit 1.8× a quarter early. At ~19× the guidance midpoint that is "not expensive" — but "we'd be a bit reluctant to chase this one after the recent rally… we'd look to buy it on a 5% pull-back."
2026-AUG-10 · Haymaker (Substack newsletter, paid) · The Haymaker Team / David Hay · Portfolio Update (Double Edition) · ↗ Read · article text · actionable insights
One-line take: the shorter Monday portfolio-update format (announced Jul-27 as "two or three… each Monday") applied to two owned names, and the interest is in the asymmetry of the two instructions — the name that has run 84% is a hold with a permission to trim, while the name that has run 13% is a wait-for-a-pullback buy. Neither is a rating change argued from valuation; both are position-management calls argued from the operating print. QNST. The Q4 FY2026 report (Aug 6, after the close) "was a pretty thorough validation of every element of the thesis we laid out in February" — that "the property-casualty insurance repricing cycle was restoring carrier profitability and reopening marketing budgets that flow disproportionately through QuinStreet's performance marketing infrastructure," and "in every quarter since, that projection has proven out." The numbers: $373.9M revenue +42.7% (4.1% above the $359.3M consensus), adj. EPS $0.50 (13.6% above $0.44), and — the line Haymaker treats as decisive — Q1 FY2027 guided to $375M against $358.8M, which "signals that management does not believe the Q4 acceleration is a single-quarter phenomenon." The evidence is presented as a sequence rather than a level: Q2 $287.8M, Q3 $346.1M, Q4 $373.9M, Q1 FY2027 guided $375M — "four consecutive quarterly records, each building on the last, with no sign of deceleration" — and read against the cycle: "a business growing at this pace in a carrier spending recovery that is still in its early innings is not a business that has pulled forward its growth, but one more likely in the middle of its cycle." The awkward fact is met head-on — the stock at ~$20.70 is above the $19 average target (five analysts, Strong Buy, high $24 from Barrington) — and dismissed on a stated rule: "analyst targets are almost always backward-looking relative to earnings inflections of this magnitude, and a 36% single-day move on a 42.7% revenue beat forces a model rebuild." On the multiple: "approximately 6.5x to 7x run-rate EBITDA on a business growing revenue 42% with expanding margins and above-consensus forward guidance" is not "necessarily stretched." The destination is unchanged and the clock has moved: 12-15% EBITDA margins on $1.5B revenue → $180-225M EBITDA → $2.7-4.1B cap at 15-18× ("$2.7 billion would be more than a double from here, while $4.1 billion would be in excess of a triple"), and "what has changed is the timeline: the pace of execution suggests we get there faster than the July base case assumed." The instruction is therefore a sizing rule, not a valuation call: "Hold the core position… Haymakers who bought QNST and have seen it grow to an outsized share of total portfolio weight following the 84% move have a rational basis for trimming 25 to 33%. But overall, the thesis is intact, the execution is exceptional, and the Q1 FY2027 guidance suggests this is not a high-water mark." That is a genuine softening from Jul-20, where the table moved QNST to a plain Hold and said hold the full position while explicitly deferring the sizing question "after the Aug-6 Q4 print" — the print has now landed and the answer is trim the excess, keep the core. Attribution is explicit: "the main credit for this big winner goes to Sy Jacobs… we have often called Sy a stock-picking superstar," with the scorecard receipt that in late 2023 "he gave us a bull story on Sprott, Inc. (SII). At one time, that was a four-bagger… and it's still nearly a three-bagger." J. Haymaker leads with the disappointment rather than burying it — re-recommended May 15 at $128.30, now ~$145 (+13% in under three months), "however, our first highlight of this name was in April 2024 at around the same upper-$120s price. In other words, it's been a 'meh' performer." The Aug 4 Q3 FY2026 print is then read as the first real test since the re-recommendation, and "it delivered a clean thesis confirmation across every metric that matters for this position": adj. EPS $1.84 +14% on "more than 8% organic adjusted net revenue growth," adj. EBITDA $367M +17% with margin "expanding 109 basis points to 15.2%," and "the headline that matters most: backlog reached a record $29 billion, up 27% year-over-year, with a trailing 12-month book-to-bill of 1.4 times on gross revenue." Supporting: guidance raised "for the third consecutive quarter" to $7.20-$7.30 ("nearly 19% year-over-year growth at the midpoint"), direct AI infrastructure now 11% of adjusted net revenue (+100bp sequentially), life sciences and advanced manufacturing (data centers and semiconductors) +24% — "the strongest quarter the company has ever reported in that segment" — and net leverage 1.8×, "hitting the below-2.0x target one quarter ahead of schedule." The honesty is deliberate: "the honest assessment of Q3 is in-line rather than a clean beat," with EPS ~1.5% below the $1.87 consensus and adjusted net revenue 1.4% light. Valuation: "at $139, Jacobs trades at approximately 19x the $7.25 adjusted EPS guidance midpointnot expensive for a company with a $29 billion backlog growing 27%, a 1.4 book-to-bill, and direct structural exposure to AI infrastructure, semiconductor fabrication, water, nuclear, defense, and energy security spending — the exact intersection of structural tailwinds that should be a solid growth-engine for the next decade." The entry discipline is the whole action item, and it is explicitly tied to the house's macro caution: "while we'd be a bit reluctant to chase this one after the recent rally – particularly, given our broad market concerns – we'd look to buy it on a 5% pull-back." Closing hope: "with the bulging backlog and the enviable position it has… once it catches its breath for a spell, it might be poised to make up for lost time." No Buys / Trims-Holds / Sells tables accompany this issue — "we will publish our updated portfolio tables later this week." Dating note: the byline is AUG 10 but the text is written on Aug 7 ("as of this writing on August 7"), so every quoted price is three sessions stale relative to the publish date; the folder follows the byline per the archive convention.

