AccAnalysisAccAnalysis
Professional Services & Staffing

Prospect research that takes seconds instead of an afternoon

Sales teams were assembling prospect context by hand from job boards, directories and industry databases. We built a research portal and an agent that produce the same picture — and the first draft of the outreach.

At a glance
Status
Delivered
Region
United States
Engagement model
Dedicated team

Status: Delivered. Described from our own delivery records. Client identity, brand and commercial terms are withheld. Figures, where given, cover the period stated and nothing beyond it.

The organisation

Context

A staffing and recruitment business whose sales process begins with identifying organisations that are hiring, understanding what they are hiring for, and finding the person who owns that problem.

That research was being done manually, per prospect, by each representative.

The brief

The problem

  • Prospect information was fragmented across job boards, company directories and industry databases, and had to be assembled by hand every time.
  • Research consumed hours that were not spent in conversation.
  • Quality varied by representative, so the same prospect might be approached with very different levels of context.
  • Outreach written from thin context reads like outreach written from thin context.
The work

What we built

A prospect research portal

Surfacing open roles, company information, key people, hiring challenges, office locations and contact details, with industry-classification search and location-and-radius filtering.

A research agent

That takes an industry, a location and related parameters and returns relevant owners and decision-makers, verified contact data, stated interests and apparent pain points.

Draft personalised outreach

Generated from that context — as a starting point for a representative, not as an automatic send.

Method

How we delivered it

  1. 1

    Watch the manual process

    To establish what “good context” means to a representative who converts well.

  2. 2

    Build the aggregation layer

    Across the sources, with verification on contact data.

  3. 3

    Add search and filtering

    On industry classification and geography.

  4. 4

    Add the drafting layer

    With the representative always editing before sending.

  5. 5

    Pilot with a small group

    And compare the resulting conversations, not just the time saved.

Sequence

How it was phased

PhaseDurationWhat happens
1Process observation
1–2 wks

What context a good representative actually gathers

2Aggregation layer
4–5 wks

Sources, verification, company and people records

3Search & filtering
2–3 wks

Industry classification, location and radius

4Drafting layer
2–3 wks

Personalised outreach generation, editing workflow

5Pilot
2–3 wks

Small rep group, conversation quality reviewed

Indicative phasing for work of this shape. Actual duration varies with data quality, access and decision speed.

Hand-over

What the client keeps

  • The portal and agent source in their own repository
  • Source integrations and verification logic
  • Search configuration
  • The outreach templates
Stack
PythonLLM-based agentsIndustry classification dataWeb applicationEmail integration
Keep reading

Related

Related case studies

Tell us what your current system can't do, and we'll tell you what it would take to change that.