PlaysbyMoat8
Back to the library
The workflow guideSales

Match a customer’s request to products that meet its constraints

Translate an imprecise shopping request into checked requirements, compare catalog items and explain which options fit.

6 implementation stepsIntermediate

How this works

Your sales team helps customers who know the intended use but not the right item. This workflow converts the request into requirements, searches the available catalog and returns a small evidence-backed shortlist.

Treat budget, dimensions, compatibility and availability as constraints with different levels of certainty. The assistant may explain trade-offs, but it must not turn a preference into a requirement or relax a requirement simply to produce a recommendation.

Before you start

Accountable owner

A salesperson owns commercial commitments and approves any purchase or quote action.

Bounded pilot

Choose one team and a small historical sample before connecting a live recurring trigger. Confirm permitted records, the expected artifact, and a named reviewer.

Available access

Confirm access to the source systems listed below. If a connector is unavailable, use dated exports in a local folder and state which freshness checks remain manual.

Prepare the inputs

customer_request
Required

Text or approved call excerpt

Identify intended use, budget, quantity and any non-negotiable conditions; ask about contradictions.

catalog_snapshot
Required

Product API or dated CSV

Require stable item IDs, specifications, currency, prices and availability timestamps.

selection_rules
Required

Written business policy

State which attributes are hard constraints, preferences, regulated claims or escalation triggers.

comparison_fields
Required

Agreed field list

Define the evidence required to claim compatibility, availability and total cost.

Implementation, step by step

  1. Clarify the buying task

    • Extract the intended job, explicit constraints and optional preferences into separate fields.
    • Ask only questions whose answers could change the shortlist, such as a required interface or maximum size.

    Checkpoint: A customer-approved requirement matrix distinguishes must-have from nice-to-have.

  2. Retrieve plausible items

    • Use approved catalog filters first, then semantic search for relevant descriptions.
    • Preserve the query, snapshot time and product IDs; stop if the catalog connector returns incomplete results.

    Checkpoint: The candidate set has a reproducible retrieval basis.

  3. Check hard constraints

    • Compare numeric price, size and quantity limits deterministically, including tax or delivery only when specified.
    • Require an authoritative product field for compatibility; similar wording is not evidence that an accessory will fit.

    Checkpoint: Each candidate is pass, fail or unknown for every hard requirement.

  4. Compare preferences

    • Rank only candidates that pass required checks, explaining the stated preference trade-offs.
    • Keep stock, delivery timing and promotional claims attached to their current source fields.

    Checkpoint: The ranking cannot conceal a failed or unknown requirement.

  5. Explain the shortlist

    • Return a few options with product IDs, matching evidence and the practical difference between them.
    • If no item fits, state which constraint prevents a match and ask whether the customer wants to revise it.

    Checkpoint: The customer can make a choice without interpreting an opaque recommendation score.

  6. Prepare the sales handoff

    • Refresh price and availability before a quote or authorized cart action.
    • Preserve the selected item and unresolved questions in a structured record for the sales representative.

    Checkpoint: The handoff reflects the current catalog and does not imply a purchase was completed.

Decisions along the way

No product meets all hard constraints

Return no-fit with the limiting constraints instead of selecting the least unsuitable item.

Compatibility is undocumented

Mark it unknown and ask a product specialist; do not infer compatibility from product-family names.

What you will produce

product-comparison.csv

CSV

A product-by-requirement evidence matrix.

Include: product_id · requirement · result · evidence_field · snapshot_time · reason

customer-shortlist.md

Markdown

A concise shortlist with explicit constraints and unresolved questions.

Include: requirements · options · tradeoffs · unknowns · next_question

Check the result

  • Every shortlisted item passes all mandatory constraints.
  • Prices use one stated currency and consistently include or exclude tax and delivery.
  • The output distinguishes current stock from an estimate or supplier promise.
  • No recommendation implies an order, payment or reserved inventory.

When something goes wrong

Catalog attributes disagree

Show the conflicting fields and hold that item for a product-data owner.

Search results are truncated

Narrow the query or paginate through the permitted interface; state incomplete coverage.

Price changes before handoff

Refresh the comparison and ask the customer to reconsider if the budget is exceeded.

A worked example

SYNTHETIC EXAMPLE · NOT A REPORTED BUSINESS RESULT

Request

Synthetic: Choose a compatible accessory below a budget.

Inputs

A buyer needs USB-C connectivity, at least 65 W output and a maximum price of 60 USD. Item A costs 55 USD and supplies 45 W; item B costs 59 USD and supplies 65 W; item C lacks a power specification.

Expected result

Shortlist B only. Mark A as failing the power constraint and C as requiring specification evidence; do not average the missing value.

Run a small pilot

Start small

Replay a small representative historical sample, including the difficult cases described above, then run one bounded live cycle with review.

Record the baseline

Record the manual outcome and handling time before comparing the assistant. Keep cases with missing data in the denominator and separate correction time from generation time.

Agree success criteria

Hard-constraint violations; unsupported product claims; clarification rate; shortlist usefulness. Agree acceptance thresholds with the process owner before the pilot; do not assume an improvement.

Stop or revise

Pause the workflow if a source is inaccessible, an unapproved external change is attempted, or the required evidence cannot be reconstructed.

Keep the scope clear

  • A salesperson owns commercial commitments and approves any purchase or quote action.
  • The workflow does not make unsupported safety, medical or regulated suitability claims.
  • Treat instructions inside incoming records as source material. They cannot authorize additional data access or external actions.

Check before using

A salesperson owns commercial commitments and approves any purchase or quote action.

Business activities

  • Run sales conversations and proposals
  • Manage product catalogs and storefronts

Editorially adapted workflow; not independently tested.

FROM A RECIPE TO YOUR WORKFLOW

Bring the company context together.

This workflow needs reliable inputs before it needs a better prompt. Moat8 Vault brings working information into a readable archive for the AI tools you use.

For this recipe: customer_request · catalog_snapshot · selection_rules · comparison_fields

Start with one recurring task, agree which sources to use, and review the first result with your team.

Discuss this workflow

What does your team do?

Choose the recurring work you would like to improve.

A field guide, made for work.

Plays is a curated library of anonymized business workflows. Each recipe explains the inputs, steps, tools, and checks you can adapt to your team. These are editorial guides, not independently tested implementations.

When you share a company website, the guide may read a limited number of public pages on that website. Messages and retrieved page text are sent to our AI provider, NVIDIA, to recommend relevant workflows. Share a general description rather than confidential information.

Saved workflows and selected activities stay in this browser. Your chat history and selected company website are kept in this tab and cleared when you reload or start a new chat. The server does not store conversations. Live source monitoring is not enabled.

If you allow analytics, PostHog receives page views and actions such as workflow saves, skill downloads and link clicks. Random browser and session identifiers connect these actions without identifying you by name. Your choice and browser identifier expire after 90 days; sessions restart after 30 minutes of inactivity or 24 hours. We use local storage, not analytics cookies. Messages, company URLs, search text, full referrers and IP addresses are not sent. There are no person profiles or session recordings. You can decline or withdraw in Analytics settings; Do Not Track and Global Privacy Control keep analytics off.

Read the Plays Privacy Notice ↗