Why consulting firms need a different 360° – and what it looks like

Why a standard 360° falls short in a consulting firm

Most 360° instruments are built for line organisations: one leader, one boss, one stable team. Leadership in a consulting, law or audit firm looks different. A manager or senior manager runs rotating project teams, is steered by rotating partners, faces clients every day – and is judged simultaneously by the analysts below and the partners above. Five features make the difference:

• From expert to leader. People are promoted for technical excellence; leadership arrives as a by-product, and many managers keep solving problems themselves instead of delegating – see Delegation as leverage in professional services.
• Client, partner and utilisation pressure at once. Under that pressure leadership behaviour shifts: more control, less listening, pressure passed down the line. See Leadership under pressure.
• Up-or-out. When every promotion round is also a selection decision, feedback is quickly read as appraisal. A developmental 360° therefore has to be separated from the promotion process – organisationally and technically.
• Very small, rotating teams. A manager often has only two or three direct reports per engagement. An anonymity rule designed for ten people fails here.
• Rated from above and below. Partners judge client leadership and quality of output; analysts experience delegation, coaching and protection from overload – both matter, but not with the same questions.

What a 360° has to deliver in professional services

These conditions translate into concrete requirements for questionnaire, rater groups and anonymity rules. The table shows how LEADBeyond 360° meets them – as a product developed for boutique consulting, not adapted after the fact.

The dimensions that count on the partner track

Of the twelve progressive leadership dimensions, four are especially visible on the partner track: Client & Stakeholder Leadership, Empowering & Delegation, Coaching & Development and Results & Execution. Add seven limiting tendencies, some of which are almost structural in expert cultures: perfectionist control, micromanagement, the overload identity ("nothing works without me") and intellectual arrogance. Whether one of them is at work in a particular leader, the report presents not as a verdict but as a pattern in the self–other comparison.

The added value over a behaviour-only 360° lies in the chain behind the behaviour: what does the manager believe about leadership (Belief), what do they trust themselves to do (Capability), what throws them off under pressure (Trigger), what does their environment experience (Behavior), and what effect follows (Outcome)? A senior manager who believes in delegation, feels able to do it and still builds every slide personally does not have a knowledge problem – the explanation sits one layer down. More in Knowing and doing: the capability–behaviour gap and the methodology.

Anonymity when everyone knows everyone

In a sixty-partner firm everyone knows everyone. A promise that "the feedback is anonymous" is not enough there – anonymity has to come from the system. In LEADBeyond 360°, individual answers are never displayed: groups below the threshold are counted into "all others", free text appears in random order and without a group label until the group reaches the threshold, and the leader sees who was nominated and who has submitted – never who answered what.

The second question is rater numbers. Research on the reliability of 360° ratings points one way: reliability rises with the number of raters, and two or three people per group is thin (Hensel et al. 2010; Nowack & Mashihi 2012). In consulting that is the normal case. The answer is not to lower the threshold but to nominate more broadly: peers from several projects, direct reports from the last engagements, several partners as supervisors where they exist. How many raters does a 360° need? and Is 360° feedback anonymous? show how to plan it.

A programme design for a manager cohort

1. Set-up (week 1): The objective is development ahead of the next promotion round – explicitly not the promotion decision. Rater groups: two partners or engagement leads as supervisors, at least three peers from different projects, at least three direct reports from recent engagements. The anonymity threshold stays at two.
2. Communication and nomination (week 2): The managing partner explains purpose and confidentiality rules in person; HR sets up the leaders; each leader nominates from the prepared pool and the LEADBeyond consultant advises on nominations so the groups are balanced.
3. Survey (weeks 3–5): Invitations go out in one batch, reminders run automatically. A window spanning a project boundary stops every response landing in the same delivery week.
4. Aggregation and review (week 6): A report is generated once the self-assessment is in, at least two rater groups reach the threshold and the deadline has passed (or every invited rater has submitted). The LEADBeyond consultant reviews every report, adds annotations and releases it.
5. Debrief (weeks 7–8): One conversation per leader, focused on two or three hypotheses – say, the gap between self-image and others' view on delegation, or the stress pattern under client pressure – see What a 360° report contains.
6. Second cycle (after twelve months): Same rater groups, same questionnaire; the report shows change per dimension. The first report stays unchanged as the reference.

A report on its own changes little. The meta-analysis by Smither, London and Reilly (2005) found only small average improvements after multisource feedback – improvement was more likely when recipients accepted the need to change, set goals and took action, for instance with a coach. That is why the debrief is part of the LEADBeyond 360° product, not an optional extra.

What LEADBeyond 360° does not do for a partnership

• Not an instrument for partner election, promotion or compensation. The assessment is built for development; feeding results into appraisal decisions costs you the raters' candour – see Confidentiality and data protection.
• Not self-service for a mass roll-out across all grades. For annual feedback to every consultant without a conversation, a survey tool is cheaper – see Consultant-led or self-service?.
• Not an industry benchmark. Percentiles refer to the LEADBeyond reference norm – initially an expert norm that becomes calibrated and empirical as data accumulates, always labelled with its source. There is no peer cohort of "all consulting firms", and we do not claim one.
• No AI analysis of open comments. Stop, start and continue answers are shown as written – in random order and threshold-protected.