1. Stocks & names mentioned

TickerNameResearchViewWhat he saidAt
QNSTQuinStreetQT · SA · STK · FAPositive"Our current view on QNST: Hold the core position" — a thesis-intact hold with an explicit sizing allowance, after Q4 FY2026 (Aug 6, after the close) sent the stock up 36% in the after-hours session to ~$20.70, a new 52-week high of $20.91 and +84% from the Feb-13 $11.25 entry in under six months. The print "was a pretty thorough validation of every element of the thesis we laid out in February" — that the P&C insurance repricing cycle was restoring carrier profitability and reopening marketing budgets flowing through QuinStreet's performance-marketing infrastructure: revenue $373.9M, +42.7% YoY (4.1% above the $359.3M consensus), adj. EPS $0.50 (13.6% above $0.44), and Q1 FY2027 guided to $375M vs $358.8M — which "signals that management does not believe the Q4 acceleration is a single-quarter phenomenon." The evidence Haymaker leans on is the sequence: $287.8M → $346.1M → $373.9M → $375M guided, "four consecutive quarterly records… with no sign of deceleration," in "a carrier spending recovery that is still in its early innings" — so this is a business "more likely in the middle of its cycle," not one that has pulled growth forward. That the price now sits above the $19 average target is dismissed on a stated rule: "analyst targets are almost always backward-looking relative to earnings inflections of this magnitude, and a 36% single-day move on a 42.7% revenue beat forces a model rebuild"; at ~6.5-7× run-rate EBITDA valuation is not "necessarily stretched." Medium-term case unchanged (12-15% EBITDA margins on $1.5B revenue → $180-225M EBITDA → $2.7-4.1B cap at 15-18× = "more than a double… in excess of a triple"); "what has changed is the timelinefaster than the July base case assumed." The one qualifier is about your book, not the company: holders whose position has grown to "an outsized share of total portfolio weight" have "a rational basis for trimming 25 to 33%" — the sizing question Jul-20 deferred "after the Aug-6 Q4 print," now answered. "The thesis is intact, the execution is exceptional… this is not a high-water mark." Credit to Sy Jacobs.read ↗
JJacobs SolutionsQT · SA · STK · FAPositivePositive with a stated entry discipline rather than a chase: "we'd look to buy it on a 5% pull-back." Re-recommended May 15 at $128.30, now ~$145 (+13% in under three months) — but Haymaker leads with the record, not the rally: "our first highlight of this name was in April 2024 at around the same upper-$120s price. In other words, it's been a 'meh' performer." The Aug-4 Q3 FY2026 print is "the first major data point since our second endorsement, and it delivered a clean thesis confirmation across every metric that matters": adj. EPS $1.84 +14% on >8% organic adjusted net revenue growth; adj. EBITDA $367M +17% with margin "expanding 109 basis points to 15.2%"; and "the headline that matters most: backlog reached a record $29 billion, up 27% year-over-year, with a trailing 12-month book-to-bill of 1.4 times on gross revenue." Plus a third consecutive guidance raise to $7.20-$7.30 ("nearly 19% YoY growth at the midpoint"), direct AI infrastructure at 11% of adjusted net revenue (+100bp sequentially), life sciences & advanced manufacturing +24% — "the strongest quarter the company has ever reported in that segment" — and net leverage 1.8×, clearing the sub-2.0× target "one quarter ahead of schedule." He books the miss honestly: "the honest assessment of Q3 is in-line rather than a clean beat" (EPS ~1.5% below the $1.87 consensus, adjusted net revenue 1.4% light). On price: "at $139, Jacobs trades at approximately 19x the $7.25 adjusted EPS guidance midpoint… in our view that's not expensive" for a $29B backlog growing 27% with "direct structural exposure to AI infrastructure, semiconductor fabrication, water, nuclear, defense, and energy security spending — the exact intersection of structural tailwinds that should be a solid growth-engine for the next decade." The restraint is tied to the house macro stance: "we'd be a bit reluctant to chase this one after the recent rally – particularly, given our broad market concerns." Closing: "once it catches its breath for a spell, it might be poised to make up for lost time."read ↗

