Best Chartsy Growth Feature Best Practices

Most bad Growth reports aren't a Chartsy problem, they're a setup problem: a missing UTM, an untested signup event, or a channel judged on three days of data. This checklist covers the habits that keep attribution accurate and the mistakes that quietly break it.

Table of Contents

• Do: Standardize UTMs Before You Trust Any Channel Report
• Do: Wire Up Signup and Billing Events Correctly
• Do: Compare Channels on Customers and MRR, Not Traffic
• Don't: Judge a Channel on One Week of Data
• Don't: Let "Direct" Traffic Hide a Broken UTM
• Do: Turn Findings Into a Recurring Review
• FAQ
• Conclusion

Do: Standardize UTMs Before You Trust Any Channel Report

UTM tags are the glue that holds traffic to revenue. If every campaign uses a different naming style, attribution turns into a mess before Chartsy ever sees the data.

Define a naming convention, source, medium, campaign, content, and put it somewhere every teammate can find it. Use utm_source=google, utm_medium=cpc, utm_campaign=summer-launch as the pattern, then apply it to every paid, social, and referral link without exceptions.

Audit what already exists. Export recent campaign links from your ad platform and scan for missing or misspelled parameters. Fix what you can and flag the rest so old data doesn't get compared against clean new data.

Once tags are consistent, run a test: fire a tagged URL, sign up with a dummy email, and confirm Growth shows the correct source. If it shows "direct," the tag isn't reaching the signup event.

For background on why this matters, see the Wikipedia entry on marketing attribution, which covers how attribution models depend on clean source data.

Key Takeaway: A messy UTM scheme caps the accuracy of every report built on top of it, no matter how good the analytics tool is.

Do: Wire Up Signup and Billing Events Correctly

Growth needs two connections to work: the visitor's source and the subscription record that follows it.

The first comes from a small tracking script placed before the closing tag on your main site. It reads UTM parameters and sets first-touch and last-touch cookies that persist across subdomains, so a visit that starts on a marketing page and ends with a signup on an app subdomain stays linked.

The second comes from the signup event itself. A no-code form can use a data-chartsy-signup attribute; a custom flow can call chartsy.trackSignup() when the form completes. Either way, test it with a real signup before trusting the data.

Finally, connect Stripe or Paddle from Chartsy's Settings and approve the OAuth request. Once that's live, Growth matches every signup to its subscription and starts rolling up MRR, churn, and lifetime value by source.

For background on the revenue concepts Growth surfaces, see Recurring revenue.

Key Takeaway: Growth is only as accurate as its weakest link, tracking script, signup event, or billing connection, so test all three before reading the report.

Do: Compare Channels on Customers and MRR, Not Traffic

Once data is flowing, open the Channel Attribution view and add four columns: New Customers, MRR, Total Revenue, and Churn Rate. Traffic alone won't be one of them.

Look for the gap between visits and outcomes. In one recent comparison, an 80-visit mention in a partner's newsletter converted into 9 paying customers and $6.4K in new MRR, while a 1,400-visit paid search campaign converted into 13 customers and $4.1K in new MRR. Ranked by traffic, paid search wins easily. Ranked by MRR per visitor, the newsletter mention wins by a wide margin.

Use that gap to reallocate, not react. If a channel's churn runs above 10%, look at onboarding before assuming the channel is bad. If a channel's ARPU is well above average, it can justify a smaller volume of customers.

Chartsy lets you slice the same comparison by plan tier, country, or cohort, so you can see whether a channel brings premium buyers or mostly free trials before you commit budget to it.

Pro Tip: Rank channels by MRR-per-visitor at least once, alongside raw customer count. It surfaces small, high-fit channels that a traffic-first view buries.

Don't: Judge a Channel on One Week of Data

A new channel's first week almost never tells you whether it's good. Signups need time to convert to paying customers, and paying customers need at least one billing cycle to show whether they stick.

Give a channel a defined test window before deciding anything, long enough to include one full renewal cycle if the plan is monthly. Write down the question you're testing before the window starts, for example "can this source produce 10 paying customers in 30 days," so a good or bad week doesn't get reinterpreted after the fact.

Treat percentage swings on a small base with suspicion. A channel moving from one customer to two is a 100% increase and also nearly meaningless. Pair every percentage with the underlying count.

Don't: Let "Direct" Traffic Hide a Broken UTM

A rising "direct" bucket is usually a tagging problem wearing a disguise, not organic growth. Growth correctly refuses to force an untracked visit into a channel, which means broken tags surface as more direct traffic rather than an error message.

Check "direct" first whenever a channel's numbers drop unexpectedly. A misspelled utm_source, a redirect that strips query parameters, or a link shortener that doesn't pass UTMs through are the usual culprits, and each looks identical in the report: a visit with no source.

Reconcile "direct" against known sends. If an email campaign should have produced 200 tagged visits and only 40 show up under that source, the other 160 are likely sitting in "direct."

Do: Turn Findings Into a Recurring Review

An insight that lives in a screenshot dies in a screenshot. Set a recurring cadence, weekly for fast-moving channels, monthly for stable ones, and build a lightweight template: channel, visitors, new customers, MRR, churn, and one action item.

Save that template as a Chartsy dashboard widget so anyone can fill it in without rebuilding it from scratch. Invite the stakeholders who actually make budget calls, marketing, product, finance, and walk through the top three and bottom three channels each time.

Close the loop by tracking what happened after each action. If you raised spend on a high-ARPU channel, note the MRR change next review. If you changed a landing page on a high-churn source, record the new churn rate. That record is what turns a one-off report into a playbook.

FAQ

How often should I review Chartsy Growth data?

Weekly for channels you're actively testing or spending on, monthly for stable, established channels. Reviewing too often on a slow-moving channel just adds noise from a small sample size.

What's the most common UTM mistake teams make?

Inconsistent capitalization or spelling across the same channel, "Newsletter," "newsletter," and "news-letter" get counted as three separate sources instead of one, which fragments the report before analysis even starts.

Should I act on one unusually strong week from a channel?

Not on its own. Confirm the pattern holds for at least one full billing cycle, and check whether the base size is large enough that the result isn't just noise.

Do I need a developer to keep UTMs and tracking clean?

Not for the basics. A no-code signup form only needs a data-chartsy-signup attribute, and a shared naming doc keeps UTMs consistent without engineering involvement. Custom signup flows and link-shortener setups are the cases that usually need a developer's help.

How many channels should I compare in one review?

Three or four at a time. Comparing everything at once tends to bury the channel that actually deserves a decision under ones that don't need one yet.

Conclusion

None of this requires new tooling, it requires discipline: consistent UTMs, tested events, comparisons built on customers and MRR instead of traffic, and a review cadence the team actually keeps. Get those five habits right and Chartsy Growth's numbers become something you can act on instead of something you have to double-check.