Put AI to work on the work that slows you down.

We help business owners automate repetitive work, get reliable numbers, and connect disconnected systems. Chris builds the technology. Rick brings the financial and operational discipline. Together, we build the solution and help your team use it.

Tell us what needs fixing See what we’ve built

Work directly with Chris and Rick. Monthly engagements from $2,800.

Illustrative
Daily scorecard · published 7:00a
$18.4kSales yesterday
212Orders
$2.1kAd spend
−$640Refunds netted
Sales · last 14 daysYesterday
Consults booked9 today
Ad spend as % of sales11.4%
Sources refreshed overnight12 of 12
Illustrative example. A daily scorecard like the one we built for a wellness brand: sales, ad spend, refunds, and bookings assembled overnight from 12+ sources. The morning starts with one screen instead of six tabs.
Featured work

Real problems. Working solutions.

A few examples of what we’ve built inside operating businesses. Your problem doesn’t have to look exactly like these.

Daily reporting

Start the day with numbers you can trust.

The problem

A wellness brand started every morning copying figures out of six tabs and still couldn’t tell which number was right.

The solution

A nightly pipeline pulls store, ads, refunds, and bookings into one warehouse and publishes a single scorecard by 7am.

16+ scattered reports became one screen, fed by 12+ live sources.

See the details

The scorecard shows sales, ad spend, refunds netted out, and consults booked. Nobody assembles anything; it is simply there when the team arrives. Each source is refreshed overnight and the report shows which ones loaded, so a stale feed is visible instead of silently repeating yesterday’s number.

Illustrative
Daily scorecard · auto-published 7:00a
$18.4kSales
212Units
$2.1kAd spend
Last 7 daysYesterday
Customer-facing AI

Help customers find the right product.

The problem

A B2B retailer’s support team answered the same thirty compatibility questions all day, while buyers who couldn’t wait simply left.

The solution

An AI agent in the storefront answers from the live product catalog, asks the questions a good salesperson would, and hands off to a person when it should.

Answers come from real product data, and the team reviews after launch what it deflected and what it got wrong.

See the details

The agent reads the same catalog the store sells from, so compatibility answers reflect current products rather than a script. When a question needs judgment, volume pricing, or a human relationship, it says so and hands the conversation to the team.

Illustrative
VisitorNeed wireless headsets that work with our Yealink phones. Which ones?
AgentThree models fit Yealink out of the box. How many desks, and is the office open-plan? That changes my pick.
Visitor12 desks, open plan.
AgentThen noise-canceling mics matter. Here’s the one I’d start with, and I can loop in a person for volume pricing.
Financial reconciliation

Catch the errors between your store and your books.

The problem

A brand that sells and rents a device had three versions of every month, and the order feed between two of its systems died quietly for five weeks.

The solution

A monthly tie-out reconciles charges, refunds, returns, and rental buyouts against the operations system order by order, plus a daily sync that fails loudly.

Two closes in a row: 3 dropped sales, $545 of refunds counted as revenue, 11 rentals booked in the wrong month, and a feed dead for five weeks. All found and fixed.

See the details

The store, the operations system, and the bookkeeper’s books never quite agreed, and each close was a week of “why is there an extra $600?” The tie-out now says exactly where and why they differ. It also produces a payout-by-product report the accountant had been told the platform couldn’t provide.

Illustrative
Month-end tie-out · June
Store charges vs ops system49 = 49
#4471 · charged 6/17, booked 6/18timezone
11 rentals · charged June, booked Julymoved
#4556 · charged + refunded same daynot revenue
Order feed · last run 6/23dead 35 days
Daily sync · 5:20aarmed
About

The people doing the work.

You work directly with Chris and Rick, from figuring out the problem to getting the solution running.

Chris Bentley

AI & Digital Operations

Chris spent 14+ years running e-commerce as a VP and P&L owner before building systems full time. He builds what you see on this page: reporting pipelines, AI agents, internal tools, and automations. For clients, he scopes the fix, builds it, tests it against real data, and keeps it monitored after launch.

→ LinkedIn

Rick Mills

Finance & Operations

Rick spent 20 years as an e-commerce CFO, owning the close, forecasting, and the reporting a leadership team runs on. For clients, he makes sure a new system produces numbers you can act on, keeps the scope tied to a business outcome, and brings the operating discipline that keeps growth from outrunning the business.

→ LinkedIn
More work

More problems we’ve helped solve.

Each of these runs inside a real company today. Open one for the full story.

