Operating Leverage Returns: 80 Basis Points of Margin, Pays Per Control Off the Floor, and a Raised Guide the Market Ignored

Results vs. Consensus

Q2 FY2026 Scorecard

Year-Over-Year Comparisons

Quarter-Over-Quarter Comparisons

The sequential comparison is the one that matters this quarter, because the whole question coming out of the September print was whether the promised margin ramp was real. It is: adjusted EBIT margin gained 52 basis points sequentially and PEO margin, which had collapsed 144 basis points below the segment's own run rate on a cluster of one-time items, recovered all of that and then some.

First-Half Comparison

The six-month view is the right frame for judging the full-year guide, because it strips out the fiscal-first-quarter acquisition drag that management has spent two calls explaining and shows what the business is actually running at.

Quality of Beat

Revenue. The 6.2% headline is 5% on an organic constant-currency basis, with foreign exchange contributing a point and acquisitions now contributing nothing after the WorkForce Software anniversary. That is a cleaner mix than the September quarter, where reported growth of 7.1% carried a point of acquisition help that was always going to disappear. Within the mix, the lowest-effort line still grew fastest: interest on client funds rose 13.1% while revenues other than client-funds interest and PEO rose 5.9%, a spread of 7.2 points against 6.7 points in the September quarter. So the revenue mix did not improve. What improved is what the revenue converted into. Ex-client-funds-interest revenue grew 5.8% and ex-float pretax earnings grew 6.4%, meaning the operating business now grows profit faster than revenue. A quarter ago the same two figures were 6.7% and 1.1%.

Margins. Adjusted EBIT margin expanded 80 basis points to 26.0%, against a full-year guide of 50 to 70 basis points that management pointedly left unchanged. All three pieces contributed. Employer Services margin turned from minus 50 basis points to plus 50. The Other segment's loss narrowed by $24.9M as corporate extended interest income rose to $77.0M from $65.0M and the cost of short-term financing supporting the client funds strategy fell to $91.3M from $100.0M. The only drag was the PEO at minus 70 basis points, and two-thirds of that is arithmetic rather than performance. The CFO's attribution of the Employer Services turn is worth reading carefully, because it names operating leverage first.

"ES margins increased by 50 basis points in Q2, driven by both operating leverage and the contribution from client funds interest revenue growth."
— Peter Hadley, CFO

EPS. Adjusted diluted EPS of $2.62 grew 11.5% against net earnings growth of 10.3%. The gap is share count, down 1.1% year over year on $515.8M of repurchases in the quarter, a pace roughly 41% above the September quarter's $366.0M. Tax helped at the margin: the adjusted effective rate fell 41 basis points to 23.2%. The reconciliation is again almost empty, with a $1.2M optimization-initiative reversal and a $2.2M venture-investment gain the only pretax adjustments, which is why GAAP and adjusted diluted EPS are identical at $2.62. There is no accounting story here in either direction.

Segment Performance

Employer Services

Employer Services grew 6.5% reported and 5% on an organic constant-currency basis, with foreign exchange worth about a point and acquisitions worth nothing now that the October 2024 deal has annualised. The segment carried interest on funds held for clients of $308.6M, up 13.1% on average balances of $37.6B, up 6%, at a yield of 3.3%, up 20 basis points. That line accounts for $35.8M of the segment's $218.9M revenue increase, so roughly 84% of the increase came from the processing and HCM software base rather than from the float. In the September quarter the equivalent split was 85 to 15 the same way, but on a segment whose margin was falling; this quarter the margin rose.

The 50 basis point margin gain is the swing factor for the whole FY26 story. It is the mirror image of the 50 basis point decline a quarter earlier, and the mechanism is exactly what management said it would be.

"we still had in Q1, as you remember, the fourth quarter of the before the anniversary of the Workforce Software acquisition. So we had some acquisition-related drag in the first quarter. The second quarter came in strongly."
— Peter Hadley, CFO

Demand held up. Management characterized new business bookings as solid and broad-based, with the fastest growth in international, U.S. enterprise, and compliance, and small business and mid-market both contributing. The 4% to 7% full-year bookings guide was maintained, as was the 10 to 30 basis point full-year retention decline. Retention declined modestly in the quarter and landed in line with plan, against a first quarter that had come in slightly better than plan.

"We delivered solid employer services new business bookings growth in the second quarter. We enjoyed broad-based strength with the fastest growth in our international, US enterprise, and compliance businesses."
— Maria Black, President and CEO

Assessment: This segment answered the question the September quarter raised. Margin expansion resumed, the growth mix improved as the acquisition contribution rolled off, and the pays-per-control lever went from switched off to modestly positive. The remaining caution is that 5% organic constant-currency growth still sits at the bottom of what a roughly 6% reported full-year guide implies, so the currency tailwind is doing visible work.

PEO Services

The PEO is where this quarter got worse rather than better. Segment revenue grew 5.5% to $1,755.2M, but zero-margin benefits pass-through costs of $1,126.2M grew 6.7% and now represent 64% of reported segment revenue. Revenue excluding those pass-throughs grew about 3%, down from 6% in the first quarter and at the bottom of the unchanged 3% to 5% full-year range. Segment earnings were essentially flat at $252.9M, up 0.5%, and margin fell 70 basis points to 14.4%.

"PEO margins decreased 70 basis points in Q2, driven mainly by zero-margin pass-through growth and higher selling expenses."
— Peter Hadley, CFO

Two causes rather than the four cited last quarter, and both are the structural ones. The pass-through mix effect dilutes reported margin every quarter that health benefit costs inflate faster than PEO service fees, which is to say most quarters. Higher selling expense is the deliberate investment in the PEO sales team that management defended in October and defended again here. The one-time items that took the September quarter down 140 basis points, state unemployment insurance timing and the employee retention tax credit filing change, did lap, which is why sequential margin recovered 144 basis points.

The demand side is what changed. Average worksite employees grew about 2% to 758,000, and management cut the full-year growth guide from 2% to 3% down to about 2% on two admissions.

