{{ v.title }}
{{ v.body }}
{{ v.cta }}{{ announce }}
Your firm sells time. AI compresses time.
Maybank Advisors rebuilds how professional services firms produce and price their work, so the efficiency can become margin, capacity or deliberate pricing—not an accidental client discount. Fixed fee by phase, never hourly. Your prompt libraries, review standards and documentation sit in your own systems and keep working without us.
Watch the 18-minute briefing
18 minutes · Four chapters · Every figure cited · Worksheet included
Cost of delivery is already falling in firms that have put AI into production work. Fees have not moved with it. The difference between the two lines is margin, and it exists only until clients, competitors or renewals reset the price.
Margin you can keep
while fees hold and cost falls
TodayMonth 12Month 18–24
What clients pay you
What the work costs you
Clients reprice
The shaded area is the margin available while fees hold and cost falls.
It is open now, and it closes without anyone deciding to close it. Repricing arrives through competitive bids, renewal negotiations and procurement benchmarks, and when it does, fees step down to meet cost. Whatever a firm has not deliberately claimed by then goes to its clients.
Deciding what that margin becomes — and rebuilding delivery and pricing to hold it — is the work we do.
- Founded 2021
- By operators, not consultants
- Fixed fee by phase, never hourly
- Quoted in writing before we start
- Your team keeps running it
- Built and held in your own systems
- Engineering in the group
- Pyxl, our sister company, builds when advice has to become software
The Window Is Closing
18–24 months, on our estimate
Our working estimate is 18–24 months from now, with shorter timing possible in more competitive sectors. The thesis is simple: a firm can become materially more efficient before clients fully reprice that efficiency. The question is what you do with the gap while it exists.
White paper
The Window Is Closing
Why we estimate professional services firms have roughly 18–24 months to turn AI efficiency into cash, what could shorten that period, and what a firm can do before repricing catches up.
- The four reasons clients have not repriced yet
- What the window is worth on a $10M illustrative fee base
- What the captured cash can fund
- A 90-day sequence, not a two-year plan
Free, no email required. Read it in the browser or print it.
What the period costs if it goes unused
What happens to a firm that spends the period deciding
The cost is not simply being late to a tool. It is spending the period when delivery can get more efficient before the commercial model catches up.
-
{{ c.when }}
{{ c.title }}
{{ c.body }}
Illustrative sequence · the mechanisms we see, not a forecast of dates
Where that leaves the firm
A firm that reaches the end of this period with no meaningful efficiency gain, no deliberate repricing position and no revenue strategy beyond billable time is more exposed than it was at the start. The question is whether the firm used the period to build options.
Build time against the period
Every one of the four things this cash can buy — a productized service line, an owned data asset, a repricing transition, an acquisition — takes eighteen months to two years to stand up.
So the time available to build is the period minus however long the decision takes. On an eighteen-month estimate, a six-month decision leaves twelve months for a build that needs eighteen.
See where you stand →Approach
Start where the work already repeats
Most AI programs stall because they begin with tools instead of tasks. We begin with the work your teams do every week.
-
{{ s.num }}
{{ s.title }}
{{ s.body }}
Sectors
Start with your own sector
The economic exposure travels across professional services. What changes by sector is where the work repeats, what the client agreements permit, what quality means, and where the first commercial decision belongs.
Services
What we do for your firm
Six practices, sequenced from your delivery model rather than imposed as a package. Every engagement ends with something your firm can keep running without us.
All services →Pyxl · internal operating evidence
We ran the operating model on a real P&L first.
Before Maybank took this model to other professional services firms, we applied it inside Pyxl. We built Pyxl Intelligence there ourselves, and it changed both the work itself and the operating system around it — reducing delivery effort, coordination, administrative handoffs and recurring work that had been rebuilt manually.
From August 2025 through Q2 2026, gross margin expanded 1,300 basis points and EBITDA margin 1,800 basis points (13 and 18 percentage points).
The first 1,300 basis points reflect stronger delivery economics and, because delivery cost sits in COGS, flow through to EBITDA. The remaining 500 basis points or so reflect further operating leverage below gross profit.
That work was paid for once, by us. What a client engages is the method and the platform that came out of it, configured to their firm — not a rebuild of the 18 months it took to get here.
Read the Pyxl case study →- +1,300 bps
- Gross-margin expansion
- +1,800 bps
- EBITDA-margin expansion
- ~9 mo
- Operating period
Internal Pyxl operating result. One company, not a forecast or promise of results for another firm.
