Though it could expand his outlet’s reach — especially, in his words, as email becomes increasingly important “as a distribution channel with social media declining as a traffic source” — Brodbeck didn’t think creating an additional newsletter was an optimal use of reporter time in the zero-sum, resource-strapped reality of running a hyperlocal news outlet.

“As much as I would love to have a 25-person newsroom covering Northern Virginia, the reality is that we can only sustainably afford an editorial team of eight across our three sites: two reporters/editors per site, a staff [photographer], and an editor,” he said. In short, tapping a reporter to write a morning newsletter would limit ARLnow’s reporting bandwidth.

But with the exponential improvement of AI tools like GPT-4, Brodbeck saw an opportunity to have it both ways: He could generate a whole new newsletter without cutting into journalists’ reporting time. So last month, he began experimenting with a completely automated weekday morning newsletter comprising an AI-written introduction and AI summaries of human-written stories. Using tools like Zapier, Airtable, and RSS, ARLnow can create and send the newsletter without any human intervention.

Since releasing the handbook, Amditis has heard that many publishers and reporters “seem to really appreciate the possibility and potential of using automation for routine tasks,” he told me in an email. Like Brodbeck and others, he believes “AI can save time, help small newsrooms scale up their operations, and even create personalized content for their readers and listeners,” though he raised the widely held concern about “the potential loss of that unique human touch,” not to mention the questions of accuracy, reliability and a hornets’ nest of ethical concerns.

Even when instructing AI to summarize content, Amditis described similar challenges to those Brodbeck has encountered. There’s “a tendency for the summaries and bullet points to sound repetitive if you don’t create variables in your prompts that allow you to adjust the tone/style of the responses based on the type of content you’re feeding to the bot,” he said.

But “the most frustrating part of the work I’ve been doing with publishers of all sizes over the last few months is the nearly ubiquitous assumption about using AI for journalism (newsletters or otherwise) is that we’re out here just asking the bots to write original content from scratch — which is by far one of the least useful applications, in my opinion,” Amditis added.

Brodbeck agrees. “AI is “not a replacement for original local reporting,” he said. “It’s a way to take what has already been reported and repackage it so as to reach more readers.”

In July, it closed Hill Now, its site covering the Capitol Hill neighborhood, and merged its coverage with another one of its sites, Borderstan, which focused on a handful of neighborhoods near Dupont Circle. Then, in December, with revenue flagging, the company also shut down Borderstan, which was run by two full-time employees and averaged 85,000 unique visitors per month.

“We were not generating enough advertising revenue to pay for the brand of journalism we were creating,” Local News Now founder Scott Brodbeck told me.

The company still operates two hyperlocal sites in D.C.’s Virginia suburbs — ARLnow in Arlington and Reston Now in Reston. The company has five full-time employees, including Brodbeck. One of the former Borderstan staffers has joined ARLnow; the other has left the company.

Local News Now is blessed with unusually fertile ground for a local news site; the D.C. metro area has the highest median income of any in America, and its suburbs are even richer. (Arlington County has the sixth highest median household income of any county in the United States. Fairfax County, which includes Reston, ranks even higher, third.)

Local News Now exceeded $500,000 in revenue for the first time in 2016. “The sites in Virginia are solidly profitable and the company itself is profitable,” Brodbeck said. ARLnow averages 250,000 unique visitors per month and Reston Now has 70,000 uniques.

Brodbeck and I discussed the thinking behind the decision to shutter the two D.C. sites, what’s next for Local News Now, and why he thinks digital advertising can still be a sustainable revenue source for hyperlocal outlets. Here’s a condensed version of our conversation, edited for length and clarity.

I would argue that the same applies to banner advertising. It gets a bad rap as not particularly effective. I’m sympathetic to that on a more national level. It seems like the biggest thing is programmatic and targeting. How do we target specific users? That is a very inexact science. On Twitter not too long ago, I was going off about how the ads I see on the web, Facebook excluded, if they’re targeting to me, I have no idea how they’re doing the targeting.

On local, it basically is targeted. It’s automatically geotargeted. If we have an advertiser who is in Arlington and trying to reach people in Arlington: Guess what? We reach people who live and work in Arlington exclusively. I think relevance is important in advertising, and local is automatically relevant to connecting local advertisers to local readers. I think that display advertising is going to continue, though I hope we continue to find other ways to connect our clients with our readers.

@damianradcliffe @jsb We've got to address mobile more at the local