"PEO new business bookings growth was solid in Q2 but did come in slightly below our expectations."
— Peter Hadley, CFO

"Our 2% growth in average worksite employees included a moderation in PEO pays per control growth."
— Maria Black, President and CEO

The second of those is the more interesting disclosure. PEO pays per control historically runs ahead of the Employer Services metric because the PEO client base skews toward faster-growing small and mid-sized employers. This quarter the two converged, which management noted explicitly. A worksite-employee guide that is entirely bookings-dependent, cut in the same quarter that bookings missed plan, is the one place in this report where the direction of travel is wrong.

Assessment: Reset the PEO's contribution expectations down. The margin baseline is structurally lower than the pre-pass-through-inflation history implies, the growth algorithm now depends entirely on selling into a market where the company just missed its own bookings plan, and the guide cut came at the bottom rather than the middle of the range. This is a real offset to the Employer Services improvement, and it is why we are upgrading on valuation and earnings quality rather than on a clean sweep of the operating metrics.

Other Segment and the Float Geography

The Other segment's loss narrowed to $148.5M from $173.4M. Corporate extended interest income rose $12.0M to $77.0M, and the cost of short-term financing supporting the client funds strategy fell $8.7M to $91.3M as short rates declined. Those two together are $20.7M of the $24.9M improvement. Management flagged the dynamic as a deliberate part of the FY26 margin story.

"This margin contribution is being driven by growth in our corporate extended interest income, while at the same time, our short-term financing costs are decreasing."
— Peter Hadley, CFO

The geography point from last quarter still holds and still matters for modelling: client funds interest revenue sits inside Employer Services, while corporate extended interest income and the related financing cost sit in Other. The composite ADP reports as "Net Impact from Client Funds Strategy" is the only place all three come together, and it was $294.3M against $237.8M, up 23.8%. What changed is the denominator. Pretax earnings grew $121.9M, so the composite's $56.5M contribution is 46% of the growth rather than the 84% it represented in the September quarter.

Assessment: The float is doing less of the work and doing it more cheaply, because the falling front end that compresses the client-short yield simultaneously cuts the financing cost of the extended strategy. That is the natural hedge inside the structure and it was invisible last quarter because the operating business contributed nothing. It is visible now.

Key KPIs

Key Topics & Management Commentary

Overall Management Tone: Management was measured and specific in a way it was not three months ago, leading prepared remarks with the three headline financial results rather than with strategy, and volunteering the two pieces of bad news (PEO bookings below plan, the worksite-employee guide cut) inside the same paragraph that delivered the guidance raise. Analyst pushback was concentrated on second-half margin cadence and on the PEO deceleration, and both received quantitative answers naming the specific mechanisms rather than the framework-level responses that characterized the October call. The only place the posture thinned was pays per control, where management declined to raise a guide its own reported number had already cleared.

1. Pays Per Control Comes Off the Floor

The metric that took 6.6% out of the stock in October, and took 5% to 7% out of every other payroll name on the same session, rounded up rather than down. U.S. pays per control grew 1% against 0% in the September quarter and 1% in the year-ago quarter. It was the second sentence of the CEO's results review rather than a buried clarification.

"Employer services pays per control growth rounded up to 1% for the second quarter, representing modestly higher year-on-year growth compared to the first quarter."
— Maria Black, President and CEO

The full-year guide of approximately flat was left unchanged, and the CFO explained why in a way that is more informative than the guide itself: the underlying trend is close enough to the rounding boundary that the reported figure could print either 0% or 1% for the balance of the year.

"It is a question of is that growth just above or just below the half percent mark."
— Peter Hadley, CFO

Management also said the contribution was broad-based across industry groups and across the small-business, mid-market, and enterprise segments, and made a point of noting that ADP's own base is not showing the down-market slowdown visible in the wider economy.

Assessment: Two things follow. First, the October guide cut has stopped being a one-way risk: a metric guided to approximately flat that prints 1% for two consecutive quarters creates cushion in the Employer Services revenue line rather than consuming it. Second, management is now under-promising on the one metric it over-promised on last year, which is the correct response to having been caught out. We move our own assumption to plus 0.5%, above the guide.

2. Adjusted EBIT Margin Expands 80 Basis Points, and the Bridge Stops Being a Leap

Adjusted EBIT margin rose to 26.0% from 25.2%. That single number does more to resolve the FY26 debate than anything else in the release, because the whole bear case coming out of the September quarter was arithmetic: a full-year guide of 50 to 70 basis points of expansion, zero delivered in the first quarter, therefore roughly 80 basis points required across the remaining nine months from sources management had named but not sized.

"This morning, we reported strong second quarter results that included 6% revenue growth, 80 basis points of adjusted EBIT margin expansion, and 11% adjusted EPS growth."
— Maria Black, President and CEO

The composition matters as much as the number. Of the 80 basis points, Employer Services supplied a 50 basis point segment gain on a segment that is 67% of revenue, the Other segment supplied a $24.9M reduction in its loss, and the PEO gave back 70 basis points on a segment that is 33% of revenue. Netting those against the consolidated result implies the operating leverage is genuine rather than a single lucky line.

Assessment: The guide was maintained at 50 to 70 basis points after a quarter that delivered 79. That is either conservatism or an acknowledgement that the second half faces a float headwind management then went on to describe in detail. Our read is that it is both, and that the maintained band is now comfortably achievable rather than aspirational.

3. The WorkForce Software Anniversary Lands, Exactly as Promised

Employer Services margin went from minus 50 basis points to plus 50 in one quarter, a 100 basis point swing that management attributed to the October 2024 acquisition annualising out of the comparison. This was the one clean, dated, self-extinguishing item in last quarter's report, and it delivered on schedule.

The strategic case advanced at the same time. Management launched the ADP Workforce Suite during the quarter, integrating the acquired workforce-management capability across the payroll and HCM platforms, and cited it as a direct contributor to the largest international win of the quarter.

