Prestige Financial Services

Key Accountabilities:
● Responsible for providing galvanizing leadership to a team of data scientists, a team of data analysts, and a team of BI developers effectively developing their capabilities and expanding their responsibilities
● Relied upon to seamlessly collaborate with CFO and Pricing Committee to develop and implement our pricing strategies
● Depended upon to manage production machine learning models ensuring targets business objectives are met
● Proficient risk counterpart for the originations committee, servicing committee, pricing committee and executive credit committee

VP of Data Science & Analytics

Key Accountabilities:
● Responsible for providing galvanizing leadership to a team of data scientists and analysts, effectively developing their capabilities and expanding their responsibilities
● Relied upon to seamlessly collaborate with CFO and SVP of Structured Finance to transition the ALLL (allowance for loan and lease losses) to follow CECL (current expected credit losses) guidelines
● Depended upon to manage production machine learning models ensuring targets business objectives are met
● Proficient risk counterpart for the originations committee, IT steering committee and executive credit committee

Key Accomplishments:
● Engineered the development and deployment of four machine learning credit scoring and pricing model reducing vendor costs by over $1M per year
● Supervised the creation of an NLP model used to monitor collectors text messages for compliance violations; monitors 2.9M text messages per year, 73% decrease in potential compliance violations
● Mentored a direct report in the creation of enhanced net loss forecasting models using state of the art machine learning and natural language processing models; reduced error rate from 30% to 5.5% at time of funding
● Created a book value prediction model that predicts the book value of a vehicle with an error of 7%.  Used to populate the book value of vehicles on the inventory page of our new dealer portal.  Pulling the actual value for each dealership and all vehicles on their lot would cost roughly $120,000 per year.

Director of Analytics

Key Accountabilities:
● Responsible for accurately developing and monitoring new credit underwriting models to approve applicants
● Relied upon to create or improve departmental reporting capabilities, effectively educating stakeholders with data-visualization and predictive modeling

Key Accomplishments:
● Orchestrated process improvements including the removal of an outdated secondary pricing system; System decisioning increased from 77% to 88%; Approval Rate increased from 28% to 32%
● Innovated cumulative net loss forecast improvements; Old method = loss timing curve only; New method = combination of loss timing curve, regression, probability of default/exposure given default/loss given default
● Championed monthly net loss projection improvements by transitioning from using static pools and roll rates to an ARIMA model, which resulted in decreasing MAPE from 22% to 8.5%
● Successfully distilled Risk Summary into a focused presentation of application statistics, monthly loss projections, vintage loss projections, recovery performance, early performance summaries, and collections statistics
● Consistently received annual promotions, from Risk Analyst I to Risk Analyst II, then Risk Analyst III, and most recently into Vice President role

Risk Analyst

1776 Faux Farmhouse

Key Accountabilities:
● Responsible for maintaining websites, developing content for our newsletters Upgrade Your Builder-Grade and The Saturday Social, and providing analytical support for all social media platforms
● Relied upon to seamlessly collaborate with brand managers ensuring that deliverables are met and that contracts are executed as required
● Depended upon to meet key business objectives: growing company revenue, providing areas for audience growth, & expanding into new product areas

Key Accomplishments:
● Helped grow our social media audience from 9k on Instagram to over 119k followers by using machine learning and content analytics; resulted in 20.7M impressions and an average daily reach of 44.92% in 2022
● Developed and released the 1776 Faux Farmhouse website complete with 2 DIY courses and future courses in development
● Engineered growth for our newsletter – Upgrade your Builder-Grade – from 200 subscribers to 5k subscribers by using automation, a/b testing, and website analytics
● Encouraged growth on TikTok going from 0 to 140k followers using the same content analytics strategy used for Instagram; resulted in 6.4M views and 250K engagements in 2022

Co-Founder

Brigham Young University

• Teaching Assistant for Stat 462: Quality Control and Industrial Statistics
• Teaching Assistant for Stat 512: Stat Methods for Research 2

Teaching Assistant

• Research on belonging among junior high students using cluster analysis

Research Assistant

Springville Dentistry

• Filed and pulled patient charts for the following day
• Filed insurance paperwork
• Performed lab work on cad cam machine
• Made crowns

Office Assistant

Riverwoods Surgery Center

• Made purchases of medical supplies while adhering to a budget
• Analyzed the cost of surgeries and looked for areas for improvement
• Stocked medical supplies

Assistant Purchasing Manager

VP Analytics | 145k Instagram | 145k TikTok | I teach creators how to grow their business using analytics.

