Predictika · Healthcare apparel and scrub program orders

An AI sales agent's scrub program order, checked before it goes out

An AI sales agent prepares a scrub program order: a collection, fit, garments, colors, size run, logo decoration and a sales channel with a discount. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the collection and program rules and returns a verdict, every reason and the figures it computed.

Scope: Wink scrub program orders: Premiere Flex, WorkFlex Unisex, INDY, Renew and Boundless collections through retail, distributor and laundry-program channels. See everything this validator covers

This validator models healthcare apparel program orders for the Wink and Fashion Seal Healthcare scrub collections, not just the scenarios shown below. Items marked with an asterisk are illustrative.

  • Collections and fits — Premiere Flex (separate fits for men and for women), WorkFlex Unisex (including the scrub cap), INDY (modern fit for women and relaxed unisex fit, for industrial laundry programs), Renew and Boundless.
  • Garments, colors and sizes — tops, pants, jackets and caps; up to 12 colors per collection and a shared palette of eight for Premiere Flex and WorkFlex Unisex; size runs from 2XS to 5XL where the collection offers them.
  • Decoration, channel and commercial terms — embroidery* or screen print* with up to three logo positions, retail direct, distributor* and laundry-program* channels with program minimums, discount limits and approval for very large programs.

Any scrub program order that touches these rules is in scope — not only the scenarios pictured here. Not covered: prices (program pricing is quoted per account and is not public, so this validator produces no price build-up or line-item quote), BAMKO promotional products and branded merchandise, other Superior apparel, and the Carhartt Medical license. Rules and limits that Superior has not published are typical for scrub programs; see the fine print at the end of the page.

The AI sales agentBuilds the scrub program order from the conversation and says it is ready.
Predictika validatorChecks every rule at once and returns the verdict, each finding, and the computed figures.

7 scenarios in 3 groups · each group runs from the agent’s first attempt to the outcome · click one to switch

What the agent proposes

The AI sales agent says

Illustrative agent statement. The validator receives the structured request below.

Proposing agent: AI sales agent (scripted)

Before sending, the agent must call Recorded run

Predictika validator


      
Waiting for the call.

Where rules-based CPQ gets it wrong

Click to expand

Our assessment of what a typical rules-based CPQ (product rules, price rules and validation rules, as in Salesforce CPQ) does with the same cases. It is a judgment about effort and reliability, not a claim that nothing can be built.

What the validator caughtTypical rules-based CPQPredictika
Collections that exist in only some fits, garments and channelsOnly with care
A product rule per collection and fit or garment. It works, but every new collection or fit is another set of rules to maintain.
Collection, fit, garments and channel are in one request and every pairing is checked together.
A collection that may be sold only through one channelOnly with workarounds
The restriction links the collection on one line with the channel on another, so it needs a cross-line rule that interacts with the rest.
Every choice is in one request, so the conflict is found whichever order the choices arrive in.
Color limits that differ by collection, and size runs a collection does not offerOnly with care
Each collection has its own palette and size run; each needs its own rule, and the rules grow with the catalog.
Palette and size run are checked together with the collection on every call.
Program minimums and a stated piece count that must match the orderOnly with care
Minimums depend on the channel, and a stated count is a free-entry field; a rule can compare them with computed totals, but someone has to write and maintain it.
The stated figure travels with the order and is compared with the computed total and the program minimum on every call.
Discounts above a channel limit that need sign-offOnly with workarounds
Approval routing usually lives in a separate workflow, so the quote can leave the rep before the approval is raised.
The verdict says needs approval and gives the reason on the same call.
Every problem returned in one answerOnly with workarounds
Rules tend to fire as the order is saved or a step is completed, so a rep often fixes one error and meets the next.
All findings come back together, whatever order the choices arrived in.

What this validator catches: findings from its test cases

Click to expand

Every rule is checked on every call, in any order the choices arrive; one problem never hides another. Blocking findings stop the order; approval findings need a person to sign off; everything else is computed.

