Trace the recommendation back to evidence; identify what makes a workbook weak or trustworthy
Trace the Recommendation Back to Evidence
Open your shared workbook. We are going to audit it together — not to fix formulas, but to answer one question: does the recommendation follow logically from the evidence? Everyone is using the same data, so we can compare reasoning quality directly.
A strong inventory workbook tells a story from raw data to recommendation. Follow this chain and check each link:
Inputs → GAFS
Beginning inventory + purchases = Goods Available for Sale. Check: do your input totals match?
GAFS → COGS + Ending Inventory (by method)
Each method assigns costs differently. Check: does COGS + Ending Inventory = GAFS for every method? If not, the method logic has a bug.
COGS → Turnover + Days-on-Hand
Turnover = COGS ÷ Average Inventory. Days-on-Hand = 365 ÷ Turnover. Check: do your KPIs update when you change the method?
MethodCompare → Recommendation
The recommendation should cite specific numbers from the comparison. Check: does the recommendation name the method, cite at least two numbers, and acknowledge one risk?
Follow these steps with your workbook open:
- Check the Inputs sheet. How many beginning inventory layers? How many purchases? How many units sold? Write these numbers down — they are your foundation.
- Check the Valuation sheet. Pick one method (start with FIFO). Trace one layer from Inputs through the COGS calculation. Does the math match what you would do by hand?
- Run the GAFS check. For each method, add COGS + Ending Inventory. Does it equal GAFS? If any method fails, flag it.
- Check the MethodCompare sheet. Which method has the highest COGS? Which has the highest ending inventory? Do these results make sense given rising costs?
- Check the Dashboard. Do the KPI tiles show different values for different methods? If they all show the same number, the dashboard is not linked correctly.
- Read the Recommendation. Does it cite specific numbers? Does it name a risk? If the recommendation says "FIFO is best" but gives no numbers, it is weak.
As you audit, watch for these red flags:
- No evidence chain: The recommendation appears but no sheet shows how the numbers were derived
- Hard-coded outputs: COGS or turnover values typed in directly instead of calculated from inputs
- Missing GAFS check: No verification that COGS + Ending Inventory = GAFS
- Invisible errors: #N/A or #DIV/0! hidden by white font instead of fixed
- Generic recommendation: "FIFO is usually best" without citing any workbook-specific numbers
- No risk statement: Every method has tradeoffs. A recommendation without a risk acknowledgment is incomplete
- Static charts: Charts that point to fixed ranges instead of Table references
Your task: Identify at least two weak spots in your current workbook (or in the areas you have not completed yet). Write them down — you will address them in Phase 4.