Beyond Implementation
Building Processes That Actually Work
Here's something I've learned the hard way: the fanciest software in the world can't fix broken processes. It can actually make them worse by cementing bad practices into expensive, rigid systems.
This is the core challenge in finance transformation today. Leaders are sold on sleek software solutions without addressing the fundamental operational issues plaguing their teams.
Fix the process first, then automate it. Never the other way around.
I've seen this pattern repeatedly in my work with finance teams. They're drowning in manual work, so they reach for technological salvation – only to find themselves with the same problems plus a hefty implementation bill.
The Process-First Approach
Before you consider any new finance technology, you need to understand and optimize your current processes. Here's my framework for doing this effectively:
Map the entire workflow: Document every step, handoff, and decision point in your current process. Make the invisible visible. This alone often reveals shocking inefficiencies.
Identify control points and bottlenecks: Where does work consistently slow down or stop? What checkpoints are truly necessary vs. historical artifacts?
Eliminate before automating: Ask the hard question about each step: "Do we actually need this?" Many processes contain unnecessary steps that should be eliminated, not automated.
Test manual improvements first: Implement process improvements manually before investing in technology. This validates your approach and builds user buy-in.
Fixing Month-End Close Hell
Want to know why your month-end close takes weeks? It's not because your journal entry system is broken. It's because your vendor rebate process requires four different people to touch the data, starting with someone manually copy-pasting rows into a CSV file.
No amount of sophisticated software can fix that fundamental inefficiency. You need to redesign the process itself, then find the right technology to support it.
A Real Life Example
Years ago, I was a supervisor managing a team of three analysts, and our month-end close took 7 days for a $10M annual business. I was sick of it.
Why did it take so long? One of the analysts was responsible for building the revenue accrual. To do this, he would spend days tracking down sales reps trying to figure out what the monthly billing estimates were going to be. The reps that did answer the phone couldn’t get off the phone with him fast enough.
In the meantime, the other analyst was responsible for booking payroll and commission accruals. The commission accruals were stuck until we knew what the revenue accrual was going to be so this analyst couldn’t start until the first process was finished.
Lastly, the third analyst was responsible for reporting who produced the P&L and delivered it to yours truly. Obviously reporting wasn’t ready until the books were closed.
You can almost visualize the bottlenecks can’t you?
No amount of sophisticated software can fix that fundamental inefficiency. You need to redesign the process itself, then find the right technology to support it.
The Technology Ladder
Once you've optimized your processes, approach technology implementation in stages:
Level 1: Maximize existing tools (Excel/Google Sheets with proper structure)
Level 2: Add automation to existing tools (Excel macros, Power Query, Python)
Level 3: Implement specialized tools for optimized processes (BlackLine, HighRadius, Anaplan, Planful, etc.)
Most finance teams jump straight to Level 3 without mastering Levels 1 and 2. This is a recipe for implementation disaster.
Remember: Excel with a clean, well-designed process will outperform a fancy SaaS solution with a broken process every single time. I've seen Fortune 500 companies successfully use QuickBooks as a sub ledger.
How We Won with this Approach
We ended up taking our close from 7 days to 2 through this approach without a single cent of CapEx that I inevitably would have been turned down for.
We started with process and mapped every single step of our close process across the three swimlanes of our team. Where we needed cross-functional input (sales and payroll), we engaged with them in the spirit of “if you help us, we’ll bug you less.”
We automated our data processing using Power Query (which comes straight out of the box with excel). We even automated the journal entry template creation. When we were all said and done the two days we could not automate was simply due to the timing of data availability in the system.
This worked at a small scale, but it can also be replicated at-scale. The flip side is also true, when I have strayed from this approach, the solution fell apart.
Start Small, Win Big
One final tip: target a single, high-impact process for your first optimization effort. Instead of a company-wide rollout, focus on a specific business unit or function.
For example, instead of implementing a planning system across the entire enterprise, start with revenue forecasting for one business unit. Get that working at 99% accuracy, then expand methodically.
This approach builds credibility, develops your implementation muscles, and creates internal champions for your broader transformation efforts.
In my next post, I'll share how this process works with external implementation partners so that you don’t have to suffer through one more SaaS implementation that falls short.
Until then, remember: process first, technology second.