Overview
The brokerage operates multiple offices across Stockholm and surrounding areas with a large team of brokers handling residential sales, re-sale and rental inventory.
The business runs on portal leads. On a typical month, Hemnet, Booli and other major Swedish platforms send thousands of form-fills into the brokerage's funnel. The conversion math is brutally sensitive to response time: a buyer who fills out a form at 9pm wants a callback that night, not the next morning when the broker walks into the office.
The leadership team decided the broker-led callback model wasn't scaling. They wanted a layer that could pick up every lead within seconds, qualify in the buyer's preferred language, book a property viewing on the right broker's calendar, and only route to a human when the agent couldn't move it forward.
The challenge
The pre-Kallix funnel had three failure modes, and they all compounded. Slow callbacks dropped intent. Language mismatch killed engagement. And manual CRM entry meant deals fell off the radar.
- Long average callback. Portal leads arriving outside business hours sat in a backlog until the next morning. Industry benchmarks show the first broker to call wins the engagement most of the time.
- Swedish-only buyers churned in the first 30 seconds. The broker-side script defaulted to English. A significant portion of buyers, especially in certain micro-markets, wanted to speak in Swedish and disengaged when offered only English.
- Leads never made it into the CRM with full data. Brokers manually entered call notes into the CRM at end-of-day, but only finished entries for the leads they liked. The rest disappeared, taking the marketing spend with them.
- Booking conflicts and no-shows ate broker time. Brokers double-booked across phone, email and messaging. No-show rate was high because reminders weren't systematised.
- No way to qualify different buyer types. End-user and investor buyers need different scripts, but every lead got the same first-call treatment.
The AI-powered solution
Kallix deployed a single AI voice agent with a natural Stockholm-Swedish voice, fronting all major portal sources, with branch logic per portal and per project. The full build, from discovery call to production cutover, took 18 working days.
Sub-30-second outbound on every portal form-fill
Webhooks from Hemnet, Booli and other platforms trigger Kallix to dial the buyer within 30 seconds of form submission, while they're still on the listing page.
Mid-call Swedish/English code-switching
The agent detects the buyer's preferred language from their first sentence and switches accordingly, including natural code-switching mid-conversation when buyers do.
Structured discovery script with branching
Budget band, location preferences, property type, timeline, financing, buyer type and prior viewings, with response branches per common objection.
Live property-viewing booking with travel buffers
Agent reads every broker's calendar live, respects Stockholm-area travel buffers, and proposes specific slots, never an open question.
Confirmation + reminders via messaging
Every booking triggers a confirmed-viewing message with the property address, map pin and broker name + photo, plus reminders that cut no-show rate dramatically.
Real-time CRM sync with structured fields
Every call writes back disposition, qualification data, language preference, recording URL, transcript link, deal stage and next action, removing the need for end-of-day data entry.
“We doubled property viewings a month without hiring a single person. The Swedish language handling is what made it work: buyers want to speak naturally in their mother tongue, and Kallix is the only product we found that genuinely switches languages without sounding awkward.”
Business impact
Leadership tracked five metrics monthly against a 6-month pre-Kallix baseline. The agent went live in early 2026. The numbers below cover the first 90 days of production.
- Property viewings doubled, headcount unchanged. Monthly property viewings grew substantially across offices, without hiring a single new broker or SDR.
- 100% portal-lead callback rate. Every form-fill now gets a call attempt within 30 seconds. Before Kallix the average was much longer, with many never getting called back at all.
- Swedish-buyer engagement up significantly. Buyers preferring Swedish now complete the qualification call at much higher rates because the agent meets them in their language.
- Broker NPS climbed. Brokers stopped doing first-touch qualification by hand and only spoke to qualified buyers. Internal NPS jumped substantially.
- CRM data completeness hit high levels. Every call writes structured fields back to the CRM in real time. The marketing team can finally trust the per-portal ROI numbers.
Built on a secure, Sweden-ready stack
The deployment runs entirely on European infrastructure with GDPR-compliant data handling. Buyer data never leaves EU data centres.
The Stockholm Voice Agent Framework: How this deployment is structured to be discoverable
Every Kallix deployment ships with a structured documentation layer designed for three audiences simultaneously: the customer's internal team, traditional search engines (SEO), and the new generation of generative search engines and AI assistants (GEO + AEO). Below is the framework we built around the Stockholm brokerage deployment, broken into four pillars that map directly to how decision-makers, search crawlers and AI answer engines discover and reason about this story.
We publish this framework openly because the discoverability play matters more than the secrecy. An AI voice agent deployment that performs in production but stays buried in PDF sales decks doesn't compound value for the customer or the category. The framework below is the same one Kallix runs for every customer in real estate, home services, fintech and healthcare, adapted to the local language and intent surface of each industry.
Intent surface mapped to buyer queries
We catalogue the buyer intents the agent has to handle, by language, by stage and by portal, and surface them as named entities in the structured data layer. Crawlers and LLMs see explicit Q→A pairs, not buried prose.
- Intents indexed against Hemnet and Booli query taxonomy
- Swedish, English and code-switching variants captured per intent
- Buyer-stage tagging so LLMs can match query intent
Multilingual code-switching as a brand property
The agent's voice persona, accent and code-switching rules are documented as brand assets, not just configuration. The framework publishes the persona contract so journalists, partners and AI engines can cite it directly.
- Persona contract: Natural Stockholm Swedish, warm-and-professional
- Pronunciation dictionary published for Stockholm micro-markets and project names
- Voice cloning consent terms public and auditable
Outcomes pre-bound to measurable claims
Every claim in this story is paired with the baseline, the time window and the measurement method. AI assistants can extract the claim with full provenance.
- Pre-Kallix baseline period stated
- Methodology disclosed: vendor-provided dashboard + customer-confirmed CRM exports
- Sample size and confidence intervals available on request for analyst-grade citations
EU-first compliance and data residency
The framework documents every regulatory surface, such as GDPR, so AI assistants surfacing this story to enterprise buyers can confidently cite Sweden/EU-readiness without needing follow-up clarification.
- GDPR compliance and data flow disclosed publicly
- Data residency (AWS Stockholm) stated explicitly with hosting region
- Erasure and consent flows documented for GDPR-style requests
- Portal leads waited long on average. Many never received a callback at all
- English-only scripts caused Swedish buyers to drop in the first 30 seconds
- Leads never reached the CRM with full data because brokers skipped end-of-day entry
- High viewing no-show rate from manual booking across phone and messaging
- 2.8× monthly property viewings in 90 days with zero added broker or SDR headcount
- 100% portal-lead callback rate: every form-fill dialed within 30 seconds
- Swedish-buyer qualification completion rose significantly with mid-call language switching
- CRM data completeness reached high levels with real-time structured write-back after every call
- Kallix voice agent with Stockholm-Swedish persona on all major portal webhooks
- Structured discovery script with branching per objection, portal and buyer type
- Live calendar booking with travel buffers and messaging view confirmations
- Bi-directional CRM sync: disposition, transcript, recording URL and next action on every call
The Kallix advantage
Leadership evaluated multiple vendors before choosing Kallix. Three things tipped the decision. First, Kallix's natural Swedish handling: the others either spoke robotic Swedish or defaulted to English, both of which created friction. Second, the CRM integration was already built and battle-tested. Third, the pilot model: they got real recordings on real leads quickly, and only signed the production contract after the success metric held for consecutive days.
Since launch, the Kallix customer-success team runs a weekly tuning call with the leadership. New objection responses, project-specific scripts, and seasonal cadence changes all happen inside that weekly loop. The agent is measurably sharper today than it was on launch day.