Competitor price monitoring

Pricing intelligence across the whole market

A scheduled scraper watches thousands of competitor prices and saves a screenshot for every one it reports.
The problem

An industrial distributor priced thousands of SKUs against competitors who changed prices silently. Checking by hand never happened.

What we built

A scheduled scraper that walks competitor sites on its own. Validation checks reject a page that doesn’t parse as expected, so a broken page is flagged instead of reported as a price, and a screenshot is saved as evidence for every price that is reported. The pricing team opens a dashboard, not forty tabs.

Every reported price carries its own evidence capture.

Illustrative
Competitor sweep · last run 4:30a
7100 Series · rollrival −8%
Safety-Walk 24"we win
Bond 847 · caseno change
Grip tape · bulkrival −3%
Edge trim 50ftwe win
Order and fulfillment systems

A CRM rebuilt around how the team works

One CRM holding 10,000+ orders, with the ship queue the warehouse actually runs from.
The problem

Orders lived in one system, fulfillment in another, and the team lived in spreadsheets between them. Nothing agreed with anything.

What we built

One CRM holding 10,000+ unified orders, with a ship queue the warehouse runs from, address editing that syncs back to the carrier, and label printing built in. The spreadsheets were retired.

10,000+ orders unified. The whole team runs on one live screen.

Illustrative
Ship queue · today
#21391 · 2 itemslabel printed
#21392 · 1 itemlabel printed
#21393 · 3 itemsaddress fix
#21394 · 1 itemqueued
Monthly financial dashboards

Month-end numbers the whole team reads

The month-end close lands as an interactive dashboard instead of a long email.
The problem

Every month the financials went out as a long, text-heavy email. Hours to write, skimmed in seconds, and no way to ask “why is that up?” without starting a reply-all thread.

What we built

The close now produces an interactive dashboard instead of a wall of text: revenue, margin, cash, and spend against plan, with the trend behind every number one tap away. It assembles itself from the same figures as before; the finance lead reviews it and hits send. It is a single self-contained file that travels inside the company’s own email and access controls, with no new service touching the books.

One of us wrote that email for years. Now the month lands as a dashboard, and the questions arrive pre-answered.

Illustrative
Monthly financials · August close
+6.8%Revenue vs plan
41.2%Gross margin
+$14kOp-ex vs plan
Revenue · last 7 monthsAugust
Pre-consultation briefs

The consultant walks in already briefed

AI writes a one-page brief before every consult from the survey, the booking, and the order history.
The problem

A brand runs video consults all day, and every call started cold. Practitioners had two minutes to skim raw survey answers across three generations of intake forms, often in three different places. Whether the person ever bought was anyone’s guess.

What we built

One record per prospect: store survey, booking slot, order history, and any earlier call transcript. Before each consult, AI turns that into a one-page brief covering what hurts, what they’ve tried, and what to ask. After the call the transcript files itself, and the outcome (bought, rented, returned) updates from the order system with no one typing it.

598 surveys from three forms, one record each. Every consult opens with a brief already written.

Illustrative
Pre-consult brief · 2:00p today
Right elbow · 8 months, worse with grippain 7/10
Tried: PT, brace, ice, one injectionspend $500–1k
Tennis 3×/wk, wants to keep playingmotivated
Ask about: the injection last springfrom survey
2 surveys · booked 2:00p · no order yetbrief ready
Product catalog editing

A product editor the client team runs themselves

An embedded editor lets the team change 3,000+ products without a developer ticket.
The problem

Every catalog change, whether a spec, a filter, or a description across 3,000+ products, was a developer ticket with a two-week fuse.

What we built

An embedded editor inside their own store admin: search, stack filters, bulk-edit, apply. The team ships its own catalog changes now, and the dev queue lost its longest line.

3,000+ products, edited in-house, without a ticket.

Illustrative
Bulk edit · 47 products selected
☑ 7100-BLK · anti-slip rollupdated
☑ 7100-YLW · anti-slip rollupdated
☑ 7150-GRY · tread 6"updated
☑ 44 more…applying
Amazon account operations

The Amazon console that replaced an agency

An eight-module console wired to Amazon’s own APIs lets one in-house operator run the account.
The problem

A brand paid an agency to manage Amazon and still couldn’t answer basic questions about its own account, listings, or ad spend.

What we built

An eight-module operating console covering sales, ads, listings, inventory, and reclassification, wired to Amazon’s own APIs. One in-house operator now runs the whole account, and the agency was retired.

8 modules, one operator, one less agency invoice.

