# A Shared End-of-Session Check Makes Every Teaching Format Comparable When teams teach the same topic in different formats, one shared end-of-session check makes their results comparable without forcing anyone to change their content. Joel Lord, who leads education at MongoDB, describes the mechanism: every session ends in the same 10-question skill check, and passing earns the same badge. "No matter where you learn, you will earn that same skill badge" (00:08:40). The check is "what makes that 10-hour workshop or that 10-minute flash badge session that we run comparable" (00:10:05). Formats include workshops from 90 minutes to a full day, webinars, short and long videos, and first-party and third-party events (00:07:58). Each skill covers one technical capability with 45 to 90 minutes of content and a hands-on lab. The check sits below certification level and is "meant to be relatively easy, but not a freebie" (00:10:05). Lord got other teams to adopt it with what Lord calls a "friendship tour" (00:15:06): sit with each team, compare content, find equivalences, and agree on learning objectives instead of imposing content (00:16:29). Teams keep their content as long as it covers the objectives. Lord joined education about 18 months before the talk, and says it didn't happen overnight (00:27:15). Badge-bearing sessions now send data back automatically (00:17:55). The reporting change: a trip report that said "3,000 badge scans" drew "so what," and now says "100 people who demonstrated skills in this very specific topic" (00:20:03). Where this breaks. "Comparable" means same instrument, not same learning. A check built to be easy measures exposure plus basic recall, so "demonstrated skills" overreads it. The figures are self-reported with no baseline or denominator: more than 500,000 badges last year (Lord says the exact number can't be shared, 00:08:40), 80 to 90% conversion in person, and about 55% online (00:14:24). The old QR-code rate isn't given, and Lord says the audience may be biased. The Atlas Search account story (00:20:47) has no comparison account. The scheme depends on agreeing objectives first, and the talk doesn't say what happens when a team's content skips one. Caption garble: the lab platform is rendered "Instruct Labs" and is probably Instruqt. <!-- atomicity: 1 idea ✓ --> ## Why This Matters When docs, workshops, and talks cover the same topic, I'd propose one shared check per topic, such as a short task or quiz tied to written learning objectives, before building any new dashboard. The first step is the agreement on objectives, which is a conversation with each team, not a tool. I'd report the check's pass rate beside what it doesn't test, so a number never reads as proof of skill. ## Links - [[3. Resources/Permanent Notes/A Contribution Metric Should Let Each Team Define What Counts]] — contrasts: that note lets each subteam define the metric, this one fixes the unit and lets each team vary the content; both are answers to cross-team comparability. - [[3. Resources/Permanent Notes/Adoption and Satisfaction Are Not Evidence That a Teaching Method Works]] — constrains: an easy check shows exposure and recall, so a pass rate doesn't establish that the method taught the skill. - [[3. Resources/Permanent Notes/Education Is the Only Marketing That Works With Developers]] — applies-in: education that ends in a measurable result gives a trust-based channel an outcome number. - [[3. Resources/Permanent Notes/A Docs Metric Earns Its Place by Naming Which Job It Does]] — extends: "100 people demonstrated skills" is a proving-side number that needs a stated method to survive scrutiny. - [[3. Resources/MOCs/DevRel MOC]] — applies-in: education measurement and cross-team alignment. - [[3. Resources/MOCs/Events/DevRelCon NYC 2026 MOC]] — applies-in: measurement cluster, education signals. ## Source - [[3. Resources/Readwise/Articles/Turning Education Signals Into Account Intelligence — Joel Lord|Turning Education Signals into Account Intelligence — Joel Lord]] - https://www.youtube.com/watch?v=YBxE5Lccd1c