How to leverage your marketing data

Pick one question, collect only what answers it, and act on the difference.

Impact
Medium
Effort
High
Filed under
Data
Updated

To put your data to work in marketing, pick one question you need answered, collect only the data that answers it, put it in one place, and change something based on what it shows. Most small teams do not lack data. They have too many dashboards and no question. A good first question is concrete, like "which channel brings the signups that pay?" or "which action in the first week predicts that a user stays?"

How to do it

  1. Write the question and the decision. "If paid signups mostly come from one channel, we will spend more there." If no result would change what you do, skip the question.
  2. Collect the minimum. For a website, that is traffic with its source and a conversion event (signup, purchase, demo request). For a product, add a handful of named events: signed up, did the first key action, invited a teammate. Fathom, for example, is a privacy-focused alternative to Google Analytics that works without cookies, with plans starting at $15 a month. Plausible is another, and Google Analytics has a free plan.
  3. Add revenue. Traffic numbers do not tell you who pays. Subscription analytics tools such as Baremetrics read your billing data (it lists Stripe, Chargebee, Braintree, Recurly, Shopify and others) and give you subscription metrics and cancellation insights. Match those to source and signup date.
  4. Put it in one place. A spreadsheet with one row per customer (source, signup date, first action, plan, status) is enough to start. If you run several ad and email platforms, a connector like Supermetrics pulls their data together, and Databox offers dashboards across 130+ integrations with a free-forever plan. Do not buy either until the spreadsheet becomes the bottleneck.
  5. Compare groups, not totals. Split customers by source, by plan, by whether they did the key action in week one. Look for the groups that retain or pay more. For the second kind of question, see key activation metric.
  6. Change one thing and re-check. Move budget, rewrite a landing page, or add an onboarding step. Wait long enough to get a meaningful number of conversions before you decide it worked.
  7. Review on a schedule. A weekly 20-minute look at the same five numbers beats a rebuilt dashboard every month.

What it looks like in practice

Say you sell a scheduling tool with a free trial. You track four things: source of each signup, trial start, whether the user created their first booking page, and whether they paid by day 30. After two months you split trials by source and find one referral partner sends fewer signups than search, but more of them create a booking page. You move your next email and landing page effort to that partner. You then check whether paid conversions from that source stay higher over the next month, instead of assuming they will.

When it backfires

  • Tracking without consent. If you use cookies or similar tracking on visitors in the EU, the rule from GDPR guidance is to "receive users' consent before you use any cookies except strictly necessary cookies." Cookie-less analytics reduce the problem but do not remove your other privacy duties.
  • Small numbers. A 10-person cohort tells you almost nothing. Wait for more data or use interviews instead.
  • Correlation is not a cause. Users who use feature X may retain better because they were more committed from the start. Test before you restructure the product around it.
  • Tool churn. Tools disappear. Splitbee has been sunset, Panelbear's address now redirects to Cronitor's real user monitoring page, and Indicative is now mParticle Analytics. Keep an export of your raw data.
  • Vanity dashboards. Page views and followers feel good but rarely lead to a decision.

Questions people ask

How do I use data to develop a strategy? Start from a business question, find the numbers that bear on it, and compare segments to find where results differ. The strategy is the change you make because of the difference.

What tools connect customer usage data with revenue? Revenue sits in your billing system and usage sits in your product analytics. Billing-based tools such as Baremetrics cover the revenue side. To tie usage to revenue, send both with the same customer ID to a spreadsheet or warehouse and join them there.

What does it mean to use data in marketing? Using what you already collect to make a specific decision, such as where to spend, what to build or whom to contact.

Sources

Want word-of-mouth like this on autopilot?

CommunityMentions gets your brand mentioned in the Reddit threads your buyers already read, plus YouTube, Medium, LinkedIn and X. We aim for 5 to 10 comments a day.

See how it works →

Related tactics

All 133 tactics →