Charles Lewis, one of the luminaries of nonprofit investigative journalism, sees a culture clash brewing as the sector continues to grow, covering what shrinking legacy media may miss and, more recently, innovating with powerful reporting techniques.

On the one hand, foundations big and small want metrics that demonstrate results analogous to assessments they apply to arts projects, social service initiatives and advocacy work.

On the other hand, Lewis wrote in a white paper last month, "veteran editors and reporters, particularly of the investigative ilk, have an inherent, almost visceral dislike of audience measurement and engagement strategies." Instead they see themselves "as intrepid hunter-gatherers of information" who overcome a host of obstacles to produce important, even heroic, journalism.

The conflict might be academic were it not for the current state of play in nonprofit funding. Established nonprofit news sites need second and third rounds of support from foundations, and startups find foundations "feeling a bit overwhelmed and besieged by proliferating prospective grantees," Lewis wrote.

In an e-mail interview, Lewis added, "Subjectivity is a serious occupational hazard for any grantor attempting to measure impact...Some foundations seem to be somewhat obsessed with these questions and issues, and others, not so much (especially smaller foundations with very few staff)"  But he expects the level of scrutiny to keep rising.

The white paper, which Lewis co-wrote with Hilary Niles, assesses the problems of measuring impact journalism and proposes the starting outlines of best practices. Written for the Investigative Reporting Workshop at American University, and underwritten by the McCormick Foundation, the paper is a lengthy but worthwhile read.  Caroline O'Donovan did an excellent job summarizing it in this Nieman Lab piece.

I'm not going to try to cover the entire scope, but here are a few themes:

Difficulties: If impact and outcomes are the true markers of an effective investigation, quantitative indicators of quality are likely to miss the mark. "Targeted reach" to decision makers may be more important than broad readership numbers. Sometimes remedial action on the problem spotlighted takes years rather than months. The investigative reporting process is more time-consuming and open-ended than advocacy work, which often assumes a desired outcome and marshals evidence to make the case.

Proxies: Audience is at least a starting point, with the qualifier that fluff may outdraw substance. Lewis and Niles recommend engagement as a more sophisticated measure of reach. Mentions, links and conversations are particularly susceptible to measurement in the digital space -- and can indicate that the work in question didn't fall with a thud in an empty forest.

Models: The authors point to several pilot projects with titles like "The Center for What Works" and "Spreading the Zing" that tackle the challenge of measuring social outcomes. Perhaps, Lewis and Niles argue, some of the framework and classification of desirable outcomes can be adapted by individual projects trying to make their case rather than reinvented from scratch.

Best practices: Lewis and Niles suggest that the Wisconsin Center for Investigative Journalism is well-along in documenting the "ripple effect" of its best work. As explained by the center's Lauren Hasler in a 10-minute video, this consists of investing time and a little money into capturing data on the spread of a story and creating a narrative with evidence of what the investigation accomplished.

The authors give passing reference to what I think is a potential ace in the hole for the nonprofits: crowd-sourcing and crowd-funding. Though they yield input rather than outcome measures, both are direct evidence of an engaged audience, willing to contribute money for investigative work or do some of it for free. Crowd-sourcing and crowd-funding fit the digital era and the non-commercial idealism of nonprofit news organizations.

Also, while Lewis and Niles are focused on the nonprofit sector, I am curious whether the assessment challenge should be applied to investigative efforts at newspapers and other legacy media as well.

Some top metro papers like those in Seattle, Milwaukee and Tampa Bay determined several years ago that even as newsroom cuts become a necessity, investigative capacity should be maintained or even expanded. When Gannett was doing research as it introduced paywalls at its 80 community papers last year, it found that investigations were the top of list of "passion topics" readers were willing to pay for.

Impact gets identified and celebrated in the Pulitzers and other contests, but I am not so sure there is a reliable way to assess the volume and quality of the investigative work, if any, that your hometown paper still provides.

Not every published investigation is a gem. We have all been asked to read lengthy pieces that reflect efforts but did not come up with all that much. Also, there is a genre that rides along with law enforcement or government auditing work, creating an impression of impact but a bogus one. And my antennae are up for financial exposes that sometimes ignore basics of risk and fiduciary obligation.

Having done some, directed some and read several Russian novels worth of investigative projects, I think there is a common, sweet-spot design to the best of them. You need to identify and document a problem of some consequence (bigger than the typical TV I-team effort to help a guy get some shoddy construction fixed).

The problem should do more than leave the reader shaking his head and saying that's unfortunate; it should be actionable. Good investigations mobilize a level of citizen awareness and often indignation. Then something happens as a result. That something can be governmental or non-governmental or both.

For instance, the recent Tampa Bay Times/Center for Investigative Reporting project identifying America's Worst Charities is likely to result in tougher regulations in a number of states. But it should also motivate prospective donors to do a little due diligence on where the money goes before responding to a heart-tugging appeal.

Lewis said in the e-mail interview that newspapers are 50 years into evaluating the impact of investigative reporting, particularly when it changes "public laws or policies." His framework might be less interesting to them except for "the intriguing issue" of collaborations between profits and non-profits like his joint creation of an investigative unit for the Washington Post with a Ford Foundation grant.  If those collaborative efforts grow, the for-profits will be pulled into the assessment discussion, he said.

The big impact of spectacular projects like the Boston Globe's investigation into pedophile priests or The Washington Post's investigation into shameful conditions at Walter Reed Hospital are self-evident. But a more modest everyday level of holding governments and other powerful institutions accountable is equally essential.

Lewis is a serial entrepreneur in launching investigative units and support organizations and some years ago won a MacArthur Foundation "genius" grant for his work. I think he is onto something again. Fleshing out the embryonic conversation about what works and why it matters is going to be essential to keep the resources to support investigative reporting flowing.