"View" is Haymaker's stance in this update. QNST and J are the two rated positions. Referenced only (not rowed): Sprott Inc. (SII) — named purely as a track-record receipt for Sy Jacobs ("in late 2023, he gave us a bull story on Sprott, Inc. (SII). At one time, that was a four-bagger… and it's still nearly a three-bagger. Great thanks, Sy!"), with no current stance stated. Deliberately not rowed: unlike Jul-31's JPM, where the scorecard entry came with an unambiguously positive view of the name, this is a thank-you aside — rowing it would overwrite the argued Negative stance of the May-15 full exit at ~$138 on "quite spendy" (9× P/S, 29× P/E) valuation, which remains Haymaker's standing position. Note the aside does corroborate that exit: a four-bagger now "nearly" a three-bagger has given back a meaningful share of the peak. Also referenced: Sy Jacobs (the source of both the QNST and SII ideas), Barrington (the $24 high target on QNST), and the property-casualty carriers whose reopened marketing budgets are QNST's demand driver. No Buys / Trims-Holds / Sells portfolio tables accompany this issue — "we will publish our updated portfolio tables later this week." Research: QT Qualtrim · SA Seeking Alpha · STK Stock Analysis.

2. Talking points

QNST — the print, and why the timing is "juicy"

What the business actually is

The February thesis, and its scorecard

The Q4 numbers

Why the report "changes the framing" — the sequence, not the level

Above the target price — and why that isn't the argument

The destination is the same; the clock moved

The Sy Jacobs credit — and the Sprott receipt

The instruction — hold the core, trim the excess

J — leading with the disappointment

The Q3 print — the operating metrics

The number that matters most — backlog and book-to-bill

The supporting evidence — guidance, mix and the balance sheet

The honest caveat

Valuation — 19× against a decade of tailwinds

The entry rule — and the macro caveat behind it

Housekeeping — the tables are deferred

3. In plain English

A jargon-free note on why each name matters here. (Companion to the table above; renders on each name's consolidated page.)

