DYSARTISANIDEAS, FORGED INTO FORM.Search the episode library ↗

THE INNOVATION FORGE / Forgeworks Grimoire

Plan recruitment travel with a Quorum Insights Engine

Build organization-level context from volume and conversion, then use Quorum Scores and Personas to ask better recruitment travel questions.

The Datamancer card from the Codex Practitioners Tarot collection.
The Datamancer reads the context behind the map.

THE IDEA, IN A MINUTE

A shared map for the next conversation

The third Forgeworks Grimoire training moves from a person-based Slate export to organization-level calculations in Excel. Quorum Scores and Personas help the team examine travel habits, investigate overlooked relationships and keep the components behind a score visible.

01 / THE RECORDING

Watch the episode.
Or find your moment.

Ask what makes a school worth another conversation

Recruitment travel plans carry a lot of inherited knowledge. A counselor remembers a productive visit. A familiar route fits the budget. A school has appeared on the calendar for years. Those experiences matter, but a team also needs a way to notice places its existing routine has overlooked.

The Quorum Insights Engine gives that conversation a structure. The third Forgeworks Grimoire training gathers information around an organization, then uses scoring and personas to make patterns easier to discuss. The Datamancer's map is a useful metaphor: each school becomes a place to investigate, rather than an isolated list of student records.

In the recording · 04:24 ↗

Choose the group before choosing the calculation

Begin by deciding what one row represents. David starts with a person-based Slate query, exports student IDs, school codes and stage indicators, then aggregates the records into one row per organization in Excel. The broader idea can also apply to ZIP codes or geomarkets, but those groupings answer different questions.

Write down the population and period before exporting anything. Which students belong to the group? Which application cycle counts? How are multiple school records handled? A clean-looking spreadsheet can hide inconsistent definitions. Keep the underlying counts visible so a colleague can trace a result back to the question you intended to ask.

In the recording · 06:23 ↗

Read the components before the combined score

Volume, conversion and engagement offer different views of the same relationship. Volume shows scale, conversion describes movement between stages, and engagement adds observed behavior. The demonstration calculates engagement averages but leaves engagement out of the final score because too few test records contain it. That distinction matters when you reproduce the build.

Consider two fictional schools with the same overall score. One brings many inquiries and relatively few applications. The other brings a small number of inquiries who often apply. Those schools need different conversations. A single number should help the team find the cases, then get out of the way while people examine them.

Pay particular attention to denominators. Three applications from four inquiries and thirty applications from forty inquiries share a percentage but carry different amounts of evidence. An empty denominator is a missing or undefined rate, not proof of poor performance. Establish a minimum-count review and a clear way to display missing values before assigning labels.

In the recording · 18:39 ↗

Make personas explain a pattern

The Excel build assigns points using quintile thresholds and explores giving conversion measures extra weight before summing the components. Quintiles describe relative position within the comparison group, not an absolute standard of quality. Document the reference population, treatment of ties and missing values, and weights. Check that fixed spreadsheet ranges still cover the data after an update.

Personas add a shared vocabulary to the numbers. Their useful purpose is to distinguish patterns that deserve different questions: substantial volume with weak later conversion, modest volume with stronger conversion, or mixed signals that need more context. Keep the definition beside the label so staff can challenge the classification.

Use these as organization-level planning prompts. They do not describe an individual student's motivation, ability or eligibility. A student at a school with little recorded engagement still deserves an accessible route to information and support. A planning score should not quietly become an admissions or financial-aid decision rule.

In the recording · 21:11 ↗

Account for the attention you already invested

A nearby school or a school visited repeatedly has had a different opportunity to engage than a distant school your team has rarely reached. David explicitly raises distance and prior recruitment activity as context for an adjusted Quorum Score. That is an important question to carry into any comparison.

An adjustment still needs inspection. Show the original score alongside the adjusted result and explain what changed. Removing a statistical association does not prove that travel caused a result or that a different itinerary will improve it. Treat unexpected changes as a reason to review the records and ask counselors what the model cannot see.

Low activity may reflect limited outreach, incomplete tracking, a small population or a relationship that needs attention. Before moving a school off the travel calendar, ask what support has actually been available there. The episode's planning frame is strongest when underperformance starts an investigation rather than ending a relationship.

In the recording · 46:33 ↗

Leave the meeting with a testable next step

Bring three contrasting organizations to the planning conversation. For each one, show the component measures, the proposed persona, the limits of the data and a question for the counselor who knows the context. Then choose a modest next step: verify a record, discuss a workshop or reconsider a visit.

Record why the team chose that step and when it will revisit the decision. Keep a dated snapshot of the definitions and results so next cycle's comparison remains understandable. The useful outcome is a travel conversation that colleagues can explain and improve together, with the score serving as a starting point rather than a verdict.

In the recording · 55:25 ↗

03 / A SMALL EXPERIMENT

Read three fictional school patterns

For each fictional organization, separate the observed measures, the missing context and one useful question for the recruitment team.

The tiny high-conversion group

Fictional Birch School has three applications from four inquiries. Fictional Cedar School has thirty applications from forty inquiries. Both show 75% conversion. What should the planning view keep visible?

Compare your reasoning

Show the counts beside the rate. The same percentage rests on different amounts of evidence. Review a minimum-count policy and more context before proposing a change in travel.

Revisit this moment · 20:38 ↗
The mostly empty engagement column

Fictional Harbor School has volume and conversion data, but engagement scores exist for only a few records. Should blank values become zero and feed a complete-looking composite?

Compare your reasoning

Keep missing information distinct from observed zero. Inspect coverage and document any omitted component. The demonstration skips engagement quintiles because the test data is too sparse.

Revisit this moment · 24:02 ↗
The familiar nearby school

Fictional Meadow School is near campus and receives several visits. Fictional Ridge School is farther away and has received none. Meadow has the higher raw score. What should the team ask before repeating last year's route?

Compare your reasoning

Compare prior outreach and distance, inspect the components and consult relationship context. An adjusted score can support investigation, but it does not establish the effect of travel or justify withdrawing access to support.

Revisit this moment · 46:33 ↗
Build a test card to take with you

Notes stay in this tab. Reloading clears them.

04 / INSIDE THE STUDIO

The seams are part
of the story.

Behind the episode

David explains that earlier attempts using independent subqueries, reports and an export/import process took too long to produce the result he wanted, so this training does most calculations in Excel. The live build also exposes a useful limit: engagement scores are too sparse in the test data to include engagement quintiles in the total. The visible omission is part of the demonstration, not a completed three-component model.

Revisit the studio moment · 06:47 ↗

FG 03 Quorum Insights Engine (Song 03 of The Forgeworks Grimoire)

The separately published Episode Three outro carries the Datamancer's map-and-compass imagery into the song.

Watch the standalone lyric video

Art from the workshop

The Datamancer, the practitioner guiding this episode, brings organizational context to the recruitment travel conversation.

Open the artwork ↗

KEEP THE IDEA MOVING

Where would you take it next?

CONTINUE EXPLORING