/channel-and-funnel-quality

name: channel-and-funnel-quality
description: Judge whether traffic is actually valuable and whether funnel drop-off is real or expected. Use when comparing marketing channels, reading a conversion funnel, or deciding where to invest. Covers volume × engagement × conversion as a matrix, vanity-traffic detection, expected step drop-off by funnel type, cohort decomposition, and mix-shift (Simpson's paradox) handling.
when_to_use: Triggered by questions about channel performance, "which channel is best", funnel conversion rates, "is this step dropping too much", or any comparison between segments. Assumes you have also loaded analytics-diagnostic-method. Before declaring a cohort delta is causal, load causal-dag-builder to surface confounders; before recommending a rollback/ship action, load causal-evidence-checklist for a Bradford Hill grade. Auto-loaded when Clamp MCP's funnels_list, cohorts_compare, or traffic_breakdown returns segment data.

Channel and funnel quality

The specialist skill for judging quality, not just volume. Most "which channel is best" answers get the wrong channel because they look at sessions or even conversion rate in isolation. The right answer is a matrix.

Read `analytics-diagnostic-method` first if you haven't. This skill assumes you know MECE, triangulation, sample size, and Simpson's paradox. It adds channel-specific and funnel-specific mechanics.

Opening moves

1. Read `analytics-profile.md`. The model (B2B SaaS vs ecom vs lead gen) dictates which benchmark table applies and what "quality" even means. 2. Nail down the primary conversion event. Volume of "users who clicked the CTA" and volume of "paid signups" are not the same question. 3. Decide the time window. Recent-7-days is too short for most B2B funnels (sales cycle is longer than the window). Use 30 or 90 days for B2B; 7-14 days is fine for high-velocity consumer funnels.

The quality matrix

A channel or segment is never "good" or "bad" alone. It has three dimensions and you need all three:

              Volume       Engagement         Conversion
              (sessions)   (pages/session,    (CVR to primary
                           duration, depth)   event, revenue/visitor)

Direct        high         high               high        → flagship; defend
Organic       high         high               medium-high → compound invest
Referral      medium       high               medium-high → cultivate
Paid search   medium       medium             medium      → maintain, scale only if CAC allows
Email         medium       high               very high   → feed the list
Paid social   high         low                low         → vanity unless targeted well
Affiliates    varies       low-medium         low-medium  → audit for fraud, gate payouts
Unknown bots  high         very low           zero        → exclude, not a channel

Use this as a default sorting key, not a verdict. A specific channel in a specific business can land anywhere in the matrix. The point is: a channel with high volume and low engagement is vanity traffic; a channel with low volume and high conversion is probably underinvested.

Vanity-traffic fingerprints

Traffic you shouldn't be proud of, even if the session count is high:

• **Pages per session near 1.0**: they landed and left.
• **Median session duration under ~10 seconds**: they didn't read anything.
• **Engaged-session rate (GA4) under ~20%**: low intent.
• **Near-zero conversion rate to any downstream event**: they never did anything measurable.
• **Heavy geo concentration in a region you don't serve** (e.g. a US B2B tool with 60% of sessions from India).
• **UA / device concentration on a narrow band** (e.g. one browser version dominates in a way humans don't).
• **Referrer list contains suspicious domains** (crypto scams, SEO spam, semalt-style fake referrers).

Treat vanity traffic as a measurement problem before a marketing problem. Filter it; then re-read the channel matrix. The new matrix is usually very different.

Channel benchmarks (cite with context, never bare)

Use these for sanity-checking orders of magnitude, **always qualified by the business model row** from the user's profile. Never quote a single cross-industry number as "the benchmark"; the variance is too large to be useful without context.

Landing page conversion (Unbounce *Conversion Benchmark Report 2024*, 57M conversions, 41k+ pages)

| Traffic source | Median CVR | |---|---| | Email | 19.3% | | Paid social (blended) | 12% | | Paid search | 10.9% | | All sources, cross-industry median | 6.6% | | Google (vs Bing / Yahoo) | 11.3% (Bing 41% worse, Yahoo 95% worse) | | Instagram / Facebook | 17.9% / 13% | | YouTube / TikTok / X | 6–9% | | LinkedIn | roughly 4× worse than Meta for landing-page CVR |

**Important caveat**: Unbounce measures "conversion on the landing page" (form fill, demo request, download). Not paid signup. Numbers are much higher than primary-conversion CVR because a form submit is upstream of revenue.

Lead CVR by source (Ruler Analytics *2025 Conversion Benchmark Report*, 100M+ data points, 14 industries)

Overall average MQL CVR: **2.9%**. By source:

| Source | Avg lead CVR | |---|---| | Direct | 3.3% | | Paid search | 3.2% | | Referral | 2.9% | | Organic search | 2.7% | | Email | 2.6% | | Paid social | 2.0% | | Organic social | 1.5% |

B2B Tech is much harder: all sources roughly 1.5%, organic social as low as 0.3%. If the profile says B2B Tech, anchor on the B2B Tech row, not the cross-industry average.

Paid ads (Wordstream 2025)

• Google Ads average: CVR 7.52%, CPL $70.11.
• Meta Ads average: CVR 7.72%, CPL $27.66.

Wide industry variance: Finance/Insurance Google Ads CPL ~$116; Business Services CVR ~5%; Automotive Repair ~14%.

Ecom (Littledata *Shopify Benchmarks 2023*, ~2,800 sites)

• Average CVR 1.4%, top 20% >3.2%, top 10% >4.7%.
• Mobile 1.2% vs desktop 1.9%.
• Add-to-cart rate 4.6%; checkout completion 45% (top 10% >66%).
• AO