NewAI Content Creation is now live in Early Access
Turning Point Training Instituteby Training Center
Skip to content
0%
Module 4 — Identifying and Defining the ProblemLesson 3 of 4
Video lesson

Baselines and the Problem Statement

BeginnerVideo lesson
8 min readBeginner

Where you are: Module 4 — Identifying and Defining the Problem · Lesson 4.3 · Based on Chapter 4 (§4.5–4.6) of Dr. Tahseen Al Doori's Troubleshooting Essentials.

What you'll be able to do

  • Define a baseline and list the four things it records.
  • Compare a faulty system with its baseline, or with an identical working system, to reveal differences.
  • Write a specific, factual and testable problem statement using the book's template.
  • Judge whether a problem statement is good enough to begin Step 2.

Watch: Baselines and the Problem Statement

The video compares a fictional lab computer with the baseline recorded at the start of term, then builds a problem statement slot by slot for Lena, the receptionist from Lesson 4.1. It ends by assembling the whole of Step 1 into one flow. Keep these questions in mind:

  • Which measurement in the baseline comparison ruled out a theory rather than suggesting one?
  • Which 5W1H question fills each slot of the template?
  • Why does a guessed cause, however technical it sounds, make a poor problem statement?

Establishing a baseline

A fault can only be recognised against a picture of normal behaviour. Is a three-minute start-up slow? Is 60% memory use high? Is a warning in the event log new, or does it appear every day? Without knowing what normal looks like, you cannot answer.

A baseline is a record of how a system looks and performs when it is working correctly. The book lists four things it records:

  • its settings — how it is configured;
  • its software versions — operating system, applications and drivers;
  • its performance figures — for example start-up time, memory and disk use, network speed;
  • its typical log entries — what the logs normally contain, including harmless warnings that appear every day.

That last item is easily overlooked. If you know which warnings appear on a healthy day, you will not waste time chasing them during a fault.

Two ways to compare

The book describes two comparisons, and either one often shows the cause immediately:

  1. Faulty state against the baseline — the same system, as it was when healthy.
  2. Faulty system against an identical working system — another device of the same model built to the same standard configuration.

The second option is especially useful in organisations where many devices share a standard configuration, such as a teaching lab or an office. Module 6 develops it further as the comparison isolation technique.

When to record a baseline. The best time is when the system is first set up and known to work, and again after any planned change. A baseline recorded during a fault is not a baseline — it is evidence.

Worked example: a baseline comparison

(Fictional scenario.) A teaching lab's PC 07 has become slow to start. Its baseline was recorded at the start of term.

MeasureBaseline (start of term)Today
Start-up time40 seconds3 minutes 10 seconds
Free storage210 GB205 GB
Startup programs614
Typical log entriesNo recurring errorsRepeated errors from a newly added updater

Read the comparison row by row:

  • Start-up time turns the user's complaint ("it's really slow") into a measured fact.
  • Free storage has barely changed, so a full disk — a common cause of slowness — is unlikely. The baseline has ruled out a theory before any test.
  • Startup programs have more than doubled, and the logs show repeated errors from a new program. Together they point at what changed.

None of this is yet a confirmed cause. But Step 2 now starts with a short, well-founded list instead of a long, vague one. The same comparison could be made against PC 08, if it is identical and working.

Writing the problem statement

The final technique of Step 1 combines everything — 5W1H answers, evidence, reproduction results and baseline comparisons — into one or two sentences, before attempting any fix. The book states that a good problem statement is specific, factual and testable.

The book provides this template:

Example prompt

Since [when], [who] cannot [what] on [where]. It happens [always / intermittently] when [how]. The most recent change was [which]. The error shown is "[exact message]".

Each slot is filled by one of your Step 1 results:

Template slotFilled by
Since [when]When
[who] cannot [what] on [where]Who, What, Where
It happens [always / intermittently] when [how]When (always or sometimes) and How
The most recent change was [which]Which
The error shown is "[exact message]"Evidence: the exact error message

Lena's problem statement

(Fictional scenario, continued from Lesson 4.1.)

Example prompt

Since 08:00 today, Lena cannot open the booking system on the reception PC. It happens always, when the booking page is opened in any browser. The most recent change was moving the PC to a new desk. The error shown is "This site can't be reached".

A technician who has never met Lena now knows exactly where to start.

What makes a statement good

  • Specific: one system, one symptom. "The reception PC cannot open the booking system", not "the computers are playing up".
  • Factual: what was observed, not what somebody guesses. "Probably a DNS problem" is a theory, and belongs in Step 2.
  • Testable: every part can be checked. You can confirm the start time, try the page, check the browser and confirm the move.
  • Short: one or two sentences.
  • Written before any fix is attempted. If you cannot state the problem in one clear sentence, you are not ready to fix it.

Define before you solve. Use 5W1H, collect exact evidence, reproduce the fault or log it, compare with a baseline, and only then write the problem statement that opens Step 2.

5W1HEvidenceReproduce or logBaselineProblem statementStep 2: Theory

Try it

  1. Record a baseline for a device you use daily, without changing anything: note its operating system version, free storage, approximate start-up time (time it once), and the number of applications that start automatically if you can see this easily. Keep it: next time something goes wrong, you will have something to compare with.
  2. Rewrite a vague report. (Fictional scenario.) A student emails: "The Wi-Fi in the library is rubbish." Invent reasonable answers to the 5W1H questions and write a full problem statement using the template.

Completion criteria: your baseline has at least three measured or recorded items with the date recorded; your problem statement fills every slot of the template, contains no guessed cause, and is no more than two sentences long (the template's short follow-on sentences may be combined).

Knowledge check

Which of these is NOT one of the four things a baseline records?

A faulty laptop is placed beside an identical working laptop and their settings are compared. What is this?

Which is the best problem statement?

In the template, which 5W1H question fills the slot 'The most recent change was [...]'?

In the PC 07 comparison, why is the unchanged free storage useful?

Answers explained

  1. C — the book lists settings, software versions, performance figures and typical log entries. Repair history may be useful elsewhere, but it is not part of the baseline.
  2. A — comparing against an identical working system is one of the book's two baseline comparisons. Reproduction (B) repeats the fault; an occurrence log (C) tracks intermittent faults; rollback (D) undoes a change.
  3. C — it is specific, factual and testable. A and B include guessed causes or fixes; D is vague.
  4. A — Which asks which changes happened just before. Where, How and Who fill other slots.
  5. D — a matching figure rules out a common cause. It does not prove the PC is healthy (A) — start-up is still slow — and nothing suggests replacing storage (B).

Key takeaways

  • A baseline records settings, software versions, performance figures and typical log entries for a system that is working correctly.
  • Comparing a faulty system with its baseline, or with an identical working system, often reveals the cause — or rules causes out.
  • The problem statement combines 5W1H answers and exact evidence into one or two sentences, using the book's template.
  • A good statement is specific, factual and testable, contains no guessed cause, and is written before any fix.
  • Do not attempt a fix until the problem can be stated clearly.

Sources

  • Dr. Tahseen Al Doori — Troubleshooting Essentials, University Edition (2026), §4.5–4.6 (including the Problem Statement Template), Chapter 4 Key Takeaway and Summary; cross-reference to §6.3 (Comparison).