"During the second quarter, we launched the ADP Workforce Suite, our integrated workforce management solution across our leading payroll and HCM platforms."
— Maria Black, President and CEO

Assessment: A margin headwind that management dated precisely, sized at 50 basis points, and then delivered the reversal of on the promised quarter is the kind of thing that should increase the credibility weight on the rest of the guide. It is the strongest evidence in this report that the October reaffirmation was honest rather than hopeful.

4. The Float Still Contributes, but No Longer Dominates

Interest on funds held for clients grew 13.1% to $308.6M on average balances of $37.6B, up 6%, at a 3.3% yield, up 20 basis points. Corporate extended interest income added $77.0M against $65.0M, and short-term financing costs fell to $91.3M from $100.0M. The composite rose 23.8% to $294.3M.

"Client funds interest revenue increased slightly more than we anticipated in Q2, helped mainly by higher average client funds balance growth."
— Peter Hadley, CFO

Management raised the full-year balance-growth assumption to 4% to 5% from 3% to 4%, held the yield at approximately 3.4%, and raised both dollar guides by $10M. Note what that says about the wage-moderation assumption embedded in the October guide: three months ago management was modelling balance growth decelerating from 7% delivered to 3% to 4% for the year, and it has now moved that up rather than down. Wages across the client base are holding better than management assumed.

"We have increased our forecast for average client funds balance growth to 4% to 5% in fiscal 2026, and we continue to expect an average yield of approximately 3.4%."
— Peter Hadley, CFO

Against that, the blended forward new-purchase rate used to price portfolio additions over the balance of the year fell again, to 3.8% from 3.9% at the October guidance date and 4.1% at the July date. That is a second consecutive decline in the price of the asset ADP is buying with its incremental balances.

Assessment: The float remains a rate-driven line whose price is set by the forward curve and whose volume is set by aggregate wages. What changed is that it is no longer carrying the earnings statement. At 46% of pretax earnings growth rather than 84%, the multiple that this earnings stream deserves is closer to the multiple the market pays for processing growth. The composite is still compounding, but the growth contribution decays from here as the comparisons harden.

5. The PEO Is the One Line Going the Wrong Way

Revenue excluding zero-margin pass-throughs decelerated to 3% from 6%. Bookings came in below management's own expectation. PEO pays per control moderated to the point where it converged with the Employer Services metric, which management noted is unusual because the PEO base normally runs ahead. The full-year worksite-employee guide was cut from 2% to 3% down to about 2%.

"we are now expecting average worksite employee growth of about 2% in fiscal 2026"
— Peter Hadley, CFO

Management kept the PEO revenue guides unchanged at 5% to 7% reported and 3% to 5% excluding pass-throughs. The first half tracked at 6.3% reported and 4.6% excluding pass-throughs, which means the maintained ranges do not require the second half to improve on the first. Holding the 4% midpoint of the ex-pass-through range needs about 3.5% in the second half, below the first half and close to the 3.4% the December quarter itself delivered. The 5% top of the range is the only outcome that requires acceleration. Asked directly whether the lower end was now the right place to model, management declined to narrow, saying the range exists because all outcomes remain live and the answer depends on bookings, pays per control, and retention across the prime selling season.

Assessment: A bookings-dependent growth guide that gets cut in the quarter bookings miss is the weakest disclosure in this report. It is contained rather than escalating, because the PEO is 33% of revenue and 18% of consolidated pretax earnings, and because the guide was cut by a single point at the top rather than reset. But it is the reason this quarter is not a clean sweep, and it is the pillar we would watch for the thesis to break.

6. Guidance Raised Twice, and the New EPS Floor Is the Old Midpoint

Consolidated revenue growth went from 5% to 6% to about 6%. Adjusted diluted EPS growth went from 8% to 10% to 9% to 10%. Employer Services revenue growth went from 5% to 6% to about 6%. The client-funds balance assumption, the client-funds interest revenue guide, and the net-impact guide all went up. Adjusted EBIT margin expansion, the effective tax rate, the bookings guide, the retention guide, pays per control, and both PEO revenue guides were all left alone. The only line that came down was PEO worksite employees.

"Putting it all together, we are increasing our fiscal 2026 consolidated revenue outlook to about 6% growth, and we are maintaining our forecast for adjusted EBIT margin expansion of 50 to 70 basis points."
— Peter Hadley, CFO

The EPS raise is worth stating precisely. Off an FY25 base of $10.01, the October guide of 8% to 10% growth implied $10.81 to $11.01. The new guide of 9% to 10% implies $10.91 to $11.01. The floor of the new range is the midpoint of the old one, and management attributed the raise to share repurchases rather than to operations.

Assessment: Raising the low end while attributing it to buybacks is a conservative way to raise a guide, and it leaves the operational upside unspent. A company that has just delivered 79 basis points of margin against a 50 to 70 basis point full-year band, and has chosen not to touch that band, has given itself somewhere to go in April.

7. Bookings Broad-Based, International Rebounds, Still Unquantified

Employer Services new business bookings were described as solid and broad-based, with international, U.S. enterprise, and compliance growing fastest, and small business and mid-market contributing. International specifically rebounded after what management called a softer first quarter, and the quarter's flagship win was a European bank with more than 75,000 employees.

"During the second quarter, we won the business of a large European bank with more than 75,000 employees."
— Maria Black, President and CEO

The 4% to 7% full-year bookings guide was maintained for the second consecutive quarter, off an FY25 base of $2.1B. As in October, no quarterly bookings growth rate, dollar amount, or comparison base was disclosed. Management was also explicit that international is structurally lower margin than the domestic mid-market and down-market businesses, offset by very high retention that makes lifetime value comparable.

Assessment: Two quarters of qualitatively strong bookings with an unchanged guide and no disclosed number is now a pattern rather than an omission. The mix shift toward international and enterprise is strategically right and near-term margin-dilutive, which is a trade we would take, and which makes the unquantified bookings line more important to size, not less.

