Ask Astra

Which channel brings paying customers?

Stop optimizing signups. Optimize for the channels that pay.

Business insightMarketingWeekly Friday 4pm. Monthly attribution audit.
Free to startNo credit card requiredUpdated Apr 2026

You'd think the channel with the most signups is your best — Astra ranks by paid customers + LTV and shows the picture is usually flipped.

The short answer

Astra ranks every acquisition channel by paying customers (not signups) and tells you which to double down on. She pulls the last 90 days from GA4 + PostHog + post-signup survey, attributes each Stripe paid signup to its real channel source, then computes per-channel: paid customer count, CAC, LTV, payback period, signup-to-paid conversion rate. The output is a Lark report ranked by 'paid customers per dollar spent' — the metric that actually matters. The signup volume leader and the paying customer leader are usually different channels by 2-3 ranks. Common pattern: paid social drives volume but converts poorly; organic search and partnerships drive low volume but convert at 3-5x. The recommendation each week is one specific reallocation: cut $X from worst channel, redirect to best. Most teams realize they should be spending 60% less on the channel they're spending 60% on.

How Astra actually does it

  1. 1
    Attribute paid signups to channel

    Stripe paid signups last 90 days. Match each to original channel via GA4 first-touch + PostHog + self-report. Resolve conflicts.

    StripeGA4PostHog
  2. 2
    Compute per-channel economics

    Per channel: paid customer count, CAC (true cost / paid count), LTV (cohort retention × gross profit), payback period, signup-to-paid %.

  3. 3
    Rank by paying customers per dollar

    Sort by ROI = paid customers / dollar spent. Volume leader and ROI leader often differ — that's the actionable gap.

  4. 4
    Identify the reallocation move

    One specific recommendation: cut $X from lowest-ROI channel, redirect to highest-ROI channel that has scaling room. Project impact on paid customers next month.

  5. 5
    Weekly report + monthly deep-dive

    Weekly Lark report on top 5 channels by paid customers + the recommended reallocation. Monthly: full attribution model audit and confidence scoring.

    Lark

What it looks like in chat

Which channel actually brings paying customers?
Astra
Last 90 days. **Volume vs revenue rankings are flipped:** **By signup volume (what you've been optimizing):** 1. Meta Ads: 412 signups 2. Google Ads: 287 3. Organic search: 198 4. LinkedIn ads: 142 5. Reddit: 89 **By paid customers (what matters):** 1. **Organic search: 47 paid (24% conversion)** ✅✅ 2. Reddit: 28 paid (31% conversion) ✅ 3. LinkedIn ads: 19 paid (13% conversion) 4. Google Ads: 14 paid (4.9% conversion) 5. Meta Ads: 6 paid (1.5% conversion) ❌ **ROI (paid per $ spent):** - Organic: ∞ (no direct spend) - Reddit: ∞ (no spend) - LinkedIn: $0.005/$ → $216 CAC - Google: $0.002/$ → $428 CAC - Meta: $0.001/$ → $1,000+ CAC ❌❌ **Reallocation: cut Meta ads entirely.** $4,200/mo redirected to Reddit community + organic SEO investment. Projected impact: -6 paid/mo from Meta loss + ~25 paid/mo from organic/Reddit boost = **+19 paid customers/mo, $0 net spend change.** Do you want me to pause Meta campaigns and start the Reddit community plan?
What you get back

Weekly Lark report: top 5 channels by paid customers, CAC per channel, recommended reallocation move with projected impact.

Cadence

Weekly Friday 4pm. Monthly attribution audit.

Ask Astra this right now

We'll spin up your workspace, hand the prompt to Astra, and you see the answer in 60 seconds. Free.

Try this with Astra

Frequently asked questions

What if I have low paid customer counts per channel — is the data significant?

Below 10 paid customers per channel/quarter, ranking is noisy. She flags low-confidence channels with a ⚠️ and recommends extending the analysis window to 6 months for those. For early-stage products with <100 paid customers total, she pivots to leading indicators (signup-to-trial, trial-to-paid) instead of paid customer counts.

How does she handle multi-touch attribution for B2B with long sales cycles?

Default first-touch for cycles <30 days, time-decay multi-touch for longer cycles. She'll show both views so you can see where deals start vs where they close. Most B2B has 'webinar started, organic search closed' patterns — first-touch and last-touch matter differently for budget decisions.

What if my best channel can't scale (organic SEO takes years)?

She'll flag scaling potential per channel: organic SEO compounds slowly but ceilings are high; partnerships scale linearly with effort; paid scales fast but CAC rises. The reallocation recommendation factors in scaling room — she won't tell you to '10x organic' if it's a 12-month bet.

Can I see channel performance by customer segment?

Yes — she'll break down by plan tier (Starter/Pro/Enterprise) or ICP segment if defined. Often the channel mix differs dramatically: Meta brings cheap signups but Starter-tier only; LinkedIn brings expensive signups but Enterprise-tier. The right channel depends on the segment you want to grow.

Run your one-person company.

Hire your AI team in 30 seconds. Start for free.

Free to start · No credit card required · Set up in 30 seconds