System 01 / inbound qualification

SageWork: Partnership Lead Scoring, Then and Now

A synthetic case study in AI-assisted lead scoring, built on a framework I designed by hand.

Ian Wilkinson
Input / employer enquiryDecision / route + priorityOutput / first response
SageWork is a fictional company. All data, enquiries and figures on this page are synthetic and for illustration only. No real client, employer or partner information is used.

01 / Situation

The problem

Imagine SageWork: a UK membership platform connecting experienced career changers and returners with employers, built around recruitment partnerships. A commercial team is already in place and closing deals, but every rep qualifies inbound interest differently. A single-vacancy small business gets the same treatment as a two hundred role employer. An agency representing a dozen clients gets read as one deal, not a channel. There is no shared way to tell the difference, or to route each enquiry to the right next step, before the time is spent.

02 / Manual baseline

Then, no AI available

What I actually built

When I joined a business in a similar position, a commercial team already existed, but a structured way to score, route and pitch partnerships did not. I built one: explicit criteria for vacancy volume, organisation scale, contact quality and sector fit, a clear view of which enquiries belonged in a self-serve flow versus a sales conversation versus a channel relationship with agencies, and enough CRM discipline, defined stages, required fields, that the pipeline data actually meant something.

Scoring a single enquiry by hand wasn't necessarily slow on its own, a rep could work through one in around ten minutes. The real cost was consistency at volume: without a shared framework, every enquiry tended to get roughly the same amount of attention regardless of its actual value, or under pressure, triage got skipped altogether and everything landed in the same pile. Either way, time went to weak leads that were never going to convert, while strong leads waited their turn instead of being prioritised.

03 / Live system

Now, AI assisted rebuild

Try the lead scoring tool

The same framework, rebuilt so every enquiry gets scored against the same criteria in seconds, not just the ones a rep has time to triage properly. Enter a synthetic employer enquiry below and see the routing decision, the reasoning behind it, and a draft first response.

04 / What changed

The delta

This is not AI replacing the judgement. The criteria, the routing logic, the sense of which relationship model fits which enquiry, all came from experience, not a model. It isn't really about speed: a rep could score one enquiry by hand quickly enough. It's consistency at volume: every enquiry gets the same rigour applied every time, so time and attention go where the value actually is, rather than spread evenly across leads that were never all worth the same.

05 / The point

What this demonstrates

This is what I mean when I say I build commercial systems: not a one-off deal, a repeatable, intelligent way of routing and prioritising demand that holds up under volume. The tools have changed. The judgement behind them has not.

Next scenario
System 02 / outbound prospecting

SageWork: AI Prospecting Agent, Then and Now

Fresh external data used to sit in the CRM until a rep had time to research and personalise outreach. Watch the same judgement, rebuilt as an agent that checks for duplicates, researches, qualifies, and drafts a full multi-channel sequence in seconds.

Open system