Interview summary examples: five summaries of one interview, each for a different reader

An interview summary is a short written account of what a candidate showed against the role's requirements, written for one reader: the evidence for each requirement, what was not covered, the logistics, and the next step. A good one describes rather than judges, quotes the candidate with timestamps, and fits on one screen. Below are five summaries written from the same invented interview, one for each reader who usually needs one, each shown beside the weak version people often send instead.

Which summary to write for which reader, and prompts that make an AI draft each one, are in how to use AI to summarize interviews. This page is the finished examples.

What makes an interview summary good?

1. It is written for one reader. A hiring manager, a debrief, a client and an ATS search box need different things.
2. It is organized by requirement, not by the order of the conversation. The reader is checking the role's must-haves, so lead with those.
3. It shows evidence, not adjectives. "Cut the forecast cycle from 9 working days to 5" beats "strong modeler".
4. It says what is missing. No evidence is a finding, and hiding it moves the problem to the next round.
5. It quotes exactly or not at all, with a timestamp so anyone can check.
6. It keeps verdicts where they belong: in the decision record, not in a pre-read.

The interview behind these examples

Invented example. Jordan Vale, Lena Brooks and Bramblecrest Outdoor Supply are fictional; the full passage is in the interview transcript example. The facts every summary below draws on:

• Role: Senior Financial Analyst. Must-haves: forecasting and modeling, business partnering, leading a budget cycle.
• Moved a monthly forecast from a 40-tab spreadsheet into a planning tool; cycle went from 9 working days to 5 [13:56]. Built the driver logic; manager reviewed [14:33].
• Forecast accuracy "within a few percent most months", no exact figure [14:46].
• Weekly review with three store-operations directors; a labor-hours model led one region to cut overtime (no figure given).
• Coordinated two budget cycles with 12 cost-center owners; "I haven't owned the cycle alone" [27:14].
• ERP and systems depth not covered. Expects $95–100k base (example figure); three weeks' notice; hybrid three days is fine.

Example 1: the hiring manager's summary

For a manager deciding whether to move the candidate forward. One screen, evidence against the must-haves, logistics last.

Jordan Vale, Senior Financial Analyst, round 2 (video, 46 min)

Forecasting and modeling: clear evidence. Rebuilt the monthly forecast
in a planning tool and cut the cycle from 9 working days to 5 [13:56];
built the driver logic himself [14:33]. Did not give an accuracy figure
("within a few percent most months" [14:46]).

Business partnering: some evidence. Runs a weekly review with three
store-operations directors; one labor-hours model led to an overtime
cut in one region. Outcome not quantified.

Leading a budget cycle: partial. Coordinated two cycles (12 owners) but
"I haven't owned the cycle alone" [27:14]. Says the gap is the
leadership trade-off conversations, not the mechanics [27:33].

Not covered: ERP and systems depth.
Logistics: $95–100k base expected; 3 weeks' notice; hybrid 3 days OK.
Suggested next step: round 3 to probe budget ownership and systems.

The weak version:

Great conversation with Jordan. Very strong analyst, good communicator,
would fit the team well. Some budget experience. Recommend moving
forward.

The weak version gives the manager nothing to check and hides the one real gap. "Some budget experience" is true and misleading at the same time.

Example 2: the debrief pre-read

For the other interviewers, sent after everyone has submitted their own scorecard. One block per competency, evidence for and against, no rating. The meeting itself is covered in the interview debrief template.

Round 2 evidence (L. Brooks): Jordan Vale

FORECASTING AND MODELING
For: 9 → 5 working days after moving the forecast into a planning tool
     [13:56]; built driver logic [14:33]
Against or missing: no accuracy figure [14:46]

BUSINESS PARTNERING
For: weekly review with three operations directors; overtime decision
     influenced by his labor-hours model
Against or missing: no outcome figure

LEADING A BUDGET CYCLE
For: coordinated two cycles, 12 owners
Against or missing: "I haven't owned the cycle alone" [27:14]

NOT COVERED: ERP and systems depth

The weak version opens with "I'd rate him a 3 overall and lean hire". Everyone who reads that first will read the rest through it.

Example 3: the client blurb for an agency submittal

If Jordan were an agency candidate, the client reads this before deciding to interview. Three reasons with proof, the honest gap, and logistics. The full document format is in the candidate submittal template.

Jordan Vale: Senior Financial Analyst

Why Jordan:
- Rebuilt a retail monthly forecast in a planning tool and cut the cycle
  from 9 working days to 5; designed the driver logic himself.
- Partners weekly with three operations directors; his labor-hours
  model changed an overtime decision in one region.
- Has coordinated two annual budgets across 12 cost-center owners.

To explore: he has not owned a budget cycle alone; he describes the
gap as leadership trade-off conversations rather than mechanics.

Available with 3 weeks' notice. Base expectation $95–100k. Hybrid
three days works.

The weak version lists his job history line by line. The client can read the CV; what they cannot get from it is what the interview proved.

Example 4: the ATS note

For your team, months from now, searching by skill or by why a candidate did not move forward. Same order every time. More on the format in how to write candidate notes in your ATS.

[Date] | Round 2 | L. Brooks | Senior Financial Analyst | video 46 min
Outcome: advance to round 3
Forecasting: planning-tool migration, 9 → 5 day cycle, built drivers.
Partnering: weekly ops review (3 directors), labor-hours model.
Budget: coordinated 2 cycles / 12 owners; has not owned a cycle alone.
Not covered: ERP depth.
Comp: $95–100k base expected. Notice: 3 weeks. Work setup: hybrid 3 days OK.
Keywords: FP&A, forecasting, driver-based model, retail, budget process

The weak version says "Good call, moving forward". In six months that note answers nothing.

Example 5: the next-round handoff

For the round 3 interviewer. Coverage and open questions, never the score. The full template is in handoff notes between rounds.

Covered in round 2: forecasting (strong evidence), business partnering
(some), budget coordination.
Please probe:
1. Budget ownership: what would he do in week one of running a cycle
   alone? He named leadership trade-offs as his gap [27:33].
2. Systems: ERP depth was not covered at all.
3. Forecast accuracy: he had no figure; ask how he would measure it.
Logistics are settled; no need to revisit pay.

An interview summary template to copy

[Candidate] | [Role] | [Round] | [Interviewer] | [Date, format, length]

[Requirement 1]: [clear / some / partial / no evidence]
[Fact or quote with timestamp]
[What is missing]

[Requirement 2]: ...
[Requirement 3]: ...

Not covered: [topics]
Logistics: [pay expectation] | [notice or start] | [location / setup]
Next step: [what, who, by when]

"Clear / some / partial / no evidence" describes the evidence, not the person, which keeps it safe for pre-reads. Swap it for your scorecard rating only in documents that are meant to carry a decision.

How do you check an AI-drafted summary before sending it?

• Every quote: find it in the transcript, with the right speaker. If it is not there, delete it.
• Every number: pay, notice, team sizes, dates and results. These are the words speech recognition gets wrong most often.
• Every negative: "has" and "hasn't" can swap, and the summary will repeat the error confidently.
• Missing evidence: AI summaries tend to smooth gaps into "some experience with". Restore the gap.
• Labels: strike personality words ("confident", "energetic") and anything about age, health or family.

Why those checks matter, with the research behind them, is in AI note taker hallucinations. For a longer in-house version of Example 1, see the candidate assessment summary template.