Why Every DevOps Engineer Should Learn Excel in 2026

DevOps · Excel for Engineers

Why Every DevOps Engineer Should Learn Excel in 2026

SECTION 01Where Excel fits into DevOps work

DevOps engineers live in terminals, YAML and monitoring dashboards, so Excel can feel out of place. But look at what happens after the technical work: the monthly cloud bill review, the quarterly capacity plan, the SLA report for a client, the list of incidents for a retrospective. Most of these are prepared and discussed in spreadsheets.

If you can open a CSV export, clean it and summarise it in ten minutes, you can answer questions like "Why did our cloud bill rise this month?" with evidence instead of guesses. That makes you far more useful in planning meetings, and it is a skill few engineers bother to build.

Core formulas (Excel 365 / 2021)

Total cost for one service:
=SUMIFS(Billing[Cost], Billing[Service], "Amazon EC2")

Cost for a service in one month:
=SUMIFS(Billing[Cost], Billing[Service], A2, Billing[Month], B1)

Find the owner team of a resource:
=XLOOKUP([@ResourceId], Owners[ResourceId], Owners[Team], "Unassigned")

Count high-severity incidents for a service:
=COUNTIFS(Incidents[Service], A2, Incidents[Severity], "SEV1")

Cleaning exported data

Extract env from a tag like "env:prod":
=TEXTAFTER([@Tag], "env:")

Split a host name like "web-prod-03" into parts:
=TEXTSPLIT([@Host], "-")

Turn a timestamp text "2026-10-14T08:15:00Z" into a date:
=DATEVALUE(LEFT([@Timestamp], 10))

Trim stray spaces from service names:
=TRIM([@Service])

SECTION 02Analysing cloud costs with pivot tables

Cloud providers let you export detailed billing data — for example AWS Cost and Usage Reports or Azure cost exports. These files can have thousands of rows, one per resource per day. A pivot table summarises them in seconds by service, account, environment tag or team.

The useful question is rarely "what did we spend?" but "what changed?". Put two months side by side, add a difference column and sort by the largest increase. Most cost surprises turn out to be a handful of resources: an oversized database, forgotten test instances, or data transfer nobody planned for.

Month-on-month cost change (summary sheet)

A: Service   B: Sep cost   C: Oct cost

D (change):     =C2-B2
E (change %):   =IF(B2=0, "new", (C2-B2)/B2)
F (rank):       =RANK.EQ(D2, $D$2:$D$40)

Then: sort by column D, largest first.
Conditional formatting on E: red if > 20%.

Power Query: combine monthly billing exports

1. Data > Get Data > From File > From Folder
2. Pick the folder holding all billing CSVs
3. Combine & Transform Data
4. Set types: Date -> Date, Cost -> Decimal Number
5. Add Column > Custom Column:
     Month = Date.StartOfMonth([Date])
6. Remove columns you do not need
7. Close & Load to a Table

Next month: drop the new CSV in the folder,
then Data > Refresh All.

SECTION 03Capacity planning and SLA maths in Excel

Capacity planning asks a simple question: at the current growth rate, when will we run out of headroom? Export daily peak CPU or request counts for the last few months, chart them, and use FORECAST.LINEAR or a trendline to project forward. It is not a perfect model, but it gives a defensible starting point for budget discussions.

SLA reporting is pure arithmetic, yet mistakes are common. Availability is usually calculated as uptime minutes divided by total minutes in the period. Knowing that 99.9% allows roughly 43 minutes of downtime in a 30-day month helps you explain what a target really means.

Capacity forecast (daily peak CPU %)

Column A: Date    Column B: Peak CPU %

Forecast CPU % for a future date in D2:
=FORECAST.LINEAR(D2, $B$2:$B$121, $A$2:$A$121)

Date when CPU is projected to hit 75%:
=(75 - INTERCEPT($B$2:$B$121, $A$2:$A$121)) / SLOPE($B$2:$B$121, $A$2:$A$121)
  (format as a date; only meaningful if the slope is positive)

95th percentile of daily peaks:
=PERCENTILE.INC($B$2:$B$121, 0.95)

SLA formulas

Minutes in period (B1 = start date, B2 = end date):
=(B2 - B1 + 1) * 24 * 60

Availability %:
=1 - (DowntimeMinutes / TotalMinutes)

Allowed downtime for target in C1 (e.g. 99.9%):
=TotalMinutes * (1 - C1)

SLA met?
=IF(Availability >= C1, "Met", "Breached")

SECTION 04Templates: cost review, capacity plan, incident tracker

Set up these three sheets once and reuse them every month. Each one is designed to be refreshed from a new export rather than rebuilt.

MONTHLY COST REVIEW
Month: ________

Total spend:            ₹_______   vs last month: ___%
Top 5 services by cost: (pivot)
Top 5 increases:        (month-on-month sheet)
Untagged spend:         ₹_______

Actions agreed:
1. ______________________  Owner: ______
2. ______________________  Owner: ______
3. ______________________  Owner: ______

CAPACITY PLAN
Service: ____________  Resource: CPU / memory / storage

Current peak (p95):          ____
Current limit:               ____
Headroom:                    ____ %
Growth per month (slope):    ____
Projected limit date:        ________

Planned events (launches, sales, migrations):
- ______________________

Recommendation: scale / optimise / no action

INCIDENT TRACKER (one row per incident)

Columns:
ID | Date | Service | Severity | Detected by | Start | End |
Duration (min) | Root cause category | Action item | Status

Useful formulas:
Duration:   =([@End]-[@Start])*24*60
SEV1 count: =COUNTIFS([Severity],"SEV1",[Date],">="&StartDate)
MTTR:       =AVERAGEIFS([Duration (min)],[Severity],"SEV1")

SECTION 05Tips for using Excel as a DevOps engineer

Use Excel where it helps and automate what repeats:

• Keep raw data untouched — load exports into one sheet and do all calculations in separate sheets.
• Use Power Query for repeats — if you clean the same export every month, record the steps once and refresh.
• Lead with the change — managers care most about what moved since last time and why.
• Know when to script — if a report needs data from APIs daily, move it to Python or your monitoring stack.
• Label assumptions — write growth rates and targets in visible cells, not hidden inside formulas.

SECTION 06Interview Q&A — data-driven DevOps

SECTION 07Test yourself — Excel for DevOps quiz

SECTION 08Frequently asked questions

SECTION 09Related reads