8. Lyric Moves Upmarket: Two 20,000-Employee Wins and 70% New Logos

The enterprise HCM platform produced the most concrete disclosure of the call. Lyric bookings again exceeded expectations, the pipeline continued to expand, and management put a number on the competitive dynamic for the first time.

"Underscoring Lyric's strong reception in the market, more than 70% of its new business bookings and overall pipeline related to new logos as it continues to fare favorably against our competitors."
— Maria Black, President and CEO

"Among our many Lyric new business wins in the second quarter were two companies with more than 20,000 employees, which represents two of our largest clients sold on the platform to date."
— Maria Black, President and CEO

New logos rather than base migration is the metric that distinguishes a product winning share from a product defending an installed base. Enterprise is historically where ADP has been most exposed to displacement, so 70% of both bookings and pipeline coming from clients ADP does not already have is a genuine change in the competitive read.

Assessment: This is the first Lyric disclosure with an actual number attached, and it is the right number. It still is not revenue, a client count, or a bookings dollar, and management again cautioned that large enterprise deals take time to convert to revenue. Treat it as evidence for the FY27-plus revenue bridge, not for FY26.

9. Workforce Now Next-Gen Crosses Into the Thousand-Employee Tier

The mid-market platform reached a milestone that matters for the addressable market rather than for this quarter's numbers: its first sale to a client with more than a thousand employees, a Midwest logistics company that bought payroll, HR, benefits administration, time and attendance, and learning together.

"We reached a milestone in the second quarter with our first sale to a client with more than a thousand employees."
— Maria Black, President and CEO

Last quarter's Next-Gen disclosure was that it had reached more than 80% of new core mid-market sales in the 50 to 150 employee band, with meaningfully lower client contacts than the current-generation platform. Extending the same architecture into a client ten times that size stretches the platform across a much wider band of the market.

"Workforce Now NextGen is being embraced by our mid-market clients for its always-on payroll processing capabilities, generative AI functionality, and expedited implementation timelines."
— Maria Black, President and CEO

Assessment: The cost-to-serve argument remains the real economic case and remains unquantified for a second quarter. What did change is the ceiling: a platform selling into the thousand-employee tier competes for a materially larger share of the mid-market than one capped at 150 employees, and it does so with the lower service cost structure. Still directional, still not modelable.

10. Embedded Payroll Finally Goes Live

The Fiserv partnership crossed from commitment to code. Cash Flow Central was embedded into the RUN small-business platform in December, delivering payroll, contractor payments, bill pay, and invoicing in a single interface. Last quarter management said this was slated for "the tail end of this year"; it arrived early.

"In December, we successfully embedded Fiserv's Cash Flow Central, an integrated accounts payables and receivables management solution, into RUN in order to help our small business clients better manage their cash flow."
— Maria Black, President and CEO

Management was equally clear about what it is worth today, which is nothing.

"That said though, we did just complete that integration in December, and so there is not a lot of contribution yet with respect to revenue and or bookings."
— Maria Black, President and CEO

Separately, the retirement services business launched its first pooled employer plan during the quarter, a structure that shifts fiduciary and administrative burden from small employers to ADP as pooled plan provider. Retirement and insurance services were also named as among the strongest down-market bookings contributors, which is the same small-business attach story from a different angle.

Assessment: Shipping on schedule after a partner's own restructuring is a mild positive on execution and changes nothing in FY26. The more interesting item is the pooled employer plan, because it monetises compliance burden rather than headcount, and headcount is the variable that is not growing.

11. Capital Return Steps Up and Cash Conversion Repairs

The board authorized a $6B repurchase programme in January, replacing the entire $5B authorization from 2022, and the dividend was raised 10%.

"Earlier this month, our board authorized the purchase of $6 billion of our common stock, which replaced in its entirety our 2022 authorization of $5 billion."
— Peter Hadley, CFO

Repurchases in the quarter were $515.8M against $366.0M in the September quarter, and dividends were $629.3M, together $1,145.1M or 108% of net earnings. That is a higher distribution ratio than last quarter, funded by materially better cash generation: operating cash flow was $1,132.9M against $1,150.3M a year ago, down 1.5%, after a September quarter that was down 22.1%. Distributions ran at 101% of operating cash flow against 155% in the prior quarter. The dividend declared and payable at December 31 was $680.5M, 8.1% above the $629.3M paid during the quarter.

Assessment: The cash-conversion concern we flagged in October has substantially resolved on a quarterly basis and has not fully resolved on a six-month one, where operating cash flow is still down 10.1% and distributions have run at 120% of it. A 20% larger buyback authorization at a share price 4% below the last print is good capital allocation. Watch the six-month gap close in the March quarter, which is seasonally the strongest for client funds and therefore for working capital.

12. AI in the Product, and the Labor-Market Question It Raises

ADP launched persona-based ADP Assist agents across payroll, HR, analytics, and tax during the quarter, including a tax registration agent that identifies missing or incomplete tax IDs and walks clients through remediation. The strategic argument is unchanged from October: proprietary payroll data across more than 1.1 million clients and 140-plus countries is the defensible input, not the model.

The more consequential AI discussion was about ADP's clients rather than its products. Asked whether the wave of enterprise layoff announcements citing AI was visible in ADP's own employment data, management said it was not.

"We are not really seeing anything discernible there."
— Peter Hadley, CFO

On its own cost base, management was explicit that AI is not reducing headcount and is not intended to.

"We remain committed to growing sales headcount."
— Peter Hadley, CFO

Assessment: ADP is the single best-positioned private observer of U.S. employment, and it is telling the market that AI-driven displacement is not yet visible by industry group, including in the verticals most commonly named as at risk. Take that seriously as a data point and note that it cuts both ways: the same dataset that says the disruption is not here is the dataset that will say it first when it arrives. On the internal side, the generative-AI efficiency claim that carried so much of the October guide is now partly redundant, because the margin arrived without needing it to be sized.