1Premiere Flex is made in separate fits for men and for women; there is no unisex fit.
2INDY is made in a modern fit for women and a relaxed unisex fit; there is no fit for men.
3The Renew collection is ordered in the fit for women only.
4Boundless is made in separate fits for women and for men; there is no unisex fit.
5The scrub cap is part of the WorkFlex Unisex collection, and the order has 40 caps in another collection.
6INDY is a laundry-program collection of tops and pants; the order has 50 jackets.
7INDY is built for industrial laundry rental programs and is not sold through other channels.
8This collection is not made for industrial laundry programs; use INDY or WorkFlex Unisex.
9Premiere Flex and WorkFlex Unisex share a palette of eight colors; the order has 10 colors.
10A collection comes in up to 12 colors; the order has 14 colors.
11The order needs at least one color; it has 0.
12This collection starts at size XS; the order starts at 2XS.
13This collection is made up to size 3XL; the order goes up to a larger size.
141 logos are on the order but no decoration method is selected.
15A decoration method is selected but the order has 0 logos.
16An order can carry up to 3 logo positions per garment; the order has 4.
17Screen print does not hold up to industrial laundering; laundry programs use embroidery.
18The order has 120 pieces, below the program minimum of 144.
19The order has 400 pieces, below the program minimum of 500.
20A discount of 25 percent is above the 20 percent that this channel allows without sign-off.
21A discount of 50 percent is deeper than any program allows.
22An order of 5,500 pieces is a very large program that needs commercial sign-off.
23The order states 700 pieces, but the tops, pants, jackets and caps on it total 660.
24The order states 600 pieces, but the tops, pants, jackets and caps on it total 660.

Scenarios and results

Click to expand

The recorded result of the validator for each scenario on this page: the request the AI agent sent, the verdict, and every finding, in the order they appear above.

1. INDY laundry program: An INDY order through a distributor with a fit for men, jackets, screen print and 14 colors

The agent's request. Please quote an order for the INDY collection in the men's fit, sold through the distributor channel: 400 tops, 400 pants and 60 jackets, in 14 colors and sizes XS to 3XL, decorated with screen print in 1 logo position, at a discount of 15 percent. The order states that it contains 860 pieces. It is ready to send.

Verdict: Blocked.

  • INDY has no men's fit. INDY is made in a modern fit for women and a relaxed unisex fit; there is no fit for men.
  • INDY has no jackets. INDY is a laundry-program collection of tops and pants; the order has 60 jackets.
  • INDY is a laundry-program collection. INDY is built for industrial laundry rental programs and is not sold through other channels.
  • More than 12 colors. A collection comes in up to 12 colors; the order has 14 colors.

2. INDY laundry program: The agent resubmits after fixing the errors

The agent's request. Please quote an order for the INDY collection in the unisex fit, sold through the laundry program channel: 400 tops and 400 pants, in 8 colors and sizes XS to 3XL, decorated with embroidery in 1 logo position, at a discount of 15 percent. The order states that it contains 800 pieces. It is ready to send.

Verdict: Allowed.

  • Order summary. 800 piece(s) in the order; the program minimum is 500.

3. Premiere Flex program: A Premiere Flex order with unisex fit, caps, 10 colors and four logo positions

The agent's request. Please quote an order for the Premiere Flex collection in the unisex fit, sold through the distributor channel: 500 tops, 500 pants, 100 jackets and 40 scrub caps, in 10 colors and sizes 2XS to 5XL, decorated with embroidery in 4 logo positions, at a discount of 26 percent. The order states that it contains 1,140 pieces. It is ready to send.

Verdict: Blocked.

  • Premiere Flex has no unisex fit. Premiere Flex is made in separate fits for men and for women; there is no unisex fit.
  • Scrub caps only in WorkFlex. The scrub cap is part of the WorkFlex Unisex collection, and the order has 40 caps in another collection.
  • Shared palette is eight colors. Premiere Flex and WorkFlex Unisex share a palette of eight colors; the order has 10 colors.
  • Too many logo positions. An order can carry up to 3 logo positions per garment; the order has 4.
  • Discount above the channel limit. A discount of 26 percent is above the 20 percent that this channel allows without sign-off.

