Most of the work costing you money shouldn't exist.
Sometimes it's a platform that has become expensive to change. Sometimes it's five people retyping the same information into three systems. The symptoms look nothing alike. The cause is usually the same: nobody has been paid to think about the whole thing at once.
That's what I do, and have for twenty years across healthcare, retail, media, and enterprise platforms. Two paths, depending on the shape of your problem.
Technology & Architecture Advisory
For organizations with real systems, real constraints, and decisions that are expensive to get wrong.
You have a platform that works but is getting harder to change. Teams are stepping on each other. An acquisition left you with two of everything. Someone needs to decide where AI actually belongs before the board asks again. This is that work.
Architecture review
An outside read on where your system is fragile, what will break next, and which fixes are worth the disruption.
AI strategy
Where AI creates leverage in your business and where it is a distraction. Evaluation criteria, build-vs-buy, and a sequence that compounds.
Platform modernization
Moving off what you have without stopping the business. Sequencing, strangler patterns, and the migrations that are safe to defer.
Integration architecture
System boundaries, API contracts, and durable synchronization between services that were never designed to talk to each other.
Technical leadership
Fractional architecture leadership for teams that need senior judgment without a full-time hire. Decision support, design review, and mentorship.
AI Automation for Growing Businesses
For owner-led businesses spending real money on work that software should be doing.
Your team is retyping the same information into three systems. Quotes take two days because someone has to chase them. Scheduling is a phone tree. None of that is a technology problem you should have to think about — it is work that should not exist, and most of it can be removed in weeks, not quarters.
Workflow automation
Intake, scheduling, quoting, follow-up, and reporting — connected so information is entered once and moves on its own.
AI-assisted operations
Call summaries, document handling, triage, and drafting, applied where they save hours rather than where they demo well.
Systems integration
Making your CRM, scheduling, accounting, and phone systems behave like one system instead of five.
Custom internal tools
The small application that replaces a spreadsheet everyone is afraid of, built to fit how your business actually runs.
- Trades and home services
- Restaurants and hospitality
- Medical and dental offices
- Spas and wellness
- Marketing and creative agencies
- Professional services
How engagements work.
No open-ended hourly arrangements. Every step has a defined end and something you keep.
Steps 01 to 03 commit neither of us to anything. Paid work begins at step 04, under a written agreement — the terms spell that out.
- 01Free · 30–45 minutes
Conversation
A call. You describe the problem and the outcome you need. I tell you whether I am the right person, and if I am not, who might be.
- 021–3 weeks
Assessment
A scoped, paid engagement where I go deep enough to be useful — reading the system, talking to the people who run it, and finding where the real constraint is.
- 03Delivered at assessment close
Roadmap
A written recommendation you own: what to do, in what order, what it will cost you to defer, and what I would do differently if I were wrong.
- 04Before work begins
Agreement
Scope, commercial terms, responsibilities, and what done looks like, confirmed in writing before any paid work starts — a proposal, a statement of work, or a services agreement, depending on the shape of the engagement. Nothing before this point is a commitment on either side.
- 05Ongoing, by agreement
Build or advise
Either I help execute the roadmap, or your team does and I stay on for design review and decision support. Both are legitimate. Neither requires the other.
Whether this is a fit.
Both lists are honest. The second one has saved us both time more than once.
This works when
- You have a decision to make and the cost of getting it wrong is high.
- You want a recommendation you can act on, not a deck.
- You are willing to hear that the problem is organizational rather than technical.
- You want your team stronger at the end of the engagement than at the start.
This doesn't
- You need staff augmentation — a pair of hands on a backlog someone else scoped.
- The decision is already made and you want validation for it.
- You want the lowest hourly rate available.
- The timeline requires skipping the part where I understand your business.
Before you ask.
- How do you price engagements?
- By scope, not by hour. Assessments are fixed-price against a defined deliverable so you know the cost before we start. Ongoing advisory runs on a monthly retainer with a clear scope and an exit. I do not sell hourly blocks — it puts my incentive and yours on opposite sides.
- How long is a typical engagement?
- Assessments run one to three weeks. Advisory retainers usually run three to six months, which is about how long it takes for a change in how decisions get made to become permanent. Automation projects for smaller businesses are often measured in weeks.
- Do you write code, or only advise?
- Both, depending on what the engagement needs. I have spent twenty years in codebases and I stay in them — architecture that has not been checked against the actual code is fiction. For automation work I usually build. For enterprise advisory I more often design and review while your team builds.
- We're a small business. Is this overkill?
- No, but the second path is the right one. The automation work is deliberately scoped for owner-led businesses: shorter engagements, faster payback, and no architecture vocabulary required. The goal is that your team stops doing something by hand, not that you learn what a system boundary is.
- What do you need from us to start?
- Access to the people who actually run the process, and honesty about constraints — budget, politics, prior decisions you are not willing to revisit. I would rather know the real boundaries on day one than discover them in week three.
- What if we already know what we want built?
- Then say so and we can skip the assessment. I will still ask what outcome the build is meant to produce, because that is the question that occasionally changes what gets built. If the answer holds up, we get to work.
Start with the problem.
Describe what's happening and what you need to be true instead. If I'm not the right person, I'll say so and point you somewhere better.
Or email directly: consulting@markfasel.dev