Guidance & Outlook

Six lines went up, two went down, eight were left alone. That is the inverse of October, where the operating metric came down and the float went up while the headline stayed put. This time the headline itself moved, on both revenue and earnings, and the float raise is a second-order contributor rather than the whole story. The GAAP diluted EPS growth guide moved with the adjusted one, from 8% to 10% up to 9% to 10%, so the two remain identical.

Management was unusually specific about the second-half shape, and the specificity is a warning as much as a reassurance.

"As we mentioned last quarter, we continue to expect a bit of a ramp in the back half of the year for margin expansion, and we currently expect to deliver more of this margin expansion in Q4 than in Q3."
— Peter Hadley, CFO

The reason is the float, and specifically the client-short portfolio. The March quarter is ADP's peak balance period because of bonus season and the annual reset of wage-base tax limits, so more of the portfolio sits in overnight instruments precisely when the year-over-year rate comparison is worst.

"So there is clearly about a 75 basis point reduction on the yield of that short portfolio in Q3 versus last year."
— Peter Hadley, CFO

Implied second-half ramp: At the raised revenue guide of about 6%, full-year revenue is roughly $21,795M and the implied second half is $11,260M, which needs 5.4% growth after 6.6% in the first half. At the 50 to 70 basis point margin band, second-half adjusted EBIT margin needs to be 27.2% to 27.6% against 26.6% a year ago, an expansion of 61 to 100 basis points against the 79 just delivered. At the 9% to 10% adjusted EPS band, second-half adjusted EPS needs to grow 9.0% to 10.9% against the 9.0% delivered in the first half. The guide is now a hold-the-line problem rather than the acceleration problem it was in October.

Street at: Consensus entered the print at $2.57 of adjusted EPS and $5.34B of revenue, and both cleared. The raise moves the full-year consensus band from $10.81 to $11.01 up to $10.91 to $11.01, a floor increase of roughly 1%. Because the margin guide was not touched, the earnings raise consensus will carry is smaller than the quarter's beat implies.

Guidance style: ADP raised the low end of the earnings guide and attributed the raise to buybacks, left the margin band untouched after beating it by nine basis points in the quarter, and cut a segment volume guide it could have narrowed instead. That is a management team resetting expectations downward on the metrics it does not control and refusing credit for the ones it does. Given the October experience, that is the right posture, and it means the reported second half is more likely to clear the guide than to test it.

Analyst Q&A Highlights

The Shape of the Second-Half Margin Ramp, and Why the March Quarter Is the Weaker One

The most useful exchange of the call, and a direct contrast with the October equivalent, when the same question produced the word "efficiencies" and nothing else. This time the answer named the mechanism, the portfolio, and the size of the rate move, and in doing so told the market to model the March quarter below the June quarter.

Q: "Just on the margin cadence, Peter, I think you know, you talked about this last quarter about it being more back-half weighted. Looked like Q2 was a little bit better than what we had modeled, including the higher float from the higher balances. So Q3 to Q4, any callouts in terms of step function change, and have you changed your investment approach given the higher float?"
— Tien-Tsin Huang, JPMorgan

A: "So the float portfolio in Q3 being calendar Q1 is our highest balance period where we have bonus season, we have tax rate, tax limits resetting. So we have more float basically in Q1, which results in more overnight balances. And this year versus last year, you know, as you know, we had a 75 basis point reduction in Fed funds between the same period last year and this year. So that creates a little bit of margin pressure in Q3 over Q3 last year relative."
— Peter Hadley, CFO

Assessment: A quantified, modelable answer where three months ago there was a framework. The disclosure is genuinely negative for the March quarter and positive for the credibility of the full-year band, because a management team hiding a soft quarter does not volunteer the portfolio, the season, and the basis points. Model the third quarter with less margin expansion than the second and the fourth with more.

Why PEO Revenue Excluding Pass-Throughs Halved Sequentially

The sharpest challenge of the call, aimed at a three-point deceleration that is larger than the segment's normal quarter-to-quarter noise. The answer decomposed it into three causes, only one of which is a demand signal, and the honesty of the decomposition is why the guide reiteration alongside it is credible.

Q: "Maybe you could help me understand the first quarter revenue ex pass-throughs grew at 6%, this quarter at 3%. That is a pretty big move or bigger move than usual we see between first and second quarter or just in the cadence of quarters. Is the 300 basis point delta there, maybe you could help us some of the drivers there?"
— Bryan Keane, Citi

A: "One is the slightly softer worksite employees we were talking about earlier, which came from, again, from a solid but slightly below our expectations booking performance and some moderation in pays per control. The second factor is you may recall, Q2 last year, we had a bunch of pull forward of SUI revenues that we would not last year, we were anticipating in the third quarter were pulled forward just due to the way the processing calendar worked in the second quarter."
— Peter Hadley, CFO

Assessment: Of the three factors, the state-unemployment-insurance pull-forward is a prior-year comparison artifact that reverses next quarter, and the wage-growth softness inside the PEO base is real but small. The bookings shortfall is the one that matters, because the entire worksite-employee guide is bookings-dependent by management's own framing from last quarter. Two of three reverse; the third is why the guide got cut.

Whether the Pays-Per-Control Pickup Is Broad-Based, and Why the Guide Did Not Move

An analyst pointed out the obvious asymmetry: the metric rounded down to zero last quarter and the guide was cut, then rounded up to one this quarter and the guide was not raised. The answer explained the rounding rather than defending the guide, and in doing so gave a more precise number than the guide contains.

Q: "So you just commented on the pickup here. I am curious if that was broad-based or if there were select contributors of that performance across certain client sizes? And as you just thought about the full year still roughly a flat outlook, last quarter you said you are rounding down to zero, here you are rounding up to one. Just curious how you thought about the second half, just given that pickup of trend."
— Bryan Bergin, TD Cowen

A: "In terms of the outlook, we had quite a lot of discussion about it. It is not an easy one to predict because we are really talking about we are very confident, I think, that we will continue to see growth. It is a question of is that growth just above or just below the half percent mark."
— Peter Hadley, CFO

Assessment: The underlying trend is roughly half a percent, which is why the reported figure can print 0% or 1% on a rounding decision. That is materially better than the October framing, in which management said it expected the metric to sit at zero for the balance of the fiscal year. A guide held at approximately flat against an underlying half point is cushion, and cushion is exactly what this guide did not have three months ago.

