# How Convenia centralized data operations for an HR platform with a 6x return on its Erathos investment

> Convenia went from loose pipelines on VMs with zero observability to a centralized data operation in BigQuery, with Slack alerts and a 6x return.

Source: https://www.erathos.com/en/clientes/convenia
Em português: https://www.erathos.com/clientes/convenia

## About

- Convenia
- Sector: HR Tech
- Location: São Paulo, SP
- Employees: 150
- Erathos customer since: November 2024
- Data stack: BigQuery (data warehouse), dbt (modeling), Astronomer/Airflow (orchestration), Power BI and Claude (visualization), Erathos (ingestion)
- Connected services: HubSpot, Conta Azul, Google Ads, Facebook Ads, MongoDB, Vindi, Convenia

Convenia is a people management platform for small and mid-sized Brazilian companies, with more than 12 years in the market, serving more than 3,000 companies and impacting over 500,000 employees. It centralizes recruiting, payroll, vacation, performance reviews, internal communication, document signing and strategic HR reporting.

Convenia, an HR management platform serving thousands of Brazilian companies, had no formal data team: data work fell on whoever had spare time on the RevOps team, with loose pipelines running on VMs and no observability at all. When Bernardo Cardozo joined to build the data function, Erathos was his first hire, replacing Kondado right from the start.

Today Convenia runs all data ingestion from HubSpot, Conta Azul, Google Ads, Facebook Ads, MongoDB, Vindi and its own internal accounting API with a lean team of 3 people, with automatic Slack alerts instead of finding problems only when a report showed up stale. The investment in Erathos delivers Convenia an estimated return of 6 times the amount invested.

## The results

- **~1 month → 15 min** — to get a new data source flowing into BigQuery
- **3 people** — run all ingestion, modeling and analytics for a 150 employee company
- **~6x ROI** — about R$147k in estimated net savings per year

## The challenge

Before Erathos, Convenia had never had a formal data team. Analytics requests fell on the RevOps team, which was already handling sales planning and commissions and kept pushing data structuring down the list. The result was loose pipelines running on VMs, built by whoever was available at the time, with no visibility into what was working and what was not.

Bernardo Cardozo joined the company precisely to fix this, tasked with structuring the entire data function: not just RevOps, but product and finance as well. At the time Convenia used Kondado, but ran into a lack of support and missing key connectors, such as Vindi, the billing system the company uses.

With no observability, the team only found out a pipeline had broken when someone needed data for a meeting and the report simply was not up to date. Bernardo estimates this happened at least once a week.

## The solution

Erathos came through a direct referral from Convenia's CEO, Marcelo Furtado, who already knew Gelson and Luca before Bernardo took over the data function. Bernardo considered an open source solution, but ruled it out because of the maintenance cost that would fall on his own team.

What tipped the decision was not just the list of ready to use connectors, but the Erathos team's willingness to build together: since signing, three to four new connectors have been created, including a custom connection to Convenia's internal accounting API, something larger tools on the market did not offer.

Today most syncs run hourly (HubSpot is the most critical case), with some connectors on a weekly refresh when the source does not support an incremental cursor. Any failure triggers a Slack alert, and issues are solved quickly and directly with the Erathos team.

One detail of the relationship: Convenia itself is also a data source in the Erathos catalog. Any company using Convenia for HR management can connect that data to its warehouse in a few minutes.

## The results

- RevOps, product and finance data centralized in a single reliable database (BigQuery), replacing a fragmented setup run by whoever had spare time.
- No more "finding out from a stale report": Slack alerts in place and visibility across the entire ingestion flow, with Airflow being implemented to strengthen that control even further.
- Connecting a new source already in the Erathos catalog takes about 15 minutes, versus an estimated month to build a connector from scratch in house.
- Custom connectors that did not exist on the market, including a direct connection to Convenia's internal accounting API.
- Data team time shifted from pipeline maintenance to governance: defining business rules with each team and curating what is actually worth analyzing.

## Return on investment (ROI)

Bernardo estimates that without Erathos, Convenia would need at least one more person dedicated to the data team just to handle ingestion, and it would take much longer to get data from new connectors into the data warehouse.

Using the average Brazilian market salary for a data engineer (R$8,000 per month, plus payroll taxes), the estimated cost of that hire would be R$172,800 per year. After the investment in Erathos, estimated net savings come to about R$147k per year, roughly 6 times the amount invested.

**~6x the investment, with about R$147k in estimated net savings per year**

## What's next

Convenia is about to test Erathos custom connectors, which will make it even easier to connect internal APIs like the one already custom built for them.

Bernardo's next big goal is closing the one gap left to get the full customer journey, from lead to in-product usage: bringing sales and support call transcripts into the data warehouse. With two new tools being contracted for this, a sales call coaching tool and an AI support bot, the aim is to identify churn risk more objectively, instead of relying on a manual tag logged in HubSpot after each meeting.

> We've always worked with data, but we never had a formal data team. It was driven by RevOps, and since their routine is very hectic, it always ended up on the back burner. I joined to structure and centralize data for the whole company, not just revenue.

The main problem with our previous tool was the lack of support and some missing connectors, like Vindi. We would have had to build pipelines on the side to cover that. Having close contact with Gelson and Luca was one of the main things that gave me confidence we'd have the support we needed whenever we had to create a new connector or fix a problem.
>
> — Bernardo Cardozo, Data Coordinator at Convenia

> Today, whenever something goes wrong, I get a Slack notification. I check whether the next run went through, and if I can't figure out why, I talk directly to the Erathos team. Without this kind of partner on a lean data team, we'd lose a lot of time chasing solutions instead of building other projects.

If I were recommending Erathos to another analyst or data engineer evaluating it, I'd say it's a Brazilian company with global potential. What really makes the difference isn't just the product, because there are other players with the same goal. It's having access to the people building the system, being able to give feedback and being heard. Every time I gave feedback, it was addressed. I don't know any other player that commits to growing alongside your team the way Erathos does.
>
> — Bernardo Cardozo, Data Coordinator at Convenia