Growth & Outbound Case Study · Airtm

AI-Powered Outreach Engine

An AI-powered outbound system built to identify high-fit prospects, personalize outreach and turn conversations into commercial opportunities for Airtm.

Client
Airtm
Services
Growth Engine
Scope
Strategy · Build · Deployment · Optimization
Markets
LATAM · USA

Impact

43.8%

Recruiter response rate

up from 18%

>71%

Brand Partnerships response rate

optimized track

$60M

Annual payment volume

AI unicorn · top-3 partner

800+

New users

nearshore talent partner

4

Recruiter partnerships

first scaled cohort

8

Countries reached

initial recruiter cohort

01

The Opportunity

Airtm has a large user base of freelancers and remote professionals who receive payments from companies abroad. Its transaction and behavioral data offered visibility into how these users work, how much they receive, and which companies and platforms they work with.

The starting point was a simple question: who else should Airtm be talking to?

That question led to two opportunities:

Track A

Community & Recruiters

Recruiters already have access to professionals working, or looking to work, for international companies. Instead of acquiring these users one by one, Airtm could build relationships with people who already had their attention.

Track B

Brand Partnerships

Companies expanding into LATAM could benefit from Airtm in two ways: accessing its freelancer base and using its infrastructure to pay distributed teams.

This became the starting point for the outbound system.

02

Defining the ICPs

An ICP was defined for each track, combining Airtm’s first-party data with external market signals.

For Recruiters, the priority was profiles whose audiences had a high concentration of remote and international professionals, rather than simply recruiters with large followings.

For Brand Partnerships, the focus was on companies expanding into LATAM, hiring freelancers or remote professionals, looking for qualified talent in the region, or managing international payments to distributed teams.

Airtm’s transaction data was especially useful here: it revealed user profiles, payment behavior and which companies were already paying people in its ecosystem.

The ICPs weren’t fixed. They were hypotheses, refined continuously as the outreach started generating data.

Community & Recruiters

High share of remote talentInternational audiencesFit over follower count

Brand Partnerships

Expanding into LATAMHiring freelancersSeeking LATAM talentPaying distributed teams
03

Discovery & Outreach

The work started with a manually selected group of high-fit profiles for each ICP, plus an integration between LinkedIn Sales Navigator and Claude to research, qualify and find additional prospects.

As the system accumulated context from validated profiles and campaigns, it began suggesting new prospects with similar characteristics.

The process:

  1. ICP
  2. Seed profiles
  3. AI-assisted discovery
  4. Qualification
  5. Outreach
  6. Results
  7. Refinement

Loops back to the ICP after every round.

Human approval stayed at key points, while AI handled much of the repetitive research, personalization and analysis.

The same infrastructure supported two different approaches:

Community & Recruiters

Volume · High

Higher-volume, automated outreach focused on building distribution partnerships.

Brand Partnerships

Volume · Low

Lower-volume, higher-value outreach to talent platforms, staffing companies and AI-related businesses, with propositions around talent access, payments or both.

04

Iteration & Results

The system wasn’t built around a fixed sequence. The idea was to learn from each round and adjust.

Recruiters

The first campaigns showed clear differences by market, recruiter profile and audience. Those results shaped the messaging and personalization for each segment.

Response rate improved from 18% to 43.8%.

In the first scaled cohort:

  • 43 recruiters identified
  • 32 contacted
  • 14 responses
  • 43.8% response rate
  • 4 partnerships activated
  • 8 countries represented

Benchmark: HeyReach 2026 median reply rate, 22.2% (96,051 LinkedIn campaigns).

Brand Partnerships

Here, the main learning was around segmentation. The first conversations showed which types of companies had the strongest fit with Airtm’s proposition.

With that signal, both the proposition and the best person to approach within each company were refined.

The optimized track reached a >71% response rate.

Response rateHeyReach 2026 median reply rate · 22.2%

  • Recruiters · first campaigns

    18%Recruiters · first campaigns18%HeyReach 2026 median reply rate: 22.2%

  • Recruiters · optimized

    43.8%Recruiters · optimized43.8%HeyReach 2026 median reply rate: 22.2%

  • Brand Partnerships · optimized

    >71%Brand Partnerships · optimized>71% · Lower boundHeyReach 2026 median reply rate: 22.2%

96,051 LinkedIn campaigns

Recruiters · first scaled cohort8 countries
  • Identified
    Identified4343
  • Contacted
    Contacted3232
  • Responded
    Responded1414
  • Partnerships
    Partnerships44
View the data as a table
MeasureValue
Recruiters · first campaigns18%
Recruiters · optimized43.8%
Brand Partnerships · optimized>71%
HeyReach 2026 median reply rate22.2% (96,051 LinkedIn campaigns)
Recruiters · first scaled cohort · Identified43
Recruiters · first scaled cohort · Contacted32
Recruiters · first scaled cohort · Responded14
Recruiters · first scaled cohort · Partnerships4
05

Business Outcomes

Response rates were useful for improving the system, but the goal was to generate real commercial opportunities.

Fast-Growing AI Unicorn

USD 60M

in annual payment volume, making it one of Airtm’s top 3 partners.

4x faster

The system identified and opened an opportunity with a fast-growing AI company that had recently reached unicorn status after raising more than $100M from leading technology investors.

The value was in how it was prospected. The need, the timing and the approach were so accurate that, once the opportunity was qualified and handed to Airtm’s Sales team, the deal to use Airtm as its payment method closed with a sales cycle 4x faster than average.

Nearshore Talent Company

800+

new Airtm users: professionals paid across Brazil, Argentina, Colombia and Bolivia.

The outreach also opened a conversation with a nearshore talent company connecting global businesses with specialized professionals across Latin America.

The relationship developed across both sides of the original hypothesis: Airtm helped the company source candidates from its user base, while the partner adopted Airtm’s virtual accounts to pay its talent, bringing 800+ new users to Airtm across Brazil, Argentina, Colombia and Bolivia.

This validated the original Brand Partnerships thesis: the same relationship could create value through both talent access and payment infrastructure.

BrazilArgentinaColombiaBolivia
06

What Was Built

The system was designed, built and deployed end-to-end.

The workflow covers research, AI-assisted drafting, human approval, outreach, response monitoring, qualification and handoff to Sales.

  1. Research
  2. AI drafting
  3. Human approval
  4. Outreach
  5. Response monitoring
  6. Qualification
  7. Handoff to Sales

Today, the system continues to run autonomously, with human approval and periodic check-ins used to review performance, adjust the ICPs and messaging, and decide where to focus the next iteration.

Shared with Airtm’s permission. Partner companies are kept anonymous.

Want a system like this for your business?

Growth Engine brings the same approach—clear ICPs, AI where it helps, humans where judgment matters—to your growth.

Let’s talk about growthExplore Growth Engine