Whether AI-Driven Layoffs Are Visible in ADP's Own Client Data

The question the entire market wants answered, asked of the one company with the dataset to answer it, on a week thick with enterprise layoff announcements citing artificial intelligence. Management answered with its own industry-group cut of the client base rather than with commentary on the headlines.

Q: "Hate to ask a question on AI impacts on hiring, but just in the context of even over the last, you know, 24, 48 hours, seeing some incremental announcements from enterprises around layoffs and citing AI. I am just wondering if you have any updated views on that topic. You know, impacts that AI could be having on the broader labor market?"
— Scott Wurtzel, Wolfe Research

A: "But in terms of the data we look at, we look at it obviously very closely. We look at it by industry, about 10 or 12 industry groups. We are not really seeing anything discernible there."
— Peter Hadley, CFO

Assessment: Management went further in the same answer, naming financial services, professional services, and technology, the verticals most commonly cited as AI-exposed, as showing reasonably healthy growth, and closing that the empirical data does not point to displacement at this point in time. This is the strongest available evidence that the terminal-value bear case on payroll processors is a narrative rather than a trend, and it is being offered by the party with the most to lose from being wrong about it. Weight it accordingly and revisit every quarter.

What Pricing Is Actually Contributing

A question about the pricing environment produced the only quantified revenue driver management disclosed all call, and one it had declined to size in October beyond confirming the assumption was unchanged.

Q: "Could you comment on the pricing environment right now? How does it feel in terms of your ability to deploy pricing? And maybe what contribution are you expecting from that in your numbers?"
— Ramsey El-Assal, Cantor Fitzgerald

A: "And, again, we are not expecting anything to deviate from what we have said before, which is around 100 basis points of contribution from price in fiscal 2026, which is a little lower, not a huge amount of difference, but a little lower than what we had in fiscal 2025 and a little higher than sort of what we were doing pre-pandemic, which was more in the half a point range."
— Peter Hadley, CFO

Assessment: One point of price against roughly six points of reported revenue growth, feathered across the client base through the year rather than applied on a single date. Pricing is running above the pre-pandemic half point and below FY25, which brackets the pricing power question usefully: this is a business that can hold real price in a soft volume environment, but it is not currently using price to offset volume weakness. That restraint is a positive for retention and a constraint on the near-term revenue algorithm.

Record Client Satisfaction Against a Retention Guide That Still Declines

A well-aimed question about an internal inconsistency: management reported its best-ever quarter and best-ever first half for client satisfaction while guiding full-year retention down 10 to 30 basis points. The answer separated the service variable from the macro variable.

Q: "So if your customers really love the product and retention is coming in line, you know, any thoughts about sort of what is impacting the market in terms of exogenous factors from the macro or the competitive environment? Anything you would share on retention?"
— Daniel Jester, BMO Capital Markets

A: "That said, it was in line with our expectation, and that expectation is really how we set out the plan for the year, and Peter can comment on this as well, but we do anticipate a bit of a moderation."
— Maria Black, President and CEO

Assessment: The retention decline is a small-business out-of-business assumption normalising toward pre-pandemic levels, not competitive loss or service failure. Management said it saw a small amount of that contribute to the second-quarter decline and expects more in the back half, with the March quarter the most definitive read. Against a 92.1% base, 10 to 30 basis points is noise in absolute terms and signal about the small-business economy. The satisfaction record and the retention guide are consistent; the first is about ADP and the second is about its clients' survival rate.

Whether a Peer's Revenue-Per-Client Warning Reads Across

A read-through question testing whether a competitor's disclosure of lower revenue per client, which had been taken by parts of the market as a category-wide attach-rate problem, was visible in ADP's base. The answer was a flat denial, followed by a counter-example from the CFO.

Q: "Your peer talked about a lower revenue per client. I was just wondering if you have seen anything on that front. In terms of the number of products that are obtained by your clients."
— Ashish Sabadra, RBC Capital Markets

A: "I have not seen any of those trends. We, you know, monitor that closely, especially this time of year as we are looking at, you know, tremendous volumes, and we have not seen anything that would lead us to believe that there is a lower revenue per client or per client employee, if you will."
— Maria Black, President and CEO

Assessment: The CFO added that the strongest down-market bookings performance came in retirement and insurance services, which is attach rate going the other way. This is the most direct competitive-position datapoint on the call, and it matters because the entire payroll complex trades as one on labor data. If the category's attach-rate problem is not ADP's problem, the group-level de-rating that has taken this stock down 14.7% over twelve months is mispricing the leader alongside the laggards.

Lyric's Move Upmarket and Whether Large Deals Become Routine

A question on whether two wins above 20,000 employees signal that the enterprise platform has reached reference-customer critical mass, or whether they are one-off lumpiness. Management claimed the former on bookings and conceded the latter on revenue timing.

Q: "So maybe on Lyric, you know, it sounded like you sold a couple of quite large deals this quarter that you mentioned in the prepared remarks. Maybe can you share a little bit of color about how maybe you won those deals or how they came together? And as you think about the larger part of the enterprise for Lyric, do you have critical mass now in terms of reference customers or and are deals like this? Should we be seeing them more frequently, or maybe just any more color about the upmarket momentum on Lyric?"
— Daniel Jester, BMO Capital Markets

A: "These are large deals. They will take some time to onboard to get to huge revenue contribution. But definitely material bookings contribution from Lyric at this time."
— Maria Black, President and CEO

Assessment: "Material bookings contribution" is the first time the enterprise platform has been characterized as moving the consolidated bookings number rather than merely exceeding an internal plan. Combined with the 70% new-logo share, the competitive read on enterprise is better than at any point in this coverage. It remains a FY27-and-beyond revenue story, and it remains disclosed without a single dollar attached.

