Every broker’s question answered in under a minute

The challenge

How many people live near this corner? How many cafes are on this block? Each question waits a day in the queue, and brokers walk into pitches without the answer. His three analysts spend their weeks on the same lookups.

How the flow works

With xMap Flows, Owen built Sage. Sage is an AI employee: a flow of 5 blocks.

1. When a broker asks in the research channel,

   A question in research (Slack event)

2. Sage reads the question,

   The research desk (AI agent)

3. counts who lives around the address,

   Who lives there (Demographics)

4. lists the shops and cafes nearby,

   What is nearby (Nearby places)

5. and answers in the thread within a minute.

   Answer in the thread (Reply)

What it solves

• A day per question: solved by AI agent and Reply
• Pitches without answers: solved by Demographics and Nearby places
• Analysts on lookups: solved by AI agent

The result

Before a pitch in Liberty Village, a broker asks how many young adults live within a kilometre. Sage answers in forty seconds: eighteen thousand in their twenties and thirties, and forty-two cafes. The analysts are back on paid studies.

Owen’s brokers stopped waiting for answers. Now they walk in with them.

An illustrative scenario: the figures are modelled to show what the flow produces, not a customer's results.

The flow, block by block

1. 1

   A question in research

   Slack event

   Starts the flow when something happens in Slack: a new record, a message, a change.

2. 2

   The research desk

   AI agent

   An AI agent: for each incoming message it thinks, uses the blocks you link to its Tools port (data, xMap AI, lookups, your tables, other flows, apps), remembers the conversation and writes an answer in {{reply}}.

3. 3

   Who lives there

   Demographics

   Who lives in an area: the number of residents, men and women, and nine age groups, from xMap’s population grid (hexagons of about 0.1 km², about 100 countries).

4. 4

   What is nearby

   Nearby places

   Places around a point from xMap’s own places database (300M+ places, refreshed every month): pick the types of place (more than 5,000, such as Restaurant, Cafe, Pharmacy, Hospital or Gas station, or a whole group of them) and how far to look.

5. 5

   Answer in the thread

   Reply

   Answers whoever started the flow, on the same channel: the app or system that called its Webhook / API gets the answer back, and a message from Slack, WhatsApp, Telegram, Teams, SMS, Outlook or Discord gets a reply in the same chat, through the account of the block that received it.

Templates to build it

Ready-made flows with the same blocks. Open one, change the city and the kind of place, and run it.

• Site advisor chat for the expansion teamA chat where your expansion team names any address and gets people, competitors and a verdict.
• Neighbourhood guide chat for buyersA chat that tells buyers about any neighbourhood, like a local agent.
• Field sales helper chatA chat that tells field reps which shops are near them today.
• Meeting prep chat for repsA chat for reps: name the business you are meeting and get a short brief with questions to ask.
• Research assistant over live location dataA research chat over live people, business and hiring data.
• Chat about places near anyoneA chat for your customers that finds places near anyone, live.

Industries

Learn how it works

• How to build a WhatsApp assistant that answers about placesCustomers ask on WhatsApp; an AI agent searches and replies.
• Location intelligenceUsing where things are, and what is around them, to make business decisions.
• Catchment areaThe area a store or service draws its customers from.
• Ask the map a questionHow many of a kind of place, how many people for each, and where the gaps are, answered on the spot.
• Count what is around each of your placesCompetitors, footfall magnets or services within a distance of each site.
• AI agents with tools and memoryAn agent that answers questions using your blocks as tools.

More stories from Consulting

• Meeting every new business in its first weekLauren runs a one-woman bookkeeping practice in Houston. Her clients are independent shops, cafes and studios.
• Twelve AI employees, one page for the partnersBram is a partner at a large management consultancy in Amsterdam. His team runs twelve AI employees that watch markets for clients.
• The client hears about every rival gym firstWing is a consultant at a global strategy firm in Hong Kong. Her client, a European gym chain, is choosing districts for its first ten clubs.
• The supplier filed. Her clients knew by nine.Lea is an independent restructuring adviser in Berlin. She helps family manufacturers when a supplier gets into trouble.

All 137 stories →