Hey, I'm Michael 👋🏻

I'm an executive by day and DIYer by night! Together with my wife Nicole Francis we've built 1776 Faux Farmhouse into a community of over 250,000 people. We've been in your shoes and know what it's like to build an audience.

Our newsletter, The Saturday Social, shows you our secrets. It's a behind the scenes look at 1776 Faux Farmhouse and shows you exactly how we grew our community using analytics.  

I've always been a huge fan of data science and analytics. From the minute I started working in finance I knew that data could transform companies if used properly. And so far it has!

My Results as an Executive:
● Orchestrated the development of internal AI-powered credit scoring and pricing models; over $1M per year reduction in vendor costs, $6.6M projected net loss saving per year, 100% increase in approval rate
● Supervised the creation of an NLP model used to monitor collectors text messages for compliance violations; monitors 2.9M text message per year, 73% decrease in potential compliance violations
● Spearheaded partnership with Zest AI to conceptualize and deploy the 1st AI-powered credit scoring model in the subprime auto lending industry

In 2020, while completing my Masters in Data Science at Northwestern, Nicole and I were slowing building a community while tackling DIY projects around our home.  

In the beginning things were slow, between August 2019 and October 2020 we grew to 5500 followers.

During that time period we tried strategies involving follow for follow, engagement communities, and tours. It doesn't work! It's all fake... There is no hack to building a community except your content.

In October of 2020 Nicole and I had an epiphany. If large corporations use analytics to become better, why can't we?

We went all in on data, data science, and analytics to improve our content. We even used machine learning to predict what content would perform best with our audience. 

And from that point, things have exploded!

Our Results at 1776 Faux Farmhouse:
● 120k+ people following our DIY inspiration on Instagram
● 140k+ people following our DIY inspiration on TikTok
● 50M+ impressions in the last 2 years
● 7X'd revenue in 2022 compared to 2021

If you want to learn more about how we did this, you can sign up for my newsletter.

THE SATURDAY SOCIAL

Get 1 actionable stat every Saturday to launch, improve, & grow your social media using analytics in less than 5 minutes! Join 5,000 other enthusiasts and content creators in this broadcast.

👉🏼 GO HERE: https://creators.1776fauxfarmhouse.com/previous-newsletters/

Michael Francis

Lean Six Sigma | Process Improvement | Data Governance | Analytics | Strategy Optimization | Process Transformation | Business Intelligence | Robotic Process Automation | Master Data Management | Operational Excellence

Entrepreneur | Experienced Analytics and Data Governance Professional | Lean Six Sigma Black Belt | Expert in Business Intelligence, Change Management, and Enterprise Transformation | Building Better Data Systems | Small Business Leader

I began my career in the consumer track of the Accelerated Career Track program at BOK Financial, focusing on retail banking. As an Assistant Manager, I honed my skills in effective communication and employee accountability.

Transitioning to Human Resources, I discovered my passion for Lean Six Sigma, which enabled me to solve complex problems using a methodical approach. While working at BOKF, I earned a Lean Six Sigma Green Belt, an MBA, and a Six Sigma Black Belt certification. I also developed proficiency in Microsoft Power BI and Alteryx for business intelligence and analytics.

At Gateway First Bank, I built the Process Improvement team from the ground up and spearheaded the enterprise Data Governance program. I oversaw data governance processes, master datasets, enterprise data warehouse, and reporting/analytics. Together, Process Improvement and Data Governance drove high-value objectives.

In December 2021, I founded Brewster Consulting Group, a professional services firm specializing in:

- Leveraging data to analyze processes and solve complex business problems
- Driving organizational value through process improvement
- Transforming strategic goals and critical business needs into successful execution plans
- Creating proficient analytics and business intelligence dashboards
- Aligning organizational initiatives with strategic goals and assisting with prioritization and resourcing

We foster an environment of data-driven decision-making and leverage data to improve business outcomes.

Key Skills: Process Improvement, Data Governance, Lean Six Sigma, Business Intelligence, Execution Strategy, Change Management, Enterprise Transformation, Process Transformation, Business Processes, Strategy Optimization, Thought Leadership, IT Solution Integrations, Leadership, Robotics Process Automation, Process Mapping

#ProcessImprovement #DataGovernance #LeanSixSigma #BusinessIntelligence #ChangeManagement #EnterpriseTransformation #StrategyOptimization #LeadershipDevelopment #RPA #ProcessMapping

Kade Brewster

56% of CEOs report no revenue gains from AI. That’s from PwC’s 2026 Global CEO Survey of 4,454 executives.

Not “disappointing returns.” No returns.