What They're NOT Saying

1. Any bookings number, for the second consecutive quarter. "Solid and broad-based" with the fastest growth in international, U.S. enterprise, and compliance, and the 4% to 7% full-year guide untouched. No percentage, no dollar, no comparison base. A guide that has now been maintained through two quarters of qualitatively strong bookings is either conservative or covering a range that the reported quarters are tracking below.
2. How far below plan the PEO bookings landed. "Slightly below our expectations" was said twice and quantified never, in the same quarter the worksite-employee guide was cut. The size of that miss is the single most useful number for judging whether the 3% to 5% ex-pass-through range holds.
3. The quarterly client revenue retention rate. Third consecutive quarter without one. Retention "declined modestly" and matched expectations, against an FY25 base of 92.1%, with the full-year 10 to 30 basis point decline guide maintained. The actual figure remains undisclosed.
4. Any Lyric revenue, client count, or bookings dollar. The new-logo share above 70% is a real advance on prior disclosure. It is also a ratio, and a ratio without a denominator cannot be modelled.
5. The Next-Gen cost-to-serve saving, again. Last quarter management said client contacts for Next-Gen clients were "meaningfully lower" than for the current-generation platform, which is the entire economic case for the mid-market migration. A quarter later, with a first thousand-employee win to point at, still no number, and still unasked.
6. What the generative-AI efficiencies actually contributed. They were named in October as a pillar of the back-half margin ramp. The ramp began, and the AI contribution was neither quantified nor mentioned as a driver of the 80 basis points, which management attributed to operating leverage, client-funds revenue, and the acquisition anniversary. Either the efficiencies are inside "operating leverage" or they have not yet arrived.
7. The Q3 float headwind in dollars. Management disclosed the mechanism, the portfolio, and a 75 basis point year-over-year Fed funds reduction, then told analysts to extrapolate it from the balances and rates in the release. Helpful, and stopping one step short of the number.
8. Whether the second-half PEO revenue guide implies the bottom of the range. Asked directly, management said not necessarily and that all possibilities remain live. The first half tracked at 4.6% ex-pass-throughs against a 3% to 5% full-year range, so the midpoint holds on a second half of about 3.5%, below the first half rather than above it. The number management would not supply is the one that decides it: whether the December quarter's 3.4% is the new run rate or the trough.
9. The six-month cash-flow gap. Operating cash flow is down 10.1% across the first half on net earnings up 8.1%, and distributions have run at 120% of it. The quarterly figure repaired to down 1.5%, which was not mentioned, and the six-month figure drew no question.
10. What a further decline in the forward curve does to the float. The blended new-purchase rate assumption has now fallen from 4.1% to 3.9% to 3.8% across three consecutive guidance dates. No sensitivity, no rate-shock scenario, no disclosure of the portfolio's reinvestment schedule beyond the balances and rates in the release.

Market Reaction

• Pre-print setup: ADP closed at $254.51 on January 27, down 1.1% year to date against an S&P 500 up 1.9%, down 2.0% over the trailing 30 days and down 14.7% over the trailing twelve months. The pre-print 52-week closing range was $249.06 to $326.81, so the stock entered the print roughly 2% above its own 52-week closing low.
• Reaction session (January 28, before the open): Opened at $249.65, a 1.9% gap down, traded a range of $248.30 to $255.73, and closed at $250.69, down 1.5% or $3.82. Pre-market indications had the stock down close to 3% before it recovered part of the gap.
• Volume: 4.1 million shares against a 30-day average of 2.2 million, roughly 1.8 times normal. Heavier than the October print in absolute terms and similar as a multiple.
• Peer reaction: Paychex fell 2.4%, Paycom fell 1.8%, Paylocity fell 1.0%, and TriNet fell 5.4% on the same session. Workday rose 0.3%. The S&P 500 finished flat.

The pattern repeated, at a quarter of the amplitude. None of the four payroll and PEO comparables reported that morning, none had company-specific news, and all four fell on a tape that closed unchanged. Three months ago the same group fell 5% to 7% on an ADP print that told the market same-store employment growth had gone to zero. This time ADP told the market it had gone back up, beat on both lines, and raised the guide twice, and the group still went down. That asymmetry is the most informative thing in this section.

Two readings are available and they are not mutually exclusive. The first is that the complex now trades on the PEO and worksite-employee data rather than on pays per control, and the worksite-employee guide cut plus the below-plan PEO bookings is the read-through that mattered. The heaviest decline in the group belonged to the pure-play PEO, which is consistent with that reading. The second is that the group is in a multi-month de-rating that individual prints no longer interrupt: ADP has fallen 14.7% over twelve months and 1.1% year to date against an S&P up 1.9% year to date, and a beat-and-raise producing a 1.5% decline is what a de-rating in progress looks like.

What the price action is not is a verdict on the quarter. The stock closed the session at 22.9 times the midpoint of a guide management had just raised, against 23.9 times the midpoint of the prior guide on the day of the October print. The multiple compressed while the earnings estimate went up, on a quarter that resolved the largest open question in the outlook. Positioning explains that; fundamentals do not.

Street Perspective

Debate: Was October a Re-Rating or an Overshoot?

Bull view: The bull case being made on the Street is that October priced a permanent impairment to the growth algorithm off a single quarter's rounding, and this print refutes it. Pays per control came back, margin expanded 80 basis points, the guide went up twice, and the underlying employment trend was disclosed as roughly half a percent rather than zero. A stock that has de-rated from the low thirties to under twenty-three times forward on an unbroken 92%-retention franchise is being priced as if the labor cycle is a structural break.

Bear view: The bear camp contends that the market is correctly pricing a business whose revenue algorithm has permanently slowed. Organic constant-currency growth was 5% with a point of currency help that reverses, PEO worksite-employee growth was cut, PEO bookings missed plan, and the float's growth contribution decays from here on hardening comparisons. On this reading the multiple compression is the market marking a mid-single-digit grower down from a high-single-digit grower's multiple, and there is further to go.

