The reminder nobody wants to send
In a toy store, the sale rarely ends when the customer says "yes." It ends when they pay. And between the "yes" and the payment there's a gap that quietly eats your margin: the layaway tricycle a parent promised to pay off before the birthday, the wholesale order for the daycare that was "due Friday," the seasonal toy sold on installments.
You know it, and you avoid it. Sending the reminder feels awkward. It sounds like collections, and collections scare off the customer you want back in December. So you don't write, the balance ages, and a conversation that could have taken ten seconds on WhatsApp turns into a tense phone call three weeks later.
At Catalizadora we build AI agents that do exactly that awkward work, on time and in your tone. Without you ever touching the phone.
What a friendly collections agent does
The agent lives in your business WhatsApp and knows three things: who owes, how much, and since when. With that, it writes the right message at the right moment, in your brand's voice.
- The day a toy goes on layaway, it confirms the balance and the deadline. The customer knows what they committed to.
- Three days before the due date, it nudges warmly: "Hi Ana, Mateo's tricycle is set aside for you. There's $60 left to complete the layaway, you have until Friday."
- On the due date, it offers to settle right there: it sends a payment link straight into the chat.
- If the date passes, it doesn't threaten. It asks whether they want a few more days or prefer to recover the deposit, and only flags you when a human is actually needed.
Every interaction lands in your CRM. You stop guessing who owes you and since when: you just see a list.
Friendly isn't weak
The difference between collections that work and collections that drive people away isn't the amount, it's the tone and the timing. A well-built agent never sounds robotic or aggressive. It uses the child's name, remembers the specific toy, and always offers a way out. That turns a reminder into a favor.
And unlike you, the agent doesn't get tired, doesn't forget, and doesn't put it off because "it's awkward." It sends on day 3 and day 7 exactly as well.
The numbers behind a silent gap
Take a small toy store with 40 layaways a month at an average ticket of $80. If 15% go unpaid because nobody followed up, that's six lost layaways a month: around $480 monthly, more than $5,700 a year in merchandise you already reserved and failed to sell.
| Without follow-up | With an AI agent |
|---|---|
| Reminders when you remember | Reminder at the exact hour, always |
| Tone depends on your mood | Same friendly tone every time |
| Balances in your head or a notebook | Everything in the CRM, visible |
| Collecting is your job | The agent collects, you sell |
You don't need to recover 100%. Recovering half of that gap already pays for the system many times over.
How we build it at Catalizadora
We follow the MAGIA methodology: Mapping, Architecture, Generation, Implementation, and Autonomy.
- Mapping: we learn how you do layaways, on what terms, what happens when someone doesn't pay, and your real tone with customers.
- Architecture: we define the triggers (layaway day, before due, on due, after) and where the payment link connects.
- Generation: we write the agent in your voice, with the messages and the rules.
- Implementation: we connect it to your WhatsApp and your CRM, and test it on real cases.
- Autonomy: we hand it over working, with code, data, and infrastructure 100% yours.
What's included and what it costs
Our entry package, MAGIA Solo, costs $4,500 USD and ships in 15 days. It includes the AI agent on WhatsApp, the reminder and collections flows, the payment links, and the CRM where every conversation lands.
We don't charge retainers or locked-in licenses. Monthly operation is pass-through: you pay hosting and tokens directly, roughly $200 to $400 USD a month, with no markup from us. If your toy store has several locations or a large catalog, MAGIA Core ($15,000) integrates more of the operation.
The key point: the system is yours. We don't lock you in.
Common mistakes the agent fixes
There are three typical ways a toy store loses money on collections, and a well-built agent fixes all three.
The first is uneven follow-up. You chase balances hard the first week of December and let them die in January, when sales drop and you need the cash flow most. The agent has no seasons: the January balance gets the same punctual follow-up as the December one.
The second is losing the context. The customer put a toy on layaway three weeks ago, you no longer remember whether it was the tricycle or the dollhouse, and you write something generic that feels cold. The agent remembers the exact product, the amount, and the date, and that makes the customer feel you're talking to them, not to a list.
The third is offering no way out. When a customer can't pay on time and only gets pressure, they hide, and you lose both the sale and the relationship. The agent always proposes an alternative: a few more days, a partial payment, or recovering the deposit with dignity. The customer comes back because you treated them well even when they couldn't pay.
What happens with the hard cases
The agent doesn't try to solve everything alone. When a conversation goes off-script (a complaint, a big negotiation, an upset customer), it hands it to you with the full context: who they are, what they owe, and what was said. You step in only where a person is truly needed, and the rest, the routine reminders that drained your time and energy, runs on its own.
Start with the balance you already have
Before chasing more customers, collect what you already sold. A friendly collections agent is the fastest way to close a gap you've been ignoring, without becoming the bad guy.
Message the AI agent on WhatsApp to see it running with a case from your store, or book a direct call: cal.com/pablo-estrada-hlqaql. In that conversation we'll tell you what you'd recover and how soon it would be live.