The short answer
Judge a channel by the customers it helps acquire and whether their contribution covers acquisition cost. Trace the path from attention to purchase, check for missing attribution, and change spend only after examining the full path. A SaaS founder reports that a post about paying influencers brought about 2,000 visitors and probably 10 or 15 clients.1 A home-services operator reports that only 10 to 20% of leads accepted despite a decent cost per lead.2
What founders did
8 real examples, each linked to the moment in the source. Results are self-reported.
- Reddit
Competitive_Tune_590
Some respondents said they had seen a comment and looked up the product later, although channel reports had not credited the forum.
Added a free-text source question at onboarding after forum traffic received little credit in analytics.
- X
Dev Raj Saini
Reported content with millions of impressions but very few meaningful enquiries.
Compared campaign impressions with meaningful enquiries in client work.
- Article
Emily Kramer
Advises supplying creators with tracking links and requesting their post analytics.
- X
Peter Quadrel
Argues against judging Meta campaigns by hook rate, click-through rate or cost per click alone.
- YouTube
Andy Tries Coding
Reported a video that outperformed his other posts; the supplied segments give no customer or revenue result from the ads.
Repeated a short-form video that outperformed the others, then ran paid ads on videos he considered winners.
- YouTube
Starter Story
The guest estimated something like 2,000 visitors and probably 10 or 15 clients from the posts.
A guest traced visitors from a post about paying LinkedIn influencers through to clients.
- YouTubeOnly 10 to 20% of leads accepted
Home Service Millionaire
The speaker reported decent cost per lead, but only 10 to 20% of leads accepting.
The speaker compared click and lead costs with lead acceptance during the offseason.
- YouTube
Auros
A guest described running regional campaign holdouts at Slack to compare exposed and unexposed areas.
Your playbook
Use this when
Use this when a channel is already running and you can connect some activity to enquiries, signups or customers.
You’ll leave with
Build a channel scorecard with conversion, attribution, acquisition cost, payback and a continue, revise or stop decision.
Define the customer outcome
- Do
Open a channel scorecard. Write down the channel, its intended audience and the customer action that would make it worthwhile. Record attention, qualified interest and acquired customers separately; define what counts as each for your product.
- Make
A scorecard with a defined customer outcome and the conversion path leading to it.
- Check
If you cannot distinguish attention from qualified interest and customers, fix those definitions before judging performance.
Preview the template
Channel: [channel] Audience: [intended buyer] Attention: [observable activity] Qualified interest: [observable action] Customer outcome: [observable conversion] Spend and effort: [costs] Attribution gaps: [possible missed sources] Customer contribution: [revenue less direct serving costs] Decision: [continue, revise or stop]
Why this step is in the plan
Dev Raj Saini says client content generated millions of impressions but very few meaningful enquiries. A guest in a Starter Story video traced a post through something like 2,000 visitors to probably 10 or 15 clients. Their cases show why exposure and customer results belong in separate scorecard fields.14
Check what attribution misses
- Do
Match tracked visits to subsequent enquiries and customers. Add a free-text source question where customers sign up or enquire, then compare answers with direct and branded-search traffic. Mark uncertain credit rather than assigning every conversion to a single touch.
- Make
An attribution-gap note beside each channel’s tracked and customer-reported results.
- Check
If customer accounts identify a channel the dashboard misses, retain it for further testing; if neither source shows buyer interest, reassess it.
Why this step is in the plan
Competitive_Tune_590 says forum visits appeared as direct traffic or earned no conversion credit, while onboarding answers revealed people who saw a comment and searched later. Emily Kramer advises brands to give creators tracking links and request post analytics. These accounts offer different checks on what a channel report captures.35
Calculate acquisition cost and payback
- Do
Add channel spend and attributable acquisition work to the scorecard. Divide by acquired customers for customer acquisition cost. Track each customer cohort’s revenue less direct serving costs over time, and note when cumulative contribution covers acquisition cost.
- Make
A channel CAC calculation and a customer-cohort payback entry.
- Check
If lead costs look acceptable but customers do not convert or payback is unsuitable, revise timing, offer or spend before scaling.
Why this step is in the plan
A speaker in Home Service Millionaire’s video reported strong cost per click and decent cost per lead, yet only 10 to 20% of leads accepted during the offseason. Peter Quadrel argues that hook rate, click-through rate and cost per click correlate poorly with revenue. Neither account supplies a payback figure.26
Test the decision before changing spend
- Do
Where feasible, compare similar unexposed and exposed audiences or regions while keeping the offer consistent. Otherwise, adjust the channel cautiously and watch acquired customers and payback, not just platform credit. Write the next change and decision in the scorecard.
- Make
A recorded continue, revise or stop decision with the customer-level result to watch next.
- Check
Increase commitment only if customer results support it; revise or stop if extra activity fails to improve the outcome that matters.
Why this step is in the plan
Decide what the channel is contributing
- What the scorecard shows
Attention is visible, but buyer interest is unclear
What to examine nextTrace enquiries or signups through to the customer outcome
Decision signalContinue measuring if you can identify meaningful progress; revise the message or audience if attention produces no qualified interest.
- What the scorecard shows
Customers appear under direct or branded search
What to examine nextReconcile tracked links with customers’ own source descriptions
Decision signalKeep the channel under review if customer accounts indicate demand that last-click reporting misses.
- What the scorecard shows
Leads arrive, but the economics are unclear
What to examine nextCalculate acquisition cost and compare it with customer contribution over time
Decision signalRevise or pause spend if customers do not cover acquisition cost within a payback period your business can support.
- What the scorecard shows
Reported results look strong enough to scale
What to examine nextCheck whether additional activity creates additional customers
Decision signalIncrease spend cautiously if downstream results improve; hold or reduce it if only platform-reported activity rises.
Continue
Keep measuring when qualified interest is progressing toward customers but the buying journey is incomplete.
Revise
Change audience, offer or timing when attention or leads are present but the next conversion fails.
Stop or scale
Stop spending when the customer path and economics remain weak; scale gradually when customer results and payback hold as activity increases.
Make it yoursChannel scorecard
A fill-in template for your own test
For each channel, record spend and effort; reach; qualified interest; acquired customers; customer contribution by cohort; tracked and self-reported sources; and the next continue, revise or stop decision.
When this works
- Use customer-level economics when acquisition costs and customer contribution can be connected.
- Use customer-reported sources and a comparison test when tracked conversions may miss earlier exposure.
When to avoid
- Avoid declaring success from clicks, impressions or inexpensive leads when the next customer action is weak.
- Avoid killing a channel solely from last-click reports when the buying path includes later searches or other touches.
Related questions
Can I reuse a sponsored creator post in my own ads?
Emily Kramer advises negotiating usage rights up front for relevant placements and suggests boosting suitable sponsored LinkedIn posts as thought leader ads.5 Treat reuse as a separate test: record the rights and added spend before judging the paid placement’s customer results.
Do I need a separate attribution tool to judge Meta ads?
Peter Quadrel argues that Meta attribution plus CAPI supplies enough data without additional attribution tools until mid-eight figures; his post does not specify what that figure measures.6 Check whether your current setup answers the customer and payback questions before adding a tool.
What if subscription revenue arrives after acquisition costs?
Devesh Kedia observes that, in BlackBuck’s reported results, new-device acquisition costs arrive upfront while annual subscription revenue is recognised over 12 months; he says renewals matter to the spending assessment.9 Track cohorts through renewal and compare cumulative customer contribution with acquisition cost instead of deciding from one reporting period’s margin.