agiliz
Case studies

Case 01

€80k/year

of manual publishing replaced by an automated engine

Before
Listings published by hand, portal by portal
After
20 marketplaces run by an AI engine

Ad Engine

The AI engine that runs 20 marketplaces and freed up 4 people to sell

A single engine that reads the sales plan and publishes, reactivates and removes listings in more than 10 countries.

  • Mobility
  • Integrations and automation
Team
4 people
To production
1 month
In production
since Sep 2026

Results

Indie Campers sells motorhomes from its own fleet on classified ad marketplaces in Europe, the United States, Australia and New Zealand. Listings were published by hand, portal by portal, by an external team. The Ad Engine reads the RV Sales team's plan, connects to each marketplace through whatever technical route it allows and, since 1 October 2026, has been the only way the company manages its listings.

Client
Indie Campers, RV Sales department
Sector
Mobility and Tourism
Technologies
Marketplace write APIs, XML and CSV feeds, reconciliation engine, AI performance analysis, dashboard in Portuguese and English
Duration
Discovery in August 2026, in production since September 2026
Team
David Esposito, Guilherme Veríssimo, Lourenço Barros, Rui Pereira

01The opportunity

  • Manual publishing with a labour ceiling

    Every listing was created by hand in each portal's form, with different categories, languages and rules. The external team managed about 100 listings a day. That number was a capacity limit, not a business goal.

  • Paid slots going unused

    Indie pays for professional packages on several platforms, each with a fixed number of slots. With no consolidated view, paid accounts had empty slots and nobody knew how much money was sitting idle.

  • A different protocol on every marketplace

    Some platforms have a write API, others only accept files they fetch themselves, others have no integration at all. The accounts held 9,429 listings, 5,121 of them paused and 3,988 expired, and nobody knew which were worth recovering.

02The solution

  1. 01

    One click to publish.

    The RV Sales team sees the motorhomes ready to list on the dashboard and approves each listing with one click. The engine does the rest: the listing goes live on all the right marketplaces, with the title, photos and description for each country and language.

  2. 02

    Everything on one dashboard.

    Active listings, paid slots used and free, and waste in euros per month, by account and by platform. No opening portal after portal.

  3. 03

    AI that analyses every listing.

    The engine tracks each listing's performance (views, enquiries and time to sale) and cross-references it by platform, country and model. It spots what works and what doesn't: for example, listings that open with exterior photos get more enquiries than those that open with the interior.

  4. 04

    Recommendations ready to apply.

    Each pattern reaches the team as a concrete recommendation, so every motorhome is listed where it sells best, with the content that gets the most enquiries.

Listing analysis and slot optimisation

With publishing automated, the second phase of the contract puts AI in charge of deciding where and how each motorhome should be listed, and makes sure no paid slot sits idle.

  • AI performance analysis

    Views, enquiries and time to sale for every listing, cross-referenced by platform, country and model. The engine identifies the worst-performing listings and why.

  • Patterns turned into actions

    Each pattern reaches the team as a concrete recommendation. Example: listings that open with interior photos get fewer enquiries than those that open with the exterior.

  • Paid slot optimisation

    Real-time occupancy of each professional package and waste in euros per month. Empty slots are reassigned to the vehicles in highest demand.

  • Smart fleet distribution

    Each motorhome goes to the platforms where it sells best, with the content that gets the most enquiries. The team stops deciding on gut feeling.

  • 100%

    Of paid slots filled, phase 2 target

  • €8,200

    Saved per month on paid slots

  • −30%

    Time to sale, phase 2 target

03How we implemented it

Through MEDIR, as in every project. It's the same screen the client follows in the portal, phase by phase.

  1. M Map

    Completed · Aug 2026

  2. E Estimate

    Completed · Aug 2026

  3. D Decide

    Completed · Aug 2026

  4. I Implement

    Completed · Sep 2026

  5. R Review

    Ongoing · phase 2

  1. Map

    We read the accounts on every marketplace and the external team's process, portal by portal.

  2. Estimate

    What publishing by hand cost: €80k a year and a ceiling of about 100 listings a day.

  3. Decide

    A contract measured by coverage of listing volume, not by hours. One technical route for each marketplace.

  4. ImplementIn production

    The Ad Engine in production, connected to each marketplace through the route it allows.

    • Write APIs connected on the marketplaces that have themSep 2026
    • XML and CSV feeds for those that only fetch filesSep 2026
    • Reconciliation engine: 5,121 paused and 3,988 expired listings identifiedSep 2026
    • Dashboard in Portuguese and English with listings, paid slots and monthly wasteSep 2026
    • One-click approval of each listing by the RV Sales teamSep 2026
  5. ReviewCurrent phase

    Since 1 October, it has been the only way listings are managed. Phase 2 puts AI in charge of deciding where and how to list.

04The impact

In its first month, the Ad Engine already had 320 active listings on 7 connected platforms. Since October, all listing management goes through the engine.

With that in place, Indie reassigned its external publishing agents. Four people who spent the day publishing listings by hand moved to work of higher commercial value.

Reading the accounts also exposed an old bad habit: the team always created new listings instead of reactivating existing ones, which triggered publishing limits and threw away each listing's history. The engine fixes this by default.

The contract measures success by coverage of listing volume, not by development hours. With the AI layer, every motorhome goes to the platforms where it sells best, with the content that gets the most enquiries.

Next case

02 · MPL Imobiliária

From Excel to an estate agency ready to triple its portfolio

Read the case

480 h

More cases

Let's build something real.

No more pilot projects that never reach production. We start with the Operations Map: two weeks, fixed price, to find out where hours, money and clients are being lost. Then we build, integrate and stay alongside the team.