Pipeline Tracker & Funnel Health Report

What you'll produce

An ATS-style pipeline tracker for your open Senior Backend Engineer req — every candidate laid out by stage — plus a short funnel-health read that turns those rows into a decision: the conversion rate between each stage, where the funnel is leaking, whether you're on pace against the 8-week deadline, and one evidence-based recommendation you'd actually bring to Dana. This is the artifact that makes you a credible advisor rather than an order-taker (Topic 8): instead of "I'm working on it," you walk in with "here's the data, here's the leak, here's the one change that fixes it." It proves the Topic 9 skills — reading conversion and time-to-hire numbers to diagnose a problem and advise with evidence — and it's the deliverable that ties your whole search together, because the rows in this tracker are the candidates you sourced (Deliverable 2), wrote to (Deliverable 3), and screened (Deliverable 4).

Instructions

1. List your stages as columns, in funnel order, using the names a real ATS would use for this search: Sourced → Contacted → Replied → Screened → HM review → Onsite → Offer → Hired. (You can collapse or split to match your loop, but keep it the funnel from Topic 4.)
2. Build the tracker table. One row per candidate with: name (or initials), source channel, current stage, days in current stage, status (active / on-hold / rejected / withdrawn), and a one-line note. Aim for 10–20 candidates — enough that the math is real. Make the rows consistent with your earlier deliverables (your screened candidate from Deliverable 4 and your outreach target from Deliverable 3 should both appear).
3. Count candidates reached at each stage (cumulative — how many ever got to that stage, not just who's sitting there now). This is what conversion is calculated from.
4. Compute stage-to-stage conversion rates. For each adjacent pair, next ÷ previous. Also compute two that recruiters live by: reply rate (Replied ÷ Contacted) and screen-to-onsite rate (Onsite ÷ Screened). Show the percentages, not just raw counts.
5. Find the leak. Identify the single stage with the worst drop-off relative to what's healthy for this kind of search — a 60% reply rate isn't a leak, a 15% screen-pass rate might be. Name it explicitly and say what the likely cause is (weak outreach? a must-have that's too strict? salary surfacing late? slow HM feedback?).
6. Check time-to-hire against the deadline. State days-open so far, your current stage of progress, and whether the remaining loop fits inside the 8-week (56-day) target. Flag it red/yellow/green.
7. Write the funnel-health read (4–8 sentences): what the numbers say, where it's leaking, the time risk, and one specific, evidence-backed recommendation to the hiring manager — phrased the way you'd actually say it to Dana, with the number that justifies it (e.g., "we've screened 6 and 4 fell out — 2 because we're rejecting strong distributed-systems engineers for not having a payments logo, which we agreed at intake was only a nice-to-have, and 1 over salary — recommend we screen on correctness-at-scale, not the brand, and hold the band"). Target a requirement you actually agreed was a must-have (or a real logistics constraint like the band) — don't recommend loosening something a rejected candidate already satisfied (if your bar is "Python or Go," a strong Python engineer can't have been cut for "not enough Go"), or a sharp hiring manager will catch the gap. When salary comes up, quote it against your intake band ($185–215K base, Dana's flex to $225K) so the recommendation lands on one consistent number across every deliverable.
8. Keep it honest. Don't invent a clean funnel. A real one-month-old, weak-inbound search leaks somewhere — show the leak and the fix, that's the whole point.

Rubric

The app's AI scores the learner's submission against these criteria and gives feedback. Levels: Needs work (1) / Solid (2) / Excellent (3). Passing = every criterion at Solid or above.

• Pipeline tracker is complete and realistic — 1: a few rows or only "good" candidates, no stages/notes · 2: 10+ candidates across the real funnel stages with status and notes · 3: a tracker that reads like a true ATS export — rejected/withdrawn/at-risk rows included, consistent with the learner's earlier deliverables, days-in-stage shown.
• Conversion rates computed correctly — 1: raw counts only, or wrong math · 2: stage-to-stage percentages including reply rate and a screen→later rate · 3: clean cumulative funnel with correct percentages and the recruiter-standard rates (reply, screen-to-onsite) called out.
• Leak diagnosis — 1: no leak identified, or "we need more candidates" hand-wave · 2: names the weakest stage and a plausible cause · 3: pinpoints the leak relative to healthy benchmarks, names the specific cause (requirement / salary / feedback speed), and backs it with the numbers from the table.
• Time-to-hire read — 1: missing or just days-open · 2: states days-open vs. target and whether the loop fits · 3: a red/yellow/green call tied to the deadline and to the pipeline's actual state, including the downside scenario.
• Evidence-based recommendation to the hiring manager — 1: generic advice or none · 2: one concrete recommendation linked to a number · 3: a specific, non-obvious recommendation phrased as you'd say it to the HM, justified by the exact metric, with a clear next action (advise-don't-just-serve, Topic 8).
• Coherence with the search — 1: a standalone spreadsheet disconnected from the story · 2: candidates and role match the prior deliverables · 3: the tracker clearly continues one coherent search — the outreach target and screened candidate appear, and the recommendation sets up the close (Deliverable 6).