how to scale

Boston MarketIn the 1990s the popular business press had a few recurring themes, such as growth in demand for PCs, anxiety over Asian economic strength and Latin American debt, and the hazards for American businesses of expanding too rapidly. On that third point the example was Boston Market so often that for lazy writers the name became a kind of shorthand for how not to scale. The chain of fast-casual restaurants began in 1985, grew explosively, borrowed too much, and by the end of the ’90s was bankrupt.

There’s more to take from that story than the hazards of leverage; with lower interest rates the Boston Market parable might have been one of nerviness rewarded. It’s also about judgment and perspective. Whenever you’re on a roll the temptation is to push harder: momentum itself is a a resource to take advantage of.

Here in the system office we spend some time trolling for ideas to scale, wondering when good practice at one state university is ready to carry over to another. We can do a lot of good when we guess right.

So imagine my surprise on finding a compelling argument against scaling at all. The authors of this paper, faculty at San Luis Obispo, begin with an observation about the prevailing metaphor for higher education, which is industrial and thus prioritizes standardization:

Industrial era manufacturing methods attempt to minimize diversity and its sources through quality control. In these metaphors, profit is assumed to be maximized through economies of scale, where variation accrues as a loss in profit.

If you replace the word “profit” with “efficiency,” you get a pretty good account of my job managing transfer credit. Uniformity is good. Idiosyncrasy produces waste. The paper continues:

Using the metaphor of complex systems instead opens possibilities for a plurality of valid “truths” to simultaneously exist, since the underlying premise is that systems are more than the sum of their parts.

For the last couple of years I’ve been fishing around for good ways to propagate ideas across communities, something less ham-fisted than policy and credit articulation. So I liked reading this paper, which brought to mind other models for group work: the “practitioner network,” exemplified in the California Community Colleges by the RP Group, for example , or the Networked Improvement Communities supported by the Carnegie Foundation, or the model called “communities of practice,” now a couple of decades old.

What they have in common is a loose-knit, organic property that feels truer to human interaction. People join or leave as time permits. Once in the group, listening and turn-taking are as important as any other contribution. Roles and obligations are fluid.

As an administrator, one part of me loves this while the other is reaching for the Xanax. Gone are the rigid to-do list and reporting deadlines; you can’t even maintain a decent listserv. I mean, for the love of God.

Intuiting my interest in such things, last year the now-dean of undergraduate education at San Francisco State gave me a book called Scaling Up Excellence. It describes this continuum in religious terms, running from Catholic (“mandating that new people and places become perfect clones of some original model”) to Buddhist (“encouraging local variation, experimentation, and customization”).

Somewhere between those theologies is the mix likeliest to work for universities like ours, thriving on local shared governance, while chained together on the state’s higher ed road crew.


A couple of weeks ago I joined administrators at a meeting of the National Association of System Heads. One of our advance readings was an article on Collective Impact, which included these five Conditions for Collective Success:

  • common agenda
  • shared measurement systems
  • mutually reinforcing activities
  • continuous communication
  • backbone support organizations

This list nicely captures what I’ve seen working around the CSU. But I’d emphasize that these are not only the minimum requirements, but also the most you’d want to do. In other words, once everyone has agreed to the goals and metrics, and you’ve provided a means of communication, shared activity, and support, get out of the way.

20090925_friedchicken_560x375The recent interest in scaling two innovations in particular has made it suddenly important for us to get this right.

  1. HIPs. In our efforts to institutionalize CSU offerings of high-impact educational practices, like learning communities, undergraduate research, and community engagement, we’re just about done with four of the five points, but the shared measurement systems remain a work in progress. Here the challenge is to do only that: create the metrics in a way that recognizes common ground where it really exists, without homogenizing everything else.

The clock is ticking: my boss in the corner office keeps upping the ante with dedicated student success money. My campus colleagues and I have successfully argued that it belongs here, with the learning: make it visibly valuable and relevant, and students will persist and graduate.

For now we’re winning, but if we don’t back our claims up soon with some research, the fad will pass, our moment of momentum missed. We’ll be the next, you know, Boston Market.

  1. Linked Learning. Last month I was witness to a bigger, splashier meeting around scaling right. For many years, grantees of the James Irvine Foundation have been quietly building a coalition to blur the boundaries between liberal education and job training. The ideas of this movement – Linked Learning – go back further than this iteration, and they will persist beyond it, too. The state of California is putting half a billion dollars into Career Pathways Trust, a development most observers attribute to Irvine’s successes.

In both of these cases, good work is getting money and support at unprecedented levels. And so both leadership groups face hard questions about how to scale – what is and isn’t in scope, what the templates must include, which details can be sacrificed while still maintaining “fidelity to the model.”

I worry about falling into that same industrial-era paradigm, and excessive homogenization.

There’s another serious risk here, and that’s simply antagonizing others who do good work. In the words of one of my colleagues in this work:

As a field we must guard against sounding ‘holier than thou,’ — that is, sounding like we are the sole guardians of quality, the only initiative that really knows what it is, and everyone else is just going through the motions. There are many other people out there who care about quality as deeply as we do.

In other words, as we draft these papal bulls on high-impact practices and linked learning, we’d better leaven the orthodoxy with a little Buddhism, and remember what we really want to scale.



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s