9 Learners, 4 Weeks, One Dashboard: Inside Elqo's Human-Skills Development Pilot

We ran a four-week pilot (18 May – 15 June 2026) to test one thing an L&D leader actually cares about: can you give people personalised development, see them measurably improve, and run the whole thing without it becoming a project? Sixteen learners enrolled. Nine completed the full four weeks with weekly data and an exit interview.

The results were consistent. Across those nine, the average performance score rose from 66.7 to 76.6 out of 100 — a gain of about 10 points, or roughly 15%. Every single completer improved. Not one went backwards. They did it on a few minutes a day, logging 512 practice sessions between them, and most reported hitting the personal goal they set at the start within the four weeks.

Metric	Result
Average score gain	+10.0 (of 100)
Learners who improved	9 / 9
Time to measurable results	4 weeks
Manager-to-program ratio	1 manager, 9 personalised programs

Three Things Made It Work

Three findings stood out, and they map directly to an L&D buying decision.

1. It was personalised, and people hit their goals

Every learner got a program built from their own intake form, aimed at their own goal. No two were the same. By week four most rated themselves as having largely or fully hit the goal they set, and they could point to the moment it showed up in real life: a won competition, a presentation that landed, a meeting they led, a conversation they no longer avoided.

2. It was fast and measurable

Improvement showed in the data inside four weeks, on five to seven minutes of practice a day. Each learner has a weekly, multi-metric score and a documented before-and-after. This is the evidence L&D usually struggles to produce: not attendance, but progression you can see.

3. It was easy to run

One person managed nine fully personalised programs at once. Curricula were built per learner, custom content was assigned, progress was tracked per person and across the cohort, and the weekly reports were generated by the platform rather than assembled by hand. The admin load that normally makes personalised development impossible at scale was not there.

What the Pilot Developed

The capability area in this pilot was professional communication — the human skill that sits underneath presenting, leading meetings, handling clients, and speaking up. But the story here is not about public speaking technique. It is about measurable improvement, growing confidence, and learners progressing against a goal they set themselves. The platform turns a soft, hard-to-measure skill into a tracked development arc, the same way a technical skill would be tracked.

That matters because human-skill development is the gap most L&D teams feel: technical training is measurable and human-skill training usually is not. This pilot measured it.

How the Pilot Was Run

This was a structured program, not a casual trial. Each learner committed to a defined protocol agreed up front:

• Three short practice sessions a day, roughly five to seven minutes total.
• A short weekly check-in survey.
• A 10-minute review call at the midpoint, and a 20-minute reflection at the end.

Every learner began with an intake form capturing their role, where the skill shows up in their work, what they most wanted to improve, and their own definition of success. That intake fed a custom development program built for them before kickoff — the same onboarding flow an L&D manager would run for a cohort.

Elqo scores every practice session and rolls it into an overall performance score out of 100, with component diagnostics beneath it (fluency, clarity, accuracy, pace, and verbal-filler frequency). Weekly surveys added a self-rated confidence score and progress against the learner's goal. The midpoint and exit reviews captured the story behind the numbers. Sixteen learners enrolled, spanning students, early-career professionals, working professionals, and founders across nine countries; nine completed the full four-week protocol and form the evidence base for this report. The rest are covered honestly in the retention section below.

Built for the Manager

Personalised development usually dies on operations. Building a different program for every person, keeping content current, and tracking who is improving is normally more work than any L&D team can carry. The reason this pilot worked is that the dashboard removed that load.

What the manager does	What it looked like in this pilot
Set up a cohort and onboard each learner from an intake form	Nine learners onboarded from a short intake; each profile captured role, context, and goal.
Build a custom curriculum per learner	Nine distinct programs built, each aimed at that person's goal. No two identical.
Upload and assign your own content	Custom prompts and scenarios assigned into individual programs and adjusted mid-pilot based on weekly feedback.
Track progress per learner and across the cohort	Scores, trends, and component diagnostics visible per person and across the group, from one place.
Generate progress reports automatically	The weekly reports behind this analysis were produced by the platform, not built by hand.

