AI Business Maturity Model
Certifications
Find a CoachFind a SpeakerSign In
About AscentMaturity LevelsHow It WorksExample JourneysStart Assessment
Back to All Journeys

Sarah M.

CPA Firm Owner
Professional Services
12 employees

2.7

AIBMM Score

Sarah runs a mid-sized CPA firm with 12 employees serving 150+ small business clients. Her team was drowning in manual work: categorizing transactions, chasing receipts, and closing books. She heard about AI tools but didn't know where to start.

Starting Point: The Struggles

Staff manually categorizing 60% of bank transactions

Spending 2+ hours per client per month chasing receipts via email

Month-end close taking 5+ days for each client

Senior staff pulled into routine work instead of advisory

The 90-Day Journey

Week 1
Discovery Assessment
Discovery

Sarah completed the Discovery questionnaire with her operations manager. The assessment revealed they were stuck at Level 2 - using ChatGPT for drafting but not connected to business systems.

Week 2
Workflow Mapping
Workflows

Mapped 8 key workflows including transaction categorization, receipt collection, bank reconciliation, and month-end close. Transaction categorization scored highest on value/readiness.

Week 3
Stage Plan Created
Stage Plans

Created Q1 Stage Plan with WIP limit of 3 Lifts. Selected: AI Bank Feed Categorization (L1→L2), Automated Receipt Collection (L1→L2), and Month-End Close Automation (L2→L3).

Weeks 4-6
First Lift: Bank Feed Categorization
Lifts

Built and piloted AI categorization rules using their accounting software's API. Started with 5 clients, expanded to 20 after 90% accuracy achieved.

Week 7
Friction Encountered
Friction Log

Integration with one major bank's feed had API rate limits. Logged as blocker, escalated to vendor. Workaround implemented while waiting.

Weeks 8-10
Receipt Collection Launch
Lifts

Automated email reminders and mobile receipt capture deployed. Reduced receipt chase time from 2 hours to 15 minutes per client.

Week 11
Metrics Review
Altimeter

Altimeter showed 40+ hours saved per month across the team. Categorization accuracy at 94%. Receipt collection SLA compliance up to 87%.

Week 12
Stage Review
Checkpoints

Graduated 2 Lifts, carried over Month-End Close (still in Pilot). Captured SOPs and prompt templates in Pattern Library for new staff training.

Lifts Executed

AI Bank Feed Categorization
Graduated
L1

L2

94% accuracy, 60% time reduction

Automated Receipt Collection
Graduated
L1

L2

Chase time from 2hr to 15min per client

Month-End Close Automation
Active
L2

L3

In pilot with 10 clients

Friction Encountered & Resolved

Issue

Bank API rate limits blocking batch categorization

Resolution

Implemented queuing system and negotiated higher limits with vendor

Issue

Staff resistance to new receipt capture app

Resolution

Created video training and incentive program for adoption

Results: By the Numbers

Transaction Categorization Time
Before

45 min/client

After

8 min/client

82%
Receipt Collection Time
Before

2 hr/client

After

15 min/client

87%
Hours Saved/Month (Team)
Before

0

After

40+

N/A

Key Takeaways

Start with high-volume, low-risk workflows like transaction categorization

Pilot with a subset of clients before rolling out to all

Capture SOPs immediately - they become training materials

Bank/vendor integrations often have hidden constraints - budget time for them

What's Next for Sarah?

Complete Month-End Close automation, begin AI-assisted advisory report generation

Start Your Journey

Sarah started exactly where you are. Begin with Discovery to chart your own path.

Start DiscoverySee Other Journeys