The 12% who are seeing both cost and revenue benefits have one thing in common: they didn’t start with AI. They started with a specific business problem, made sure the data underneath it was clean, and then applied AI to that problem with clear metrics.

The other 88% started with “we need an AI strategy” and worked backward. That’s how you end up with pilots that demo well and deliver nothing.

AI is not a strategy. It’s a tool. Tools only work when you know what you’re building.

I Built an AI Invoice Auditor in an Hour. It Found $3,600 in Duplicates.

I built an automated invoice auditor last week. Took about an hour. I tested it against a set of sample invoices with duplicates hidden in them. I had Claude make duplicate invoices that represented a real situation I've encountered with an organization. Inside of these invoices I hid $3,600 in duplicate charges. It caught every single one.

Here’s the setup.

A vendor sends invoices as PDFs. It gets dropped into a folder. That’s the trigger. No manual uploads. No “remember to check this.” Just drop the file and the workflow fires.

Zapier picks up the new file, sends the contents to Claude’s API with a prompt that tells it exactly what to look for: duplicate line items, same candidate billed twice, charges that don’t match the contract rate sheet.
Claude analyzes the invoice and posts the results to a Slack channel. Invoice number, total amount, line item count, any duplicates found, dollar amount flagged, recommended action. The processed file moves to a “completed” folder automatically.

Total build time: about an hour. Total cost to run: pennies per invoice. I tested it against 5 sample invoices with intentional duplicates planted in them and it flagged $305/month in duplicate charges. That’s $3,600 annualized from a test set.

That’s a background screening vendor with maybe 15–20 line items per invoice. Imagine this running against a vendor billing you hundreds of line items monthly. The savings scale fast.

Most companies pay invoices without auditing them. Not because they don’t care, but because manually checking every line item against a contract isn’t realistic. AI makes it realistic.

Video walkthrough of the full build coming soon. I’ll show every step so you can replicate it.

I built a working app this week.

It took an afternoon.

No dev team. No sprint planning. No six-month roadmap.

The context: the 2026 NCAA Wrestling Championships are this week. I wanted a tool that could simulate team scores and bracket outcomes across all weight classes before results came in.

So, I built one.
It's not complex. It's a clean simulation tool that lets you run scenarios and see how team scores could shake out based on individual placement results. Exactly what I needed, nothing more.

The point isn't wrestling.

The point is that a functional, purpose-built tool now takes an afternoon to build, not a quarter.

Most organizations still operate as if custom tooling requires a formal project, a budget cycle, and a developer on retainer. That assumption is about three years out of date.

The businesses figuring this out first won't just move faster. They'll stop tolerating the 20% gap in every off-the-shelf tool they bought to solve a specific problem.

Link to the simulator in the first comment if you want to check it out.

Growing your sales and marketing team is a good thing. Until it isn’t.

Most companies scale these teams based on revenue growth or gut feel. We’re growing, so we need more salespeople. We’re launching a new product, so we need more marketing headcount.

The problem is that growth becomes the justification. Nobody goes back and asks: is this spend actually producing results?

Marketing adds headcount and agency spend but can’t tell you cost per qualified lead. Sales grows the team but pipeline per rep stays flat, or even drops. 

Tech stack expands with every new hire who brings their favorite tool. 

Nobody’s measuring output against input.

This isn’t a sales and marketing problem. It’s a measurement problem.

What is our fully loaded cost per closed deal? And how has that number changed as the team has grown?

If nobody can answer that confidently, you don’t know whether you’re investing in growth or just funding headcount.

Most RIFs start with a number. Cut 15%. Reduce headcount by 30.

The problem isn’t the decision to reduce. It’s how it gets executed.

Leadership picks a target. Managers scramble to protect their people. HR builds a list. And somehow the person who was the only one who knew how to run the month-end close process is gone by Friday.

Before you cut a single role, you need to know: Who actually does what? Not job descriptions, but real work. Who holds institutional knowledge that isn’t documented anywhere? Which roles touch processes that break if that person leaves?

Most leaders can’t answer those questions. Not because they’re bad leaders, but because that level of granularity doesn’t live in an org chart or an HRIS system. It lives in people’s heads.

A well-executed RIF requires process documentation, not just a headcount target. Understand the work first. Then decide who’s essential to it.

The alternative is cutting 15% today and spending 6 months rebuilding what you didn’t realize you lost.

Most businesses are making decisions on gut feel and spreadsheets held together with hope.

That works... until it doesn't.

