6.2Recurring problems (playbook)
The essentials in 30 seconds
- Diagnose: a symptom = likely causes = a sequence of actions — you don't fix at random.
- Roots: types (text vs number), spaces, deleted references, frozen source, volatile functions — 90% of blockers.
- Check: numbers aligned to the right, pivot Refresh, proof on edge cases, not just on an isolated cell.
Wassim thread — the table taped next to the screen
Wassim keeps this table next to his screen in Algiers. Verdict after a quarter: three rows served him most — the #N/A of XLOOKUPs, the SUM that returns 0, and the pivot not up to date. For each, he eventually applied the sequence (symptom, causes, actions) instead of retyping at random.
Observed symptom, checked causes, actions in order: the reflex holds in three beats.
One reflex per symptom
Most Excel blockers are classics. Linking each symptom to its likely causes and a sequence of actions saves precious time.
| Symptom | Likely causes | Actions |
|---|---|---|
| #N/A (XLOOKUP/VLOOKUP) | key not found, spaces, text vs number | TRIM; homogenize the type; 4th « if not found » argument |
| #REF! | deleted reference (row/column removed) | undo; rebuild the reference; prefer structured tables |
| #DIV/0! | division by zero or empty cell | =IF(divisor=0,"",num/divisor) or IFERROR(...) |
| #VALUE! | text where a number is expected | check the types; VALUE; clean the input |
| =SUM returns 0 | numbers stored as text (left-aligned) | retype as numbers / convert (VALUE) |
| Formula shown as-is | cell formatted as Text, or = missing | set to General; retype starting with = |
| Pivot not up to date | not refreshed / frozen source | right-click ▸ Refresh; source as a structured table |
| Slow workbook | volatile functions, excessive CF, large volumes | limit OFFSET/INDIRECT; lighten the CF; Power Query/model |
| #SPILL! | spill blocked by an occupied cell | free the space below/right of the dynamic result |
Ready-to-say script — diagnose without retyping at random (30 s)
« My symptom: [#N/A / SUM at 0 / frozen pivot]. I check the likely cause from the table (spaces, text vs number, source), I apply the targeted action (TRIM, VALUE, Refresh) and I prove it on an edge case. »
Copy-paste template — Excel band-aids
=IFERROR(XLOOKUP(A2,Customers[Code],Customers[Name],"Not found"),"Not found")
=IF(B2=0,"",A2/B2)
=TRIM(A2)
=VALUE(A2)
SME / small budget
No formal table: a symptom, the likely cause, the action and who proves it, noted down.
SME Algiers
Integrate the real exports into the diagnosis: local separator, French dates, known owner.
Remote / hybrid
Add a screenshot to the diagnosis: at a distance, a written symptom and proof avoid duplicates.
Intercultural
Make the findings explicit in writing: what seems obvious (which format, which source) isn't for everyone.
48h challenge — apply tomorrow
Spot 1 current error (#N/A, SUM at 0, frozen pivot), apply the table's row (named symptom + checked cause + targeted action) and prove it.
- I know how to link a symptom to its likely causes.
- I check types and source before correcting.
- I prove each correction (edge case, Refresh, total that adds up).
Wassim's SUM returns 0 although the amounts are visible. Which reflex does the table apply?
The same error coming back over and over isn't ten incidents: it's a design problem to fix at the root (types, ranges, structure).