Illustrative
Console · this week
+12%Amazon sales
4.2xAd ROAS
3Low stock
Sales · last 7 days
Executive AI

An AI chief of staff for a CEO

An always-on assistant a CEO calls or messages, in daily use since February 2026.
The problem

A CEO wanted a personal AI assistant that just works, on his phone, every day, with no setup or maintenance on his side.

What we built

An always-on assistant on dedicated hardware he owns, reachable by voice, text, or browser. It drafts messages and sends them only after he approves the wording, runs hourly self-checks, and lets him switch features off and back on by asking. In daily use since February 2026.

The CEO uses it every day and never sees the machinery.

Illustrative
CEO · 6:12awhat’s on deck today?
Assistant3 meetings, the board deck draft is ready for you, and your 2pm asked to move. Want me to offer 4pm?
CEOyes. and remind me about the lease thing thursday
AssistantDone and done.
See how Smarty works →
Smarty

Meet Smarty. An assistant built around your day.

Email, calendar, reminders, and research, brought together in an assistant you can call or message.

  • Catch up on important emails.Ask what needs you today and get the short version, with the sender and the ask.
  • Draft messages and manage calendar changes.Smarty drafts the reply or the reschedule, reads it back, and sends only when you say so.
  • Get briefings and follow-up reminders.A morning briefing, an afternoon list of open items, and reminders that fire on your local clock.
See how Smarty works
Voice call · 12:24pIllustrative
YouFind David’s cell and text him I’m running ten minutes late.
SmartyFound him. The text reads: “Running about ten minutes late, see you shortly.” Send it?
YouSend it. And read me the email from the accountant.
SmartySent. The accountant’s email is from 9:40 this morning. She needs one form signed by Friday. Want me to draft the reply?

Illustrative conversation, written to show how Smarty’s real features work.

How it works

From recurring problem to working solution.

Step 1

Show us the problem

Tell us what takes too long, breaks too often, or leaves you guessing.

Step 2

Agree on the first useful fix

We define the scope, approach, and cost before the work begins.

Step 3

Put it to work with your team

We build, test, and help your team use the solution, with ongoing support defined in your engagement.

Pricing

Choose the support your business needs.

Both options are month to month, with hours reserved for your work each month.

Focused support
$2,800per month
15 hours reserved each month.

For one priority at a time: one system built, tested, and kept running.

Discuss your priorities
More capacity
$5,000per month
30 hours reserved each month.

For several active priorities, or one larger implementation you want finished sooner.

Discuss your priorities
Need more hours?

Additional hours are $160 per hour, when available.

Doesn’t fit a monthly rhythm?

We can price the work as a project instead.

  • No setup fee and no onboarding fee
  • Runs in your accounts or ours, your choice. The code is yours either way.
  • Monitored after launch, not handed over and forgotten
  • Month to month, no long-term contract
Questions

Before we get started.

What does fractional mean?

You get experienced operators part time, without a full-time hire. You pay for the specific work you need instead of a salary and a hiring cycle. We do this for AI systems, operations, and finance.

What can you help us with?

Repetitive work that should run on its own, numbers you can’t fully trust, and systems that don’t talk to each other. The examples above are the shape of it: reporting, customer-facing AI, reconciliation, order systems, and internal tools your team can run.

Can you train our existing team?

Yes. Helping your team use the solution is part of every build, and we design tools so your people can operate them without a developer on call.

How long does a first project take?

It depends on what’s broken. Some of the systems on this page were useful within days; others took months of iteration. We’ll tell you which yours looks like, along with the scope and cost, before you commit to anything.

Who owns the code and systems?

You do. Every system runs in your accounts or ours, whichever you prefer, and the code and data are yours either way.

Are software and hosting costs additional?

The monthly fee covers our time. Before work begins we list any third-party subscriptions, hosting, or AI usage a solution needs and who pays for them, so there are no surprises on the first invoice.

What happens when something breaks?

Tell us, and we fix it. Systems we build during an engagement are monitored after launch, and fixes come out of your reserved hours. Where it matters, we build a way to roll a change back so a bad update can be undone quickly.

What happens if we stop working together?

Engagements are month to month. What we built stays with you: the code, the accounts, and the documentation. If you want support after the engagement ends, we agree on that at the time.

What’s taking more time than it should?

Tell us about one process you’d like to improve. We’ll help you figure out whether we’re a fit.

Or email us directly at
hello@fractionalfixers.ai

Opens a ready-to-send message in your email app. Both of us read every one.