Our take: The bulls have the better of it now, which is a change from our October position. The specific thing that changed is the conversion: ex-float pretax earnings growth went from 1.1% to 6.4% on similar ex-float revenue growth, which is the difference between a business that has lost its operating leverage and one that had a quarter of acquisition drag. The bear case's revenue points are all correct and all already in a 22.9 times multiple. The bear case's earnings-quality point, which was the strongest argument three months ago, is substantially weaker after this print.

Debate: Does the Float Tailwind Survive the Rate Path?

Bull view: The optimistic framing has strengthened on the volume side. Management raised the balance-growth assumption to 4% to 5% from 3% to 4%, which reverses the wage-moderation assumption embedded in the October guide, and raised both dollar guides. The ladder continues to work: client extended yields rose 70 basis points year over year and client long rose 10, because older low-coupon paper keeps rolling into higher-coupon paper. Falling short rates cut the financing cost of the strategy by $8.7M in the quarter, a natural hedge against the client-short yield decline.

Bear view: The skeptical framing points at the client-short portfolio, where the yield fell 130 basis points year over year to 3.3%, and at the blended forward new-purchase rate, which has now declined at two consecutive guidance dates to 3.8%. Management has explicitly guided the March quarter, its largest balance quarter, to weaker margin expansion than the June quarter because of exactly this. And the composite's growth contribution mechanically decays: $237.8M to $294.3M is a 24% increase, and the full-year guide of $1,270M to $1,290M against FY25's $1,073M implies the year's growth is front-loaded.

Our take: Both sides are right and the question has become less important, which is the whole point. When the float supplied 84% of pretax earnings growth, the rate path was the thesis. At 46%, it is one of three drivers. We continue to model the float conservatively, at the low-to-middle of the raised guides, and we no longer regard a further decline in the forward curve as thesis-breaking. It is a headwind to the growth rate that the operating business has demonstrated it can now partly absorb.

Debate: Is the Whole Payroll Complex Correctly Priced as One?

Bull view: The differentiation case is that ADP is being dragged down by a group it does not belong to. It is the only member of the complex with a $37.6B float, the only one with a 92%-plus retention base spanning 1.1 million clients across 140-plus countries, and it explicitly denied the attach-rate deterioration a competitor disclosed. Its down-market bookings strength came in retirement and insurance services, which is attach rate expanding. On this view the correlated selling is an indexing artifact and the leader deserves a premium that has vanished.

Bear view: The bear camp argues the correlation is fundamental, not technical. Every company in the group monetises the same variable, which is the number of people employed in the United States, and every one of them is telling the market that variable is flat to marginally positive. ADP's own guide of approximately flat pays per control for the full year is the group's problem stated in ADP's language. Differentiation in product and retention does not change the fact that the volume driver is shared, and the multiple should compress across the group until it turns.

Our take: The bears are right about the driver and wrong about the dispersion. Employment is the shared variable and it is not growing, which is why we are not modelling a pays-per-control recovery. But ADP converts that flat variable into 6% revenue growth and 11% EPS growth through pricing, bookings, mix, the float, and a shrinking share count, and no other member of the complex has all five levers. Selling the leader at the same multiple compression as a pure-play PEO whose stock fell 5.4% on someone else's earnings is the mispricing we are underwriting with this upgrade.

Model Update

The table below revises the assumptions we set at initiation three months ago against management's newly raised guide. Eight of the eleven lines move up, one moves down, one is unchanged, and the PEO ex-pass-through line is modelled separately for the first time.

Valuation framework: At the January 28 close of $250.69, ADP trades at 22.9 times the midpoint of management's raised FY26 adjusted EPS guide of $10.91 to $11.01, against 23.2 times at the pre-print close and 23.9 times the then-guide midpoint at the October reaction close. On the completed FY25 adjusted EPS of $10.01 the same close is 25.0 times, against a pre-print 52-week closing range spanning 24.9 to 32.6 times on that identical denominator. The stock has now spent a full quarter compressing while the earnings estimate rose. Market capitalization at that close is approximately $101.0B on the 403.0 million shares outstanding at December 31, and the $680.5M dividend declared and payable at the quarter end annualizes to a yield of roughly 2.7%.

Risk-reward: Base case upside of 9.3% plus a 2.7% dividend against bear case downside of 9.8% is a materially better skew than the 5.3% up against 12.0% down we published in October, and the improvement came from both directions: the price fell 4.0% and the earnings estimate rose. The base-case multiple of 25.0 times assumes the stock recovers roughly to where it traded before the October print on a guide that is now higher, which requires no re-rating beyond the reversal of an overshoot. The bull case requires the second half to clear the top of the margin band and pays per control to hold at a reported 1%. The bear case requires the PEO deceleration to spread to Employer Services bookings and the forward curve to keep falling, which would put FY27 growth in the mid single digits and justify a low-twenties multiple.

Thesis Scorecard Post-Earnings

The six pillars below are the ones established at initiation in October, graded against what this quarter's print and call revealed. Three status tags moved.

Overall: Thesis strengthened. All three bear points established at initiation were challenged by this print, one bull pillar upgraded, and a new, narrower bear point opened in the PEO. The October thesis was that ADP's FY26 earnings growth was being supplied by the client-funds float rather than by the operating business at a multiple that had de-rated to fair rather than to cheap. The first half of that sentence is no longer true: the operating business produced 6.4% ex-float pretax earnings growth and 79 basis points of margin expansion. The second half has become more favorable, because the price fell while the guide rose.

Action: Upgrade to Outperform. The trigger we published in October was a second-quarter print showing year-over-year adjusted EBIT margin expansion, and it arrived at 80 basis points alongside a pays-per-control recovery, a twice-raised guide, and a larger buyback authorization. At 22.9 times the midpoint of that raised guide, with a 2.7% dividend and a share count shrinking 1.1% a year, the risk-reward has inverted from the fully-valued setup we described three months ago. What would break the call is the PEO deceleration spreading to Employer Services bookings, or a second-half margin print that comes in below the 50 basis point floor of the band management chose not to raise.