xMap AI

Jev decision

Jev decision asks TypeSafe’s Jev model one typed question about each place, using everything the blocks before it found: yes or no, pick one of your options, or a score on your own levels. Every answer is checked against the type you asked for and comes with how sure Jev is, so Filter, Branch and Sort can act on it straight away.

In the editor: Asks TypeSafe’s Jev model one typed question about each place: yes or no, pick one of your options, or a score on your levels. Answers come with how sure Jev is, ready for Filter and Branch.

Settings

Question type

Pick one

The shape of the answer you want. It decides which fields the block adds and the settings that follow.

Yes or no The chance that the answer is yes, from 0 to 100, and a plain Yes or No (Yes from 50 up).
Pick one One of your options, exactly as you wrote it, and how sure Jev is.
Score A score from 0 (your first level) to 100 (your last), the level’s name, and how sure Jev is.

Question

Text, several lines

Asked about each place, using everything the blocks before found about it.

Needed before the flow can run.

Options

Text, several lines

One per line, 2 to 50.

Needed before the flow can run.

Levels, lowest first

Text, several lines

One per line, 2 to 10. The score runs from 0 (the first level) to 100 (the last).

Needed before the flow can run.

Field it adds

Text

Yes or no: the chance of yes, 0 to 100, plus _answer (Yes or No). Pick one: the option, plus _confidence. Score: 0 to 100, plus _level and _confidence.

Needed before the flow can run.

What comes out

Each item that leaves the block carries these fields, to use in the blocks after it as {{field}}.

Question type: Yes or no

With Field it adds left as jev_yes.

jev_yes: The chance that the answer is yes, 0 to 100.
jev_yes_answer: Yes when the chance is 50 or more, else No.

Question type: Pick one

With Field it adds left as jev_pick.

jev_pick: One of your options, exactly as written.
jev_pick_confidence: How sure Jev is, 0 to 100.

Question type: Score

With Field it adds left as jev_score.

jev_score: The score, from 0 (first level) to 100 (last level). Filter reads a field ending in score as Fit score.
jev_score_level: The name of the level it chose.
jev_score_confidence: How sure Jev is, 0 to 100.

How to use it

Banking and payments

Only independent shops for the field reps

A payments company sells card terminals to independent shops; chain stores buy centrally and waste a rep’s visit.

Before: Reps found out at the door that a shop belonged to a chain.

With the flow: Only the shops Jev judges independent reach the CRM, each with how likely it is.

FMCG distribution

Size every new restaurant

A wholesaler routes large restaurants to its key-accounts team and the rest to van sales.

Pick one always answers with one of your three sizes, so Branch can split on it without surprises.

B2B sales

Score prospects on your own scale

Score uses your own levels, from Poor prospect to Great prospect, and the 0 to 100 score sorts them.

Healthcare

Does the clinic take walk-ins?

A diagnostics lab wants to partner with clinics that see walk-in patients.

Filter on the chance of yes, not just the answer, to keep only the places Jev is fairly sure about.

Real estate

Is the area right for a premium store?

A developer asks the same question about every plot the brokers send, using the shops around it.

Before: Each analyst judged the area in their own way.

With the flow: Every plot gets the same question and an answer with a percentage the team can compare.

Hospitality

Which hotels might stock our drinks?

The confidence field lets you send only the answers Jev is sure about and check the rest by hand.

Telecom

Possible dealers, ranked

A telecom operator wants to know which phone shops could become dealers.

Score with your own four levels turns a long list into a shortlist the regional manager can read.

Good to know

One question per place, up to 500 places a run, 8 at a time. Places past that go on without an answer, and the step says so.
Jev reads every field the place has, except pictures, IDs and coordinates, and long texts are cut. Add POI Intelligence or Map business scan before it to give it more to go on.
Answers are checked: a chance or a confidence is always 0 to 100, Pick one is always one of your options, Score is always on your levels. An answer of another kind is left out, and not charged.
Pick one takes 2 to 50 options; Score takes 2 to 10 levels, lowest first. Write them one per line.
Without TypeSafe connected on the server, the block gives sample answers, so check the step says it answered with Jev.
Name the field after the question (independent, size, prospect) so the fields that come out are easy to read in emails and sheets.
Credits: 15 for each place answered.

