David Maister - Professional Business, Professional Life
David’s ResourcesAbout David
NEW! Browse my materials by topic of interest:StrategyManagingClient RelationsCareersGeneral

Passion, People and Principles

Metrics for Internal Service providers

post # 201 — September 28, 2006 — a Managing post

Ted Harro, a regular participant in the conversations on this blog, has submitted this topic for us to react to (These are Ted’s words):

Many people serve as internal consultants or advisors. The more they act with the same intensity as outside providers, the more value they will bring to their company, the better they will get, and the more fun they should have.

Internal consulting is a bit different from being an external consultant. You have a very limited market, and you can’t fire your clients. Since prophets are rarely celebrated in their own land, you are positioned differently with your ‘clients,’ and must engage with them differently

One of the challenges is figuring out the right metrics to use in monitoring and managing internal consulting groups (or any other staff advisors.)

A few obvious metrics used by outside providers still apply to in-house providers: client satisfaction (post-engagement surveys, internal referrals) and staff (employee satisfaction surveys, skill development monitoring)

You could also try to monitor “market positioning” – share of core capability work done in-house, share of overall capability work ‘under management’ (for example, acting as a broker for outside resources, such as the in-house legal department managing the outside law firm providers.)

The most difficult area is financial performance. Most internal groups are cost centers. Few companies still use “internal pricing and bill-back” systems anymore because it turned into an administrative nightmare.

A few initial thoughts:

Measuring the utilization of in-house staff (hours engaged/people hours available) still works, though it is a hygiene measure and does not describe effectiveness.

Leverage could work (ratio of total people to senior practitioners, defined by credentials and pay grade)

It would be tough to use a margin measure. I suppose you could do a ratio of overhead cost to employee cost.

Rates are the real question. If in the outside world rates are the market’s perception of your unique value, how do you do that in the inside world? Overall department budget? ROI/hours invested?

(but ROI can be a tricky number for a lot of internal consultants)?

Clearly, if it could be measured, ‘value’ is the ultimate measure for internal consultants.

In many firms, internal groups generally track it by how much budget they get, but this seems incomplete to me somehow.

(End of Ted’s Question)

Ted has posed so many questions that I’m going to ‘hold my fire’ until all of you have the chance to jump in. (Also, I’m traveling in Europe and I’m a bit jet-lagged!)

However, I would extend the challenge by pointing out that, in these days of outsourcing, internal groups probably will need to develop metrics that are at least part-way comparable to those being used by the outside providers pitching your management for YOUR job!

So, what metrics are used where you work to measure internal consulting / advisory / staff work?


Jennifer said:

Great questions! My metrics are client satisfacton and staff satisfaction. But I’m still struggling with the main contradiction. In an external consulting role – I convinced clients I could add value, they paid my fees, and I hired more staff.

In my internal consulting role, the people I serve, and the way my budget is allocated are only vaguely related. I don’t think going back to hourly rates is the way to go, but I do struggle with the fact that I can convince internal people I’m going to add value, because my marginal cost to them is nothing, but can’t easily increase my budget, even if I am clearly adding value to my internal clients. Another way of putting your questions.

posted on September 28, 2006

Dennis Howlett said:

Both Jennifer and Ted are right. Applying bald financial measures is neither helpful nor meaningful for all the reasons stated. And I’m a finance guy saying this!

Internal surveys are inevitably skewed either in favour or against depending on the organisation and they are lagging indicators. What’s needed in my view are real-time indicators. 

I’m very much in favour of light touch measurement because even though those may be relatively crude, they don’t get in the way of what people are trying to do so simple polls included in the company’s intranet or available through RSS is one way.

I’d encourage the setting up of group blogs where you have Project blog and People blogs tied together so that issues, milestones etc can be both discussed and monitored in real time. This helps resolve the effectiveness problem because folk can discuss the detail on an ‘as you go’ basis. Also avoids the interminable round of progress meetings that don’t achieve anything other than murtual back slapping or verbal fist fights.

I’ve reently done something like this on a topic where we used GoogleGroups, and in 4 days we had a 78 piece threaded discussion among 45 of us – most were lurkers listening in with maybe 10 real contribs. We’re now distilling this into a single blog post where all sides are represented and where the arguments are summarised with appropriate conclusions. The general sense was that EVERYONE learned something of value – even when disagreeing.

We have another discussion under way where so far the chat count is around 22. Same applies. The point is that we’re treating it as both a learning and action exercise which takes our thinking forward and so adds demonstrable value – if not capable of direct financial measurement.

Hope that helps.

posted on September 28, 2006

David (Maister) said:

It helps a lot, Dennis. And it also goes to show how much more progress we can make in general when we stop asking “How can we measure?” and start asking “How can we improve?”

It unleashed a totally different mindset.

posted on September 28, 2006

Charles H. Green said:

It seems to me that “metrics” can get overdone, particularly when the thing being measured is complex and/or intangible. In those cases, a distinction may be useful.

When I helped run the consultant evaluation process in a consulting firm, it became clear to me we could do one of two things. Either we could define huge numbers of specific measures—ideally behavioral in nature—or we could invest a lot in process—senior people discussing consultants in depth. Firms like Bain went the first route (this was in the 80s; we bent more in the second direction).

I have seen a similar distinction made in evaluating complex training programs; you can either devise very convoluted 2nd and 3rd generation behavioral metrics—or you can instead just get very quantitative about the subjective impressions of trainees and their managers. Suppose you taught a course on trust: do you invent measures of trustworthiness, or do you simply ask people whether attendees have become more trustworthy?

The value of the latter approach came clear to me years later when David and I co-authored The Trusted Advisor. I got an inquiry from someone who had developed four pages of behavioral metrics to describe each of the four components of the trust equation (credibility, reliability, intimacy, self-orientation).

The real kicker comes in when you try to link measurement and rewards, as we are all taught to do these days to be good do-bees.

Measuring reliability with tradititional behavioral metrics isn’t so hard. But try measuring self-orientation. Worse yet, trying figuring out to monetarily incent people to become less self-oriented. There’s an absurd contradiction lurking there.

And it’s not far below the surface. If you have to work too hard, get too many generations and proxies away from very simple end-user based metric, then something’s not right. Look for a Plan B.

If you can’t measure results, then measure people’s opinion of results. If linking measurements to results results in bizarre behavior, back up a few levels and remember the KISS principle.

Frankly, as I get older and grayer, I find myself more and more recommending values, principles and culture over measurement and reward.

posted on October 9, 2006