I've spent years helping companies clean up the mess that happens when data strategy is an afterthought. The same problems keep showing up:

No single source of truth. Everyone has their own version of the numbers.
Reporting built on reporting built on reporting. Nobody knows where anything comes from.
Decisions get made slowly because nobody trusts the data.

So, I built the Data-First Doctrine.

It's a practical framework built for businesses. It's something that actually works in building data-driven processes.

If you'd like to learn more, download our framework in the comments below. 

It's free. It's a PDF. It takes 20 minutes to read.

If you run a business and you've ever said "I just don't trust our numbers" — this was written for you.

Download link in the comments.

Your most expensive employee might be a bad process.

Not a person. A process.

The approval that requires 4 people when 1 would do.
The report built manually every week that could run automatically.
The handoff that drops something 20% of the time and nobody tracks it.

These aren't dramatic failures. They're quiet taxes your business pays every single day.

Most companies don't fix them because they don't know exactly what they're costing.

So, they stay. And they grow.

We quantify the drag before we recommend anything. No vague recommendations. Actual numbers.

What does process inefficiency cost your business per year? Most of our clients are surprised by the answer.

Oklahoma State University has learned of the passing of U.S. Air Force Maj. John "Alex" Klinner. He was one of six crewmembers on a KC-135 refueling aircraft that went down in the Middle East. The Cowboy family is grateful for his service to our country and we are keeping his wife, Libby, and young children in our thoughts and prayers.

Alex was graduate of the MBA program in OSU's Spears School of Business in 2025.

If your job is primarily executing repetitive tasks, whether that’s manual labor, data entry, or writing boilerplate code, the next five years are going to be uncomfortable. That’s not fearmongering. That’s the trajectory.

Being scared about it is understandable. But fear doesn’t change the reality. You have to adapt.

Here’s the part most people miss: the same technology that’s automating task-based work is also destroying the barriers to creating things.

Want to build a software application? Five years ago that required a development team or months of learning to code. Today, you can build functional apps with AI tools and no engineering background. The barrier went from $$$$ to practically zero.

Want to design a physical product? AI-assisted design tools and consumer-grade 3D printers let you go from concept to prototype on your desk. No factory. No six-figure tooling investment. Just an idea and the willingness to iterate.

Want to launch a media company, a course, a consulting practice? The tools to produce, distribute, and market professional-quality content are available to anyone with a laptop.

The people who will thrive aren’t the ones clinging to tasks that machines will do better. They’re the ones who use these tools to create things.

The future belongs to builders. The cost of entry just dropped to almost nothing. What are you going to make?

I did not write this, although I wish I had. 

However, this is an excellent overview on where we stand right now in the new AI world. 

I would encourage everyone to read this.

Open article: A Guide to Which AI to Use in the Agentic Era by open.substack.com

A Guide to Which AI to Use in the Agentic Era

X (Twitter) is blowing up with discussion about OpenClaw. 

It seems like an awesome tool, but just word to the wise from people a lot smarter than me on InfoSec. Do not use it in a professional environment right now. Too many security vulnerabilities.

A custom manufacturer wanted to become more efficient and make their processes repeatable. 

They decided to invest in a new ERP. Top of the line software, Gartner Magic Quadrant solution. 

Their belief was that this would fix their issues. 

Unfortunately, they didn't understand: 

- Their current workflows
- Their current limitations
- The things they'd have to create to effectively implement (Master Customer List, etc.)
- How the physical shop interacts with their systems in the production process

Instead of doing the proper legwork, they signed a large contract with the ERP vendor to implement. 

Then they found out that they didn't have master customer, master product, or many other data tables needed to properly configure the system. So, project delayed six months. 

Next, they got into configuration, and they realized they didn't have any material to make decision points on configuration options. Six more months to map current processes and inform configuration decisions. 

Several cycles of these types of issues, you get the picture. 

At this point it's been 3 years, 4 different fired project management firms, 3 different implementation partners, countless delays and millions of dollars wasted. 

I happen to know this isn't an uncommon story either with major implementations. In fact, this might be more the norm, than the exception. 

Successfully completing a major implementation is kind of like winning a championship in sports. You don't just wake up one day and decide to do it. 

It takes months, maybe years of practice, prep, and discipline to be in a position to have success. And even then, it's difficult. 

But, if you don't properly prepare for it, you're almost guaranteeing failure.

The average company uses 106 SaaS applications. Only about half of those licenses are actually being used.

But software is just where the conversation starts. The real vendor redundancy problem is broader than your tech stack.

Here’s where we see it most:

• Software and SaaS — The obvious one. The average company has 15 duplicate training platforms, 11 project management tools, and 10 overlapping collaboration apps. Departments buy their own tools without checking what already exists. One Zylo study estimated $18M in annual license waste at the average enterprise.

