Build a Sales Qualification Assistant for Inbound Leads with Prismfy

The focus is practical lead research from live public web data, so sales and growth teams can build account context, prospect profiles, and briefing notes without claiming private-data enrichment.

Problem framing

Inbound leads are easiest to qualify when you combine what the lead told you with what the company already publishes publicly. A company domain, product category, geography, and website copy often tell you enough to decide whether the lead is a fit, a maybe, or a no.

The wrong way to do this is to pretend you have hidden data. The right way is to build a small assistant that checks public pages, recent posts, and visible company signals, then returns a routing recommendation.

Prismfy fits because it gives the assistant a live public-web search step. That lets you qualify from evidence instead of guesswork.

Why this matters now

Inbound volume is useful only if the team can react quickly. If every form fill requires manual research, the response time slips. If every lead gets the same generic follow-up, the conversion quality drops.

An assistant built on public-web evidence can help with both problems. It can flag obvious fit, spot a mismatch, and surface the one or two public facts that justify the next step.

Step-by-step solution

1. Start with the lead submission. Capture the company name, website, and what the lead asked for.
2. Search the company domain for product pages, pricing, docs, or about pages.
3. Search the broader web for recent public signals such as launches, news, hiring, or events.
4. Turn the results into a simple qualification object: fit, urgency, route, and evidence.
5. Hand off to sales, CS, or support based on the public evidence and the inbound request.

The assistant should help the rep decide, not replace the rep’s judgment.

Code example

This TypeScript example takes an inbound lead domain and builds a compact qualification result from public-web searches only.

type PrismfyResult = {
  title: string
  url: string
  content?: string
  engine?: string
}

type PrismfyResponse = {
  results: PrismfyResult[]
  cached?: boolean
}

async function prismfySearch(query: string, domain?: string, timeRange = "month") {
  const body: Record<string, unknown> = {
    query,
    page: 1,
    timeRange,
  }
  if (domain) body.domain = domain

  const response = await fetch("https://api.prismfy.io/v1/search", {
    method: "POST",
    headers: {
      Authorization: `Bearer ${process.env.PRISMFY_API_KEY}`,
      "Content-Type": "application/json",
    },
    body: JSON.stringify(body),
  })

  if (!response.ok) {
    throw new Error(`Prismfy search failed: ${response.status}`)
  }

  return (await response.json()) as PrismfyResponse
}

function qualifyInbound(lead: { company: string; domain: string; request: string }) {
    return Promise.all([
        prismfySearch(`${lead.company} product pricing docs`, lead.domain),
        prismfySearch(`${lead.company} recent news hiring launch`, undefined, "month"),
    ]).then(([official, publicWeb]) => {
        const officialTop = official.results[0]
        const publicTop = publicWeb.results[0]

    const evidence = [
      officialTop && { title: officialTop.title, url: officialTop.url, snippet: officialTop.content?.slice(0, 160) },
      publicTop && { title: publicTop.title, url: publicTop.url, snippet: publicTop.content?.slice(0, 160) },
    ].filter(Boolean)

        return {
            fit: evidence.length > 0 ? "likely" : "unclear",
            urgency: publicTop ? "current public activity found" : "no clear recent signal",
            route: lead.request.toLowerCase().includes("demo") ? "sales" : "triage",
            evidence,
      note: "Public-web qualification only. No hidden enrichment or contact discovery used.",
    }
  })
}

You can use the same output to route to a rep, request more information, or ask a follow-up question. The important part is that the decision is visible and explainable.

Practical notes and caveats

Keep the qualification scope narrow. Public-web signals can tell you whether the account looks relevant, but they should not be stretched into contact discovery or private intent claims.

A strong qualification object usually includes:

• the inbound request
• the company domain
• one or two public evidence URLs
• a simple route decision
• a short note explaining the decision

If the search results are weak, the assistant should say that plainly and route to manual review. That is a better outcome than pretending the lead is qualified.

Use domain when the company owns the answer. Use timeRange when the inbound request depends on recent public activity. Do not make the assistant search everything every time.

Why Prismfy fits this workflow

Prismfy fits because inbound qualification is a live public evidence problem. You need a quick POST /v1/search call, a small result set, and a way to turn those results into a routing decision the team can trust.

FAQ

What kind of lead research fits Prismfy best?

Prismfy fits public-web account research: company pages, product pages, recent announcements, public documentation, and other open sources that help a team build better account context before outreach.

Does this workflow promise private-data enrichment?

No. The safe positioning is public-web prospect research, briefing, and account context. It helps SDR and growth workflows without claiming access to private contact or CRM data.

Related Prismfy guides

• Build a Prospect List Sourcing Workflow for SDR Outreach with Prismfy
• Build a Pre-Demo Account Briefing Template with Prismfy
• Add Real-Time Web Search to an OpenAI Agent Workflow with Prismfy

Try Prismfy

Create a Prismfy key, test POST /v1/search, and wire the search step into the workflow you care about first.