The headline for an L&D buyer: one person ran nine fully personalised development journeys at once and still produced a clean weekly audit trail for every learner. The same workflow scales to a cohort without adding headcount.

Headline Results: Every Learner Progressed

The table below shows each learner's performance score in week one and week four, the change, and their session count. The cohort moved from a 66.7 average to 76.6, with no exceptions to the upward trend.

Learner	Wk 1	Wk 4	Change	Sessions
Bec	72.6	80.2	+7.6	96
Odette	69.4	78.2	+8.8	21
Yegana	67.5	81.1	+13.6	89
Petrichor	63.9	80.4	+16.5	9
Eminent	69.0	75.5	+6.5	102
Joshua	66.9	76.3	+9.4	43
Ewa	70.2	78.8	+8.6	80
Amy	65.7	76.5	+10.8	67
Promil	54.7	62.2	+7.5	5
Cohort average	66.7	76.6	+9.9	512

The biggest absolute gains came from those starting lower or practising most: Yegana (+13.6), Amy (+10.8), and Joshua (+9.4). Petrichor posted the largest percentage gain (+26%) on only nine sessions — a reminder that the model rewards quality of practice as well as volume. Bec logged 96 sessions and moved her early average from the high-50s into the mid-80s, one of the largest jumps in the cohort.

They Hit the Goals They Set

Improvement on a score is one thing. Hitting the goal you came in with is what makes a development program feel like it worked. At week four, learners rated their own progress against the personal goal written in their intake form.

Learner	Goal set at intake	Self-rated at week 4
Odette	Speak concisely so the team doesn't ask her to repeat	80–90% there
Joshua	Improve storytelling and voice range	75–80% there
Promil	Stop losing his train of thought; fewer fillers	~75% there
Petrichor	Start a sentence with clarity	“To a large extent” (won a debate final)
Eminent	Speak more confidently, converse more	Goal met
Amy	Hold a conversation without anxiety	Goal met
Yegana	More energy and confidence; fewer fillers	Goal met; confidence 8/10
Bec	Reduce fillers, feel confident and concise	“Definitely” meaningful progress

Confidence moved alongside the scores. Where learners gave a confidence rating versus their starting point, the typical answer was 7 or 8 out of 10. Promil went from a self-rated 3 to 7.

The Underlying Diagnostics Moved Too

The performance score is a composite. The component diagnostics show where the work happened, with the steepest gains for learners who started with clarity or second-language challenges.

Learner	Fluency Δ	Clarity Δ	Accuracy Δ	Fillers / session
Bec	+4.2	+15.8	+17.5	3.8 → 3.6
Odette	+11.0	+19.1	+18.6	2.6 → 2.5
Yegana	+8.6	+27.9	+31.6	2.3 → 1.4
Petrichor	+19.7	+15.7	+9.9	2.5 → 2.6
Eminent	-3.0	+14.4	+16.6	5.4 → 4.3
Joshua	+8.2	+19.6	+20.1	5.2 → 4.6
Ewa	+1.1	+1.4	+0.3	1.2 → 1.8
Amy	+6.9	+19.4	+21.3	7.5 → 5.3
Promil	+7.1	+7.7	+4.9	0.7 → 2.2

What Changed: The Patterns

Across nine exit interviews, four midpoint reviews, and weekly surveys, the same findings surfaced repeatedly. These are the outcomes an L&D buyer should care about.

Confidence was the headline outcome

Asked what changed most, learners named confidence more than any single technique. Confidence is what turns a development program from a checkbox into a behaviour people keep up on their own.

“I really do start conversations with people now. Before, I avoided them because I felt I was going to be judged. Now I can just start a conversation.”

— Eminent, university student

Results transferred to real situations, fast

Every completer pointed to a concrete moment where the development showed up in their actual work or study inside four weeks. Transfer to the real task is the proof point L&D needs.