• Shipping and logistics — Multiple departments setting up their own carrier accounts instead of consolidating volume for bulk discounts. Nobody in finance knows how many UPS or FedEx accounts are active.

• Staffing and recruiting agencies — HR uses one firm, a department head uses another, and a regional office has their own relationship. Three vendors, three fee structures, no leverage.

• Marketing and creative agencies — A web agency, a social media agency, a branding consultant, a PR firm, and a freelance designer. Some of their scopes overlap significantly, but nobody’s mapped it.

• Insurance and benefits brokers — Especially after mergers or acquisitions. Legacy broker relationships survive because nobody revisited them.

• Office supplies and MRO — The classic tail spend problem. Low-dollar, high-volume purchases spread across dozens of vendors with no consolidated pricing.

The pattern is the same every time: each department made a reasonable decision in isolation. But nobody has a centralized view of total vendor spend by category.

Procurement teams at best-in-class companies manage 91% of their spend under contract. The average? 57%.

That gap is where the money leaks.

The fix isn’t complicated. Build a vendor inventory. Map spend by category. Identify overlap. But somebody has to own it and in most mid-market companies, nobody does.

McKinsey released a study in late 2025 that should make every CEO pause.

57% of U.S. work hours could be automated with technology that exists right now. Not future technology. Not “by 2030.” Technology that’s available today.

For context, their 2023 estimate was 30%. It nearly doubled in two years.
Now this doesn’t mean 57% of jobs disappear. It means 57% of the activities people do at work every day are technically automatable. The data entry. The status updates. The report generation. The approval workflows.

The real question for business leaders isn’t “will AI take jobs?” It’s: how much are you paying people to do work that a system could handle?

If more than half of your workforce’s time is spent on tasks that don’t require human judgment, that’s not a technology problem. That’s an operational cost problem you haven’t measured yet.

The companies that figure this out first won’t just be more efficient. They’ll be structurally cheaper to operate. And that’s a competitive advantage that’s hard to undo.

According to Gartner, 30–40% of IT spending in large enterprises is shadow IT.

Shadow IT is technology your employees are buying, subscribing to, and using without IT’s knowledge or approval. And in some organizations, Everest Group estimates it’s closer to 50%.

This isn’t malicious. It’s practical. IT can’t keep up. Only 12% of IT departments can respond to new technology requests fast enough. So, employees solve the problem themselves. They sign up for a project management tool. They expense an AI writing assistant. They find a workaround.

Each individual decision makes sense. But collectively, you’re looking at a third of your technology spend that nobody’s managing, nobody’s negotiating, and nobody’s auditing.

That means duplicate subscriptions across departments. No volume discounts. No centralized security review. No visibility into what’s actually being used versus what’s just being paid for.

65% of SaaS applications running in the average organization were never approved by IT.

Let that sink in. Two-thirds of the software your company is paying for—storing data in, running workflows through, giving access to customer information, was adopted without anyone in IT signing off on it.

And it gets worse. 48% of enterprise applications are completely unmanaged. Nobody is monitoring usage. Nobody is reviewing security. Nobody knows when the renewal date is.

Meanwhile, 41% of employees are actively acquiring or creating technology that IT doesn’t know about. Gartner projects that number hits 75% by 2027.

This isn’t just a security issue, though it is that too. It’s a cost issue. The average company wastes $135,000 annually on unnecessary SaaS tools.

The question for leadership: do you actually know what software your company uses? Not what’s in the IT asset inventory. What’s actually running, actually being expensed, actually touching your data.

Nobody budgets for broken processes. But they pay for them every month.

It shows up as:

Overtime that never seems to end.
Vendor invoices that get paid twice.
Customers who churn because something fell through the cracks.
Managers spending Friday afternoons chasing down status updates.

None of those line items say "broken process" on them. They show up as payroll, refunds, and turnover.

The companies we work with aren't failing. They're growing. But growth magnifies inefficiency. What works at 20 employees breaks at 50. What works at 50 breaks at 100.

The hidden cost isn't just the money lost. It's the ceiling it puts on how far you can scale.

We've helped clients recover six figures in operational waste they didn't know they had.

Curious what that looks like for your business?

taplio

The Battle

Michael Francis

Kade Brewster

Followers

Posts

Engagements

Likes

Performance ScoreKade Brewster's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Michael Francis

Kade Brewster

Followers

Posts

Engagements

Likes

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Kade Brewster's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreKade Brewster's LinkedIn Performance