Gap analysis: measuring the distance between where you are and where you said you would be

A gap analysis compares where something is now with where it needs to be, and lists what would close the difference. It applies to almost anything with a standard attached — skills against roles, processes against a certification, service levels against a contract, systems against what the business now needs — which is why the term appears everywhere and means slightly different things each time.

The structure is always the same three columns, and the whole exercise stands or falls on the first one. Current state assessed optimistically produces small gaps, comfortable conclusions and no change. Most gap analyses fail here rather than in the planning that follows.

What the analysis needs

• The dimension being assessed, listed at a level someone can actually judge — a specific requirement, capability or process step.
• Target state, stated as something checkable rather than aspirational. "Compliant with clause 7.5" or "two people can run the close" rather than "better documented".
• Current state, assessed against evidence rather than impression. This is where the honesty either exists or does not.
• The gap, described in a sentence rather than as a score. Scores compress away the information you need to act.
• Impact or priority — not all gaps are worth closing, and saying so explicitly prevents an undifferentiated action list.
• The action that would close it, and roughly what it costs in money or time.
• An owner and a target date.
• Status, so the analysis becomes a live tracker rather than a snapshot.

Assess current state against evidence, not against memory. "We do document that" is worth checking: ask to see the last three instances. Nearly every gap analysis that later proves wrong was wrong in the current-state column, usually because the person assessing was also the person responsible for the gap.

Doing it

1. Define the target first and get it agreed. A gap analysis against an unagreed standard produces an argument about the standard rather than about the gaps.
2. Break the target into assessable items — usually twenty to fifty rows. Fewer hides detail; many more never gets completed.
3. Assess current state with evidence, and have someone other than the owner sanity-check the generous-looking rows.
4. Describe each gap in a sentence. "Partial" tells you nothing; "the procedure exists but was last reviewed in 2023 and does not cover the new system" tells you what to do.
5. Prioritise by impact and effort, and be explicit about the gaps you are choosing not to close.
6. Write actions with owners, dates and rough costs, so the output is a plan rather than a finding.
7. Track status on the same table. A separate action list drifts out of sync within a month.
8. Reassess on a cycle — annually for compliance, quarterly for capability — and keep the old assessment for comparison.

Where it is worth the effort

Gap analysis earns its cost when the target is externally imposed and non-negotiable — a certification, a contractual service level, a regulatory requirement — because there the current-state column is genuinely decision-relevant and the gaps must be closed or explicitly accepted. It is much weaker as a general improvement tool, where the target is something you invented and the analysis becomes a way of listing everything imperfect. If you cannot name who set the target and why, that is a sign to skip the formal exercise.

Ettex Records fits the analysis once it is more than a one-off document: custom tables with typed fields for requirement, target, current state, gap, priority, action, owner, date and status; relations between tables so gaps link to the standard or project they came from; grid and kanban views with saved filters — open gaps by owner, high priority not yet started, everything overdue; formulas and rollups to summarise by area; revision history where every change is tracked, so you can see how the assessment moved between cycles; and CSV import to start from a spreadsheet. The standard and the written conclusions live in Ettex Docs, evidence collected during the assessment fits Ettex Forms, and the resulting work goes to Ettex Board.

Said plainly: Ettex has no gap or compliance module. There is no library of standards to assess against, no scoring engine, no maturity model, no automatic evidence collection and no compliance reporting. You define the rows and do the assessing; Records keeps the result structured and auditable. Formal certification programmes at scale want purpose-built software.

How gap analyses go wrong

• Current state assessed by the person accountable for it, with no independent check.
• Targets that are aspirations rather than checkable statements.
• Gaps recorded as scores, which cannot be acted on.
• Every gap treated as needing closure, producing an action list nobody starts.
• Actions without owners, dates or costs — findings dressed as a plan.
• The tracker separated from the analysis, after which the two disagree.
• Done once before an audit and never repeated, so the trend is invisible.
• Applied to an invented internal standard, generating work for its own sake.

A gap analysis is only as good as its current-state column. Agree the target, assess against evidence, describe each gap in a sentence, decide which ones you are not closing — and keep the tracker in the same table.