Surus can close an auction in seconds, but getting paid can take weeks
Surus runs one of the world’s largest B2B auction platforms for industrial and corporate assets, called Escrapalia.
More than 110,000 auctions have run through the platform to date.
Companies auction off machinery, real estate, vehicles and surplus stock they no longer need.
Once a bid is won, Surus has to:
- Verify the award.
- Send the invoice.
- Track the stock.
- Notify the buyer.
- Monitor whether payment arrives on time.
- …And manage any incidents along the way.
All of this currently runs on manual work.
Staff call buyers, email them, and follow up with each one individually until the payment is in.
This becomes a threefold issue for Surus:
- Communication with winning bidders depends on someone picking up the phone or sending an email.
- Monitoring payment status and deadlines also depends on a person remembering to check, which raises the risk of a missed reminder or a payment slipping through unnoticed.
- Surus’s internal team spends a large share of its time on this repetitive follow-up, time it could spend on higher-value work instead.
Surus estimates this manual process pushes its operational costs 10 to 15% above where they should be.
A call centre that never sleeps and never forgets
Mozart Fintech is a Uruguayan startup that built a conversational AI agent for payment collection.
By way of illustration, instead of a Surus staff member picking up the phone, Mozart’s virtual agents call a customer – who won an auction on Escrapalia three weeks ago – and that still hasn’t paid.
How complete is the agent?
- It already knows the buyer’s past interactions, so the second or third call about the same debt doesn’t start from scratch.
- It can offer a payment plan or a debt restructuring option on the spot, negotiating rather than just reminding.
- Once the buyer commits to paying, the agent checks the bank directly to confirm the funds have landed, rather than waiting for someone at Surus to notice.
The expected effect:
- More buyers pick up the phone,
- More of them commit to paying,
- More of those payments are recovered.
As a result, Surus improves its cash flow while freeing its internal team from manual chasing.
Mozart will also hand Surus a live dashboard that will track:
- Contactability,
- Recovery rate,
- Amount recovered,
- How long successful calls take,
- How many calls it takes to get one buyer to commit.
How the EU-LAC Digital Accelerator turned an idea into a structured pilot
- Service n°1: PoC design & Prototyping
The accelerator helps define the scope of the proof of concept:
– Hypothesis: e.g. buyers are as likely, or more likely, to answer and engage with an AI agent as with a human caller.
– KPIs: e.g. 70% of customers promise to pay within three attempts.
– What it actually requires to run: e.g. Data access, Compliance, Phoning service.
The accelerator will then review preliminary results, evaluate the final results against the agreed KPIs, and support the conversation about scaling or contracting further work.
- Service n°2: Stakeholder mapping
The accelerator helps identify and prioritise the organisations that matter most for Surus’s move into AI-driven collections in Spain:
– Potential partners in the fintech and payments ecosystem,
– Compliance advisors familiar with automated debtor contact,
– Existing large clients, such as its IBEX 35 accounts and insolvency administrators, who could serve as reference cases.
Later, the partnership can turn that map into a prioritised outreach action plan:
– who to approach first,
– through which channel,
– with what message,
– and on what timeline.
One pilot, two different wins
| 🔨 SURUS (Corporate) |
📞MOZART |
|
|
