Creators

From 7% to 21% open rate in less than 2 months.

Luana Carolina tripled her open rate and multiplied absolute opens in less than 2 months with Email Intelligence, while doubling her send volume.

182%

in open rate

5.1x

more absolute opens

3.9x

more clicks

Luana Carolina is a digital creator who runs an email relationship operation with her audience via ActiveCampaign. Her weekly newsletter is the main connection channel with her base and the starting point for her product sales campaigns.

Goals

  • Increase newsletter open rate
  • Identify which contacts were engaging and which were hurting reputation
  • Reduce the impact of disengaged contacts on sends
  • Increase sales campaign conversion

Solutions

  • Email Intelligence installation with automatic field and segmentation creation
  • Progressive testing strategy by engagement level
  • Calibrated cleanup of contacts with no interaction
  • Targeted sends to engaged segments

The problem

When Luana reached out to Inbox, her ActiveCampaign list had roughly 104,000 contacts and her weekly newsletters went out to the entire base, with no segmentation. The result repeated campaign after campaign. Average open rate of 7.49% across newsletters 102 to 105. Clicks below 0.4%.

Over 40,000 contacts in the base were highly disengaged or had no recorded interaction at all. They received every email, every week, even without ever opening one. They diluted metrics, consumed paid send volume, and potentially hurt sender reputation with providers.

The diagnosis was clear, but with no individual visibility into each contact, there was no way to act. ActiveCampaign showed campaign performance. It did not show contact behavior.

The optimization strategy

Email Intelligence was installed on January 30, 2026 and onboarding happened on February 2. The strategic decision from the start was not to tackle everything at once. Instead, we chose a progressive testing cycle. Each phase answered a different question and each answer calibrated the next phase. The goal was to understand how the base behaved before deciding how to treat it.

Phase 1: Diagnosis

Before touching any send, we needed to understand the base's actual behavior. In the two weeks following installation, we segmented newsletters by engagement level. Highly Engaged, Moderately Engaged, Lightly Engaged, three levels of Disengaged, No Interaction, and Undefined. Each segment received separate sends.

The numbers confirmed the hypothesis and revealed the scale of the problem. Highly Engaged contacts opened between 19% and 44% of messages. No Interaction contacts, between 1.5% and 2%. They were all receiving the same email on the same day, and the final average was diluted to the historical 7%.

Phase 2: Careful cleanup

With the mapping in hand, the first action was to stop sending to the No Interaction segment. The next newsletter went out without it. The Highly Engaged segment, which previously sat around 19% to 44%, jumped to 48.26% open rate. It was the highest rate of the entire period.

The next temptation was to reactivate inactive contacts that were still inside the engaged segments. We tested. Opens came in at 21% to 36%, acceptable numbers. But bounce sat between 4% and 8%, and unsubscribes between 3% and 5%, high numbers that touched the domain's reputation directly.

The practical conclusion: reactivation works, but it has a cost. It can't be a continuous action. It needs to be calibrated, isolated, and monitored.

Phase 3: Calibration

With a cleaner base, two more tests came to better understand the nuances of who was still on the list.

The first tested ActiveCampaign's predictive send against fixed-time sending. In 112,000 sends, half went out at 4pm and half via AC's predictive algorithm. The fixed time won by 1.33 percentage points, 19.72% against 18.39%. A small difference per campaign, thousands of opens per month in aggregate. Predictive sending was dropped from the strategy.

The second tested the temporal granularity of the disengaged. Lightly and Moderately Disengaged were split into four windows: 30 days, 60 days, 120 days, and more than 120 days without interaction. The 30-day group still opened between 13% and 16%. Beyond 60 days, they collapsed. Beyond 120 days, they opened less than 2.2%. The decision was to stop sending to that last group. Continuing to fire to them was burning reputation for nothing.

Phase 4: Conversion

With the base segmented and calibrated, campaigns changed in nature. Four rounds of sales campaigns went out, targeted at the engaged segments. The result consolidated the thesis of the previous phases. There were 10 direct sales. One hundred percent came from the engaged segments. Zero from the disengaged.

The reading is simple: segmentation is not just for raising open rate. It is for selling better to those willing to buy.

The results

In January, before Email Intelligence, Luana sent every newsletter to the entire base and got an average open rate of 7.49%. In March, with the base segmented and the send strategy calibrated, the rate rose to 21.12%. A +182% evolution in two months.

The compound effect was even more expressive in volume. Absolute opens grew from 33.9k to 174.2k per month, a 5.1x increase. Clicks went from 1.3k to 5.1k, almost four times more. And all of this happened while total send volume more than doubled, going from 382k to 830k monthly sends.

The honest reading of this result is that send volume grew alongside opens, and that is not a coincidence. Email Intelligence made it possible to send more to those who engage and stop spending sends on those who haven't opened in months. The base is the same. The send behavior changed.

Sales campaigns confirmed the pattern. Four rounds, totaling 10 direct sales. One hundred percent of the sales came from the engaged segments. Zero from the disengaged. It is not an absolute surprise, but it is an empirical confirmation of the obvious that no one had been able to measure before.

Want to see these results in your account?

Luana started with 104k contacts and 7% open rate. Email Intelligence is in free beta now. Same install. Same process.

Join the Beta

Free. No card. No commitment.

See other cases