Works well with

Map business scan: Finds businesses on online maps live: name, category, rating, address, phone, website and whether it is open.
POI Intelligence: Adds details to each place found by Map business scan: categories (cuisine), price level, opening hours, phone, website, amenities and a short description, and if you want them its latest reviews.
Filter: Keeps only the items that match a rule.
Branch: Splits items in two.
Sort and limit: Puts the items in order, and can keep only the first ones: the 10 best rated, the 20 nearest, the highest fit scores.
xMap LLM: Tell the AI what to do in plain words.
Email: Sends one email per run with all the items, or one email per item.
HubSpot: Powerful CRM that offers tools for sales, customer service, and marketing automation.

Templates that use it

New openings to the field rep on WhatsApp: New restaurants and cafes graded by Jev and sent to your rep on WhatsApp every morning.
New outlets, graded and routed to each rep: New outlets graded, researched and routed to the nearest rep, with what to pitch at each.
New shops likely to want card payments: New shops likely to be cash only, given to the nearest sales agent every weekday.
A vetted list of subcontractors, every month: Contractors vetted from their ratings, likely to finish on time, every month.
Independent phone shops to recruit as dealers: Independent phone shops to recruit as dealers, sent to the nearest territory manager.
Small businesses matched to the right bundle: Small businesses matched to the right bundle, with a pitch, sent to the nearest agent.
New garages and tyre shops on my van routes: New garages and tyre shops sorted by what they do, scored and sent to the right van route.
Agri input dealers for my field officers: Agri input dealers sorted by what they sell, routed to the nearest field officer.
Young independent businesses for tax season: Young independent businesses graded as tax clients, in a sheet before tax season.
New exchange and money transfer offices, for compliance work: New exchange and money transfer offices that need compliance help.
New warehouses that need security systems: New warehouses graded as clients for access control and cameras.
Forest plots checked for clearing, every month: Forest plots checked for clearing from space, with a field check flag.
Busy restaurants for food waste collection: The busiest kitchens graded as food waste collection clients.
Shipyards for my coatings and paint sales: Shipyards graded as coatings buyers, with the purchasing manager.
Independent shops to sign up as marketplace sellers: Independent shops graded as marketplace sellers, with the owner.
Gift shops and home stores to stock my products: Independent gift and home stores graded as stockists, in a sheet every week.
Truck repair and machine shops for my spare parts reps: New workshops graded as spare parts buyers, to the reps.
Metal fabricators for welding supply sales: Fabricators graded as welding supply buyers, with the owner.
Garment factories for my fabric sales: Garment factories graded as fabric buyers, with the merchandiser.
Clinics without online booking, for my scheduling software: Clinics likely without online booking, with the practice manager.
Computer shops to recruit as hardware resellers: Computer shops graded as hardware resellers, with the owner.
New independent cafes and restaurants for my point-of-sale software: Newly opened independent cafes and restaurants, graded as point-of-sale buyers.
Independent gyms for my membership software: Independent gyms graded as software customers, to the sales team.
Aquarium shops for my fish supply wholesale: Aquarium shops graded as wholesale customers, with the owner.

And 16 more in the template gallery.

Stories with it

First through every new bakery's door (Food distribution)
A loan officer who sees the street before she lends (Small business lending)
The forty clinics that need a visit (Pharmaceuticals)
Closing the gaps in the care network (Health insurance)
The next charging stations, where drivers have none (EV charging)
Calling the companies about to move (Business telecom)

Help articles

Typed decisions with Jev: Yes or no, pick one, or a score, with confidence.

Try it

Build a flow with Jev decision on the home page, no account needed, or book a call. Try it live

Settings

Question type

Question

Options

Levels, lowest first

Field it adds

What comes out

Question type: Yes or no

Question type: Pick one

Question type: Score

How to use it

Only independent shops for the field reps

Size every new restaurant

Score prospects on your own scale

Does the clinic take walk-ins?

Is the area right for a premium store?

Which hotels might stock our drinks?

Possible dealers, ranked

Good to know

Works well with

Templates that use it

Stories with it

Help articles

Industries that use it

Try it