Accountability
When a number does not move, see whose job it was.
A normal CRM tells you a rep missed their number. MADDOX shows why, and whose job it was to fix. “We coached it and nothing changed” is really one of three different situations: the coaching never happened, which is the manager’s; it happened and the rep did not act on it, which is the rep’s; or the rep acted on it and the number still did not move, which points at the coaching content itself. Each finding goes to the right place.
How it works
Three failures, three different owners
Each arm points somewhere specific, which is what makes it accountability rather than a report.
Never coached
The item was identified and no coaching happened. This belongs to the manager, and it shows on the manager’s record, not the rep’s.
Coached, not implemented
The rep was properly coached and did not act on it. This one belongs to the rep, and only this one.
Implemented, still flat
The rep did the thing and the number still did not move. That points at what is being taught, not at either person, and it routes to the coaching content.
In the product
Each manager’s own record, on five measures.
At a glance
- Cadence breaches
- 2
- Unverdicted commitments
- 4
- Sessions without transcript
- 1
Sample data — illustrative product UI, not a performance claim.
The mechanism
What managers are measured on, and the rules that keep it fair
Measuring whether managers coach is the part that changes behavior, and it is also the part most likely to go wrong. Three design decisions do the work.
What a manager is actually measured on
Five axes carry named occurrences, and one further reading — whether the team’s identified issues are actually moving — is a team-level fact rather than an event, so it deliberately has no occurrence rows behind it.
- Cadence — sessions owed and held
- Transcript — sessions run without one
- Verdict recording — commitments they authored and never answered
- Coaching target — whether what they set out to coach was coached
- Report compliance — their reports’ own overdue obligations, so an unmet duty surfaces where somebody can chase it
A missed 1 on 1 is recorded on the day, not inferred later
Every night, MADDOX records each weekly 1 on 1 that was owed and not held once its deadline passes. The record keeps the cadence and deadline that applied at the time, so changing your cadence policy next quarter cannot rewrite whether last quarter was missed.
If a late session eventually happens, the record is marked resolved rather than deleted, because “held, three weeks late” and “held” are different facts, and only one of them explains a number.
Three failures, three owners, and one of them is not a person
When a number has not moved, that is not one finding. The coaching never happened, which belongs to the manager. Or it happened and was not implemented, which belongs to the rep and only to the rep. Or it was implemented and the number still did not move — which points at what is being taught, and routes to the coaching content itself rather than at either person. Separating them is what makes it accountability rather than a report, because each arm has somewhere specific to go.
One accountability score per manager, from nine inputs
The Owner sees one score per manager: a weighted average of nine inputs. It is calculated from snapshots taken at the time, so changing a weight never rewrites last quarter. The Owner sets the weights with sliders, and a weight of zero removes that input rather than counting it as a zero. An input without enough data is left out, and a manager with nothing measurable yet reads as unmeasured, with the reason, never as zero.
Click through from the score and you land on the manager’s whole record: each input, the occurrences behind it, and the findings recorded against them.
- Cadence kept: 1 on 1s held against what was owed
- 1 on 1 grade: how their 1 on 1s scored against the coaching rubric
- Coaching grade: the craft of their coaching sessions
- Prep rate: 1 on 1s that were actually prepared for
- Coaching depth: how far coached items really got
- Promises kept: the dated promises they made, and kept
- Obligation clearance: how fast their team’s overdue work is cleared
- Team movement: whether their reps are improving
- Escalated share: the one input where lower is better, weighted least
A weekly digest band, ordered by what demonstrably did not happen
Every week the Owner’s digest carries a band about the managers: 1 on 1s owed and not held, coaching promises past their date and 1 on 1s that fell short of the contract come first, because they record work that did not happen. Coaching graded below the rubric follows, because that work happened and fell short. Last comes each team’s own open findings, as counts that name nobody. Ranking a bookkeeping gap level with a missed session would teach an Owner to skim the band instead of reading it.
Leave and waivers are recorded, not inferred
Where a person’s leave is on record before the nightly run, the period is suspended outright: no obligation is created for the time they are away, so no accusation exists to undo. Leave recorded after a breach has opened is different, because a record already says a named person missed something. MADDOX does not clear it on its own; it lists it for the Owner as waivable because of leave, and the Owner decides.
A waiver always names the Owner who gave it and requires a reason. It is not a resolution: “held three weeks late” and “agreed not to apply” stay different facts, because reporting a 1 on 1 that never happened as one that happened late would be a false statement about a named person.
A rep never sees their manager’s failure, and a named list has a ceiling
The finding that a manager did not coach belongs to the manager and appears on their record. Upward visibility of it is deliberately not built: it would change the finding from accountability into a grievance. What a manager sees about their own reports is counts by category and a pointer to the gated view — one level down, never the whole subtree, and never the underlying item titles of people below them.
Where a team carries more overdue obligations than a person can act on, a handful are named and the remainder is disclosed as a count. Thirty named accusations under one heading is a wall nobody reads; showing the first few silently would be worse, because it reads as the whole list.
Who it is for
For leaders who mean it
Any organization can install a coaching tool. Measuring whether managers actually coach is the part that changes behavior.
- Owners and VPs who want coaching to be a real expectation
- Managers who would rather be measured on the right things
- Reps who have been coached badly and blamed for the result
Related
Questions
The things people actually ask.
Does a rep see that their manager failed to coach them?
No. That finding belongs to the manager and appears on the manager’s record. Upward visibility of a manager failure is deliberately not built: it would change the finding from accountability into a grievance.
What does a manager see about their own reports?
Counts by category and a pointer to their own scorecard, not the underlying item titles of people below them. One level down, never the whole subtree. Where a team has more overdue obligations than a person can act on, a handful are named and the rest are disclosed as a count — thirty named accusations under one heading is a wall nobody reads, and silently showing only the first few would be worse.
Is a missed 1 on 1 automatically a breach?
It is recorded as an expectation breach against the coach, with the specific week it was missed, because the cadence obligation sits on the person who owed the session. Leave on record suspends the obligation in advance, and only the Owner can waive a breach after the fact, with a stated reason that stays on the record.
Are managers graded on their team results?
Partly, but their coaching craft is graded separately on what they actually did in the room, and none of those items is an outcome. Folding an outcome into a craft rubric grades a manager twice on the same fact and lets a lucky quarter cover a badly run meeting.
Does an escalation mean someone is in trouble?
No. Every escalation is labeled as raised by the system and not yet reviewed by a human, every time it is shown. It is a prompt for a conversation, not a verdict.
Find out whether the coaching is happening.
Request access and look at your managers’ cadence record. Most organizations assume the coaching is happening. The record is usually the surprise.