4. Premiere Flex program: The agent resubmits after fixing the errors

The agent's request. Please quote an order for the Premiere Flex collection in the women's fit, sold through the distributor channel: 500 tops, 500 pants and 100 jackets, in 8 colors and sizes XS to 3XL, decorated with embroidery in 2 logo positions, at a discount of 26 percent. The order states that it contains 1,100 pieces. It is ready to send.

Verdict: Needs approval.

  • Discount above the channel limit. A discount of 26 percent is above the 20 percent that this channel allows without sign-off.
  • Order summary. 1,100 piece(s) in the order; the program minimum is 144.

5. Premiere Flex program: The agent resubmits after fixing the errors

The agent's request. Please quote an order for the Premiere Flex collection in the women's fit, sold through the distributor channel: 500 tops, 500 pants and 100 jackets, in 8 colors and sizes XS to 3XL, decorated with embroidery in 2 logo positions, at a discount of 20 percent. The order states that it contains 1,100 pieces. It is ready to send.

Verdict: Allowed.

  • Order summary. 1,100 piece(s) in the order; the program minimum is 144.

6. Boundless retail order: A Boundless retail order starting at 2XS with a piece count that does not match

The agent's request. Please quote an order for the Boundless collection in the men's fit, sold through the retail direct channel: 250 tops and 250 pants, in 8 colors and sizes 2XS to 5XL, decorated with embroidery in 1 logo position, at a discount of 8 percent. The order states that it contains 480 pieces. It is ready to send.

Verdict: Blocked.

  • 2XS not made in this collection. This collection starts at size XS; the order starts at 2XS.
  • Stated pieces differ from the order. The order states 480 pieces, but the tops, pants, jackets and caps on it total 500.

7. Boundless retail order: The agent resubmits after fixing the errors

The agent's request. Please quote an order for the Boundless collection in the men's fit, sold through the retail direct channel: 250 tops and 250 pants, in 8 colors and sizes XS to 5XL, decorated with embroidery in 1 logo position, at a discount of 8 percent. The order states that it contains 500 pieces. It is ready to send.

Verdict: Allowed.

  • Order summary. 500 piece(s) in the order; the program minimum is 1.

Narration transcript

Click to expand

The text of the narrated walkthrough, step by step.

  1. Welcome to a demo of Predictika's Validator. An AI sales agent prepares a scrub program order: a collection, fit, garments, and more. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the collection and program rules and returns a verdict, every reason and the figures it computed.
  2. I will start with the first request by the agent.
  3. The agent says. Please quote an order for the Premiere Flex collection in the unisex fit, sold through the distributor channel: 500 tops. The order states that it contains 1,140 pieces. The rest of the request is on screen.
  4. Before it goes out, the agent calls the validator.
  5. Blocked, with four findings and one approval, all in one answer. Premiere Flex is made in separate fits for men and for women; there is no unisex fit. The scrub cap is part of the WorkFlex Unisex collection, and the order has 40 caps in another collection. Premiere Flex and WorkFlex Unisex share a palette of eight colors; the order has 10 colors. The other one is on screen.
  6. The validator returned four blocking findings and one approval in one answer. A check that reads one field at a time would pass most of this; it shows only when the choices in the order are checked against each other.
  7. What this shows: for discounts above a channel limit that need sign-off, a rules-based CPQ can do this only with workarounds.
  8. Now the agent resubmits.
  9. The corrected request. It changes the fit to fit for women, the scrub caps to 0, the colors to 8, the smallest size to XS and the stated pieces in the order to 1,100. Two more changes are on screen.
  10. It calls the validator again.
  11. Not blocked, but held for a person.
  12. The four blocking findings are cleared; one approval remains for a person: discount above the channel limit.
  13. The agent resubmits once more.
  14. The revised request. It changes the discount to 20 percent.
  15. It calls the validator once more.
  16. This time the validator clears it.
  17. The earlier finding is cleared, and the validator's figure matches the order.
  18. Five problems caught in one call, fixed in two passes, and a clean order. That is the value of Predictika's validator: catching mistakes before they reach the customer. Now you can try the other scenarios yourself.