QNST — QuinStreet Positive

QuinStreet runs online marketplaces that connect people who are actively shopping — for car insurance, a mortgage, a roofer — with the companies that want their business. It doesn't sell insurance; it sells the introduction, and it only gets paid when the introduction works (a click, a call, a signed customer). So its revenue is essentially a share of what insurers and home-services companies choose to spend on marketing.

That is why Hay bought it in February at $11.25. Car and home insurers had spent two years raising premiums to repair their finances; once profitable again, they compete for customers, and that spending flows straight through companies like QuinStreet. The bet was on the timing of that turn.

The August 6 results settled it. Revenue was $373.9 million, up nearly 43% on the year and comfortably ahead of forecasts, profits beat by more, and — the part Hay cares most about — management guided the next quarter above what analysts expected, which says the surge isn't a one-off. Line the quarters up and you get $287.8m, $346.1m, $373.9m, and a $375m forecast: four records in a row, each bigger than the last. The stock jumped 36% overnight to about $20.70, up 84% from his entry in under six months.

An obvious objection: the stock is now above the average analyst price target of $19. Hay's answer is a general rule worth remembering — analysts' targets lag reality when a company's earnings step-change like this, because a 43% revenue beat forces every one of them to rebuild their model from scratch. On the numbers he cares about, it still trades at about 6.5-7 times its current run-rate cash profits, which is not demanding for something growing this fast.

Where he thinks it goes: if the business gets to $1.5 billion of revenue at a 12-15% cash-profit margin, that's $180-225 million of profit, worth $2.7-4.1 billion at normal multiples — more than double, potentially triple, today's value. That target hasn't changed; what has changed is how fast it looks like arriving.

So the instruction is about your portfolio, not the company: hold the core position. If the 84% gain has made it an uncomfortably large slice of what you own, trimming a quarter to a third is reasonable risk management — but the thesis is intact and he doesn't think this is the top. He also gives the credit away: the idea came from Sy Jacobs, whose late-2023 Sprott call was at one point a four-bagger.

J — Jacobs Solutions Positive

Jacobs is an engineering and design firm — the people who plan and manage the building of data centres, semiconductor fabs, water systems, power plants and defence facilities. It doesn't pour the concrete; it designs the project and runs it, which is a higher-margin, less capital-hungry way to profit from a construction boom.

Hay is unusually blunt about his own record here. He first flagged it in April 2024 in the high $120s, re-recommended it in May 2026 at $128.30, and it is only now around $145. Two years, roughly flat, then a 13% move: "a 'meh' performer," in his words.

The August 4 results are why he still likes it. Profits per share rose 14%, cash profits rose 17%, and the margin improved — but the number he singles out is the backlog: $29 billion of work already won but not yet done, up 27% in a year. Alongside it sits a "book-to-bill" ratio of 1.4, which simply means that for every dollar of work Jacobs completed over the past year it signed $1.40 of new work. Both figures describe revenue that is largely already contracted, which is far better evidence about the next few years than one quarter's earnings. Management also raised its full-year forecast for the third quarter running, work directly tied to AI infrastructure is now 11% of revenue, the data-centre-and-semiconductor segment grew 24% (its best quarter ever), and debt came down faster than promised.

He doesn't oversell it: on the quarter itself, earnings actually came in about 1.5% below what analysts expected. His point is that the headline number was in line while the forward indicators were excellent.

At roughly 19 times this year's expected earnings he thinks the price is fair rather than cheap — reasonable for a company sitting at the intersection of AI build-out, chip manufacturing, water, nuclear, defence and energy security. But he won't chase it after the recent run, especially with his broader worries about the market, so the instruction is a patient one: buy it on a 5% pullback. His view is that once the stock pauses for a while, the backlog should let it make up for the years it went nowhere.


Summary derived from the paid Haymaker newsletter (text in transcript.txt) for personal study. Not investment advice. © Haymaker / David Hay for source material.