The finance view
Alban Howorth
Controller, Pyxl
Founder, Howorth Accounting
Pyxl Intelligence is the operating-intelligence system Maybank built inside Pyxl.
“As Pyxl’s long-time Controller, I see the impact of AI through Pyxl Intelligence in the financial results, not just in anecdotal productivity gains. Over the past nine months, Pyxl has expanded gross margin by 13 percentage points and EBITDA margin by 18 percentage points. The gross-margin improvement reflects stronger delivery economics, and that gain flowed through to EBITDA. EBITDA expanded further as the company also generated additional operating leverage below gross profit. From a finance perspective, this is a meaningful improvement in the operating model, not simply a technology-efficiency story.”
Executive briefing · 18 minutes
AI is compressing professional services work. Who captures the difference?
Research + Pyxl operating case · captions and transcript available
Executive briefing · 18 minutes
AI is compressing professional services work. Who captures the difference?
An evidence-led briefing on margin, capacity, pricing and the operating model required to turn AI productivity into firm economics.
Research from HBS/BCG, Microsoft Research and Thomson Reuters, plus the measured operating experience of Pyxl, an affiliated professional services firm.
What you will learn
Why task productivity and firm economics are different.
What changed inside Pyxl—and what showed up in gross margin and EBITDA.
How to determine whether one service line inside your firm is worth assessing.
- 18 minutes
- Four chapters
- Sources included
- Worksheet included
- No call required
AI assessment
Find out where your firm actually stands
Ten questions across delivery, data, tooling and governance. You get a scored readout of where your model is ready, where the binding constraint sits, and the two or three moves that matter most.
- Three minutes · Ten questions · No prep
- Results on screen
- No call required
Sample readout
61
out of 100
Ready in parts
Delivery model78
Data and confidentiality44
Tooling and access69
Governance and people52
Strongest: the work repeats enough to systematize.
Weakest: nobody owns the permitted-use position.
Sample report · your score, four dimensions and three prioritized moves
Engagements and fees
What this costs
Fixed fee by phase. The assessment is $4,500, credited in full against a build started within 90 days. The build phase is scoped per firm and quoted in writing after the assessment. Governance is $2,500/month and optional.
{{ t.kind }}
{{ t.name }}
{{ t.price }}
{{ t.duration }}
{{ t.body }}
{{ t.cta }} →Ranges are indicative and depend on scope, number of teams and markets. Every engagement is quoted in writing before it starts.
Proof
What changes when the work is rebuilt
The Pyxl case is a named affiliated-company operating case with approved results. The other sector examples are composites of recurring engagement patterns until founding-cohort clients agree to be named, and are not presented as attributed client case studies.
Tools and white papers
Take something useful with you
Practical material from the work: The Window Is Closing, the Engagement Margin Model, the Prompt Library Playbook, Governance for Professional Practices and the Prompt Review Checklist.
{{ r.kind }}
{{ r.title }}
{{ r.body }}
{{ r.cta }} →Insights
Writing from the work
Short arguments and working models on where AI changes delivery, margin, governance and the commercial model.
The same engagement, rebuilt
Hours you spend on it−25%
Fee you charge for itUnchanged
Illustrative · the difference is a fee decision, not a technology one
Fees
What a reviewed prompt library does to your rate card
Efficiency gains do not have to become client discounts. Where the value lands is a fee decision.
Model it on your numbers →What your client agreements permit
What associates assume is allowedMost
What the contracts actually permitSome
What partners assume is allowedLittle
Directional · the gap, not the quantity, is the point
Governance
Confidentiality standards that hold across offices
Local workarounds are the real risk. A single standard, written down, is cheaper than remediation.
Read the white paper →Two ways firms structure this
Steering committee, no budget lineStalls
Named owner with budget authorityShips
Maybank method · the structure decides the outcome
Operating model
Who owns artificial intelligence inside a firm
A steering committee without a budget line tends to stall. A named owner with budget authority tends to ship.
Read the playbook →Common questions
Before you call
If your question is not here, we will answer it directly.
Ask us directly →{{ f.a }}
Contact
Let's talk about your firm
The fastest route is the assessment or a 45-minute partner review. If you would rather write first, the form works too.
Thank you, we have your inquiry
We respond within one business day.
We could not send that
Nothing you typed has been lost. Try again, or open a prefilled email and we will pick it up from there.
We could not confirm that was received
It may have reached us and it may not. Rather than tell you it worked when we do not know, we would rather you sent it again or emailed us.
Prefer email? Open a prefilled message instead.