Analysis Detail

Analysis · Funnel & Channel Detail

Parts 1.1–2.3 of the assignment, on one page. Sections share a global channel filter; methodology lives in the Appendix.

1.2 · Weekly Cohort Conversion

SQL: fct_weekly_cohort_activation · fct_activation_curve · standalone

No Results

Page / 5 10 of 44 records

Activation rate, CPS, CPA, LTV:CAC, plus the pre-signup ad funnel (CTR, CPC, click→signup) for paid channels. unknown kept as its own bucket. Organic / referral / unknown have $0 spend → NULL on all spend-derived metrics.

SQL: fct_channel_performance · int_channel_spend_rollup · standalone

On CTR. Click-through rate is ~2.2% across all three paid channels in this dataset — uniform to a striking degree. Real-world CTRs vary widely by channel (paid search 3-6%, paid social 0.5-2%, LinkedIn 0.4-1%). The flatness here looks like seed-generated synthetic data. Treat CTR as illustrative; CPC and click→signup are the diagnostic metrics.

Signups

10,097

Activated

1,219

Paid Spend

$1,053,267

Expert Payouts (12m)

$822,818

No Results

No Results

2.2 · Campaign Deep Dive

Paid-channel campaigns ranked three ways: by volume, by activation rate (n ≥ 100), by LTV:CAC. LinkedIn outreach sweeps all three.

SQL: fct_campaign_performance · standalone

Recommendation: scale the top-ranked LinkedIn outreach campaigns by LTV:CAC — currently outreach_finance, outreach_healthcare, outreach_tech. They sweep all three rankings, activation rates ~2–3× paid search/social, CPA roughly an order of magnitude lower. Pause paid_search and paid_social pending an attribution audit.

No Results

2.3 · Unknown Channel Investigation

~15% of signups have no recorded signup_source. Funnel reach is roughly half the known average at every stage; rejection rate of unknowns who do submit is ~1.9× the base rate. Not lost attribution — low-quality traffic.

SQL: dim_experts · int_expert_attribution

Not partial attribution loss. 0 of the unknown signups have any utm_campaign populated — if the tagging dropped just `signup_source` we'd expect partial-tag rows, but there are none. These are bare landing-page hits.

Could the unknowns be paid-click bot traffic?

Pearson r = 0.155 across 364 days. Weak positive correlation — consistent with a small fraction of unknowns being driven by paid-click activity that didn't propagate through attribution, but the bulk of the unknown bucket is uncorrelated with paid metrics. Day-to-day CTR variance itself is small (1.5–3.1%), so the test isn't very powerful. The dominant signal still points to bare landing-page traffic without UTM tagging, not bot click-stuffing.

Recommendation. Treat unknown as a traffic-quality signal, not a re-attribution target. Add a referrer / IP / device-fingerprint filter at the top of the funnel to keep low-quality traffic out of the verification queue.

← Marketing Lead Dashboard Memo →

Analysis Detail

Analysis · Funnel & Channel Detail

Filters

1.1 · Full-Funnel Conversion

1.2 · Weekly Cohort Conversion

1.3 · Q3 vs Q4 Comparison

2.1 · Channel Performance

2.2 · Campaign Deep Dive

2.3 · Unknown Channel Investigation

Could the unknowns be paid-click bot traffic?