• Petrichor won her inter-faculty competition final.
• Promil handled an impromptu spot in front of 50+ peers without freezing.
• Yegana led a live activity and team introduction at work.
• Bec, Odette, Amy, and Ewa reported clearer client consults, project recaps, supervisor and team meetings.

Personalisation was the differentiator

Learners repeatedly singled out the custom-built program as the standout — the partner-dashboard workflow being validated, and what users perceived as making Elqo different from generic, one-size courses.

“You made customized programs according to the weakness that the person felt. It was more like we are having an exclusive session for ourselves.”

— Joshua, participant

Short daily practice built a habit that stuck

The bite-size format lowered the barrier enough that people actually showed up daily. Several built a consistency habit they had previously failed to keep.

“Elqo is like Duolingo for development. If you're a busy professional, it's great to keep you practicing every day since you only need five minutes.”

— Odette, participant

Results by Learner Type

Grouping the cohort by use case maps the evidence onto the segments an L&D leader buys for. Each has its own full case study:

• Client & customer-facing roles (Bec, Ewa, Amy) — group average 69.5 → 78.5. The cohort L&D usually has to justify with revenue, with measurable gains in exactly the contexts where performance affects trust and sales.
• Team leaders & high-potential talent (Odette, Yegana) — group average 68.5 → 79.7. Both led real meetings during the pilot with tighter, more confident, more structured delivery.
• Early-career & emerging talent (Joshua, Petrichor, Eminent, Promil) — group average 63.6 → 73.6. The largest percentage gains and the most dramatic real-world results in the pilot.

Retention, Drop-Off, and What We'd Change

An honest report covers the people who did not finish. Of 16 enrolled, nine completed fully, three received bespoke programs but did not finish all four weeks, and four dropped after intake or week one. The drop-offs cluster around competing time pressure (several sat exams or had heavy work periods mid-pilot) and the friction of self-directed enrolment without a manager or program owner holding the cohort accountable.

That is a feature of the deployment model, not a verdict on the product — and it points straight at the L&D opportunity. The most engaged learners — Eminent (102 sessions), Bec (96), Yegana (89), Ewa (80), Amy (67) — are exactly the profile a sponsored, manager-run cohort produces, where the program is scheduled and tracked rather than left to individual motivation. Even the lightest completers improved: Promil moved +7.5 on five sessions; Petrichor posted the largest percentage gain on nine.

Product feedback the pilot surfaced

Learners were asked directly for criticism. Sharing it openly shows a live roadmap. The recurring asks:

• Feedback should say how to fix a flagged issue, not just name it.
• The in-app AI coach should feel like a constant companion, not a side feature.
• Longer practice formats for real presentations (the current maximum is three minutes).
• More content variety over a month, and an optional human practice partner.
• Minor reliability fixes flagged by a few users (session logout, an occasional submission bug).

None of these blocked results. They are the punch list of a product being used hard by real people, and most are already in progress.

Demand Signal: Learners Wanted to Keep Going and Pay

All nine completers said they would recommend Elqo, and several had already referred a friend, partner, or family member during the pilot. Offered a discounted continuation, learners accepted on the call — including one who immediately enrolled her partner. The program earned word of mouth and willingness to pay inside four weeks, from a standing start.

“I told my best friend to use it. I would definitely recommend.”

— Promil, participant

The Bottom Line for L&D

This pilot shows the full loop L&D needs for human-skill development: set a learner's goal, build them a personalised program, measure them weekly against objective metrics, and produce a documented before-and-after the learner recognises and values. It did that for nine people at once, on five-to-seven-minute daily sessions, with one manager and a clean audit trail for every learner.

The result was measurable improvement, fast goal attainment, rising confidence, strong satisfaction, and willingness to pay. The same engine that ran nine individuals is the cohort workflow built for organisations. The proof that it develops real people is in this report. The next step is running it inside a team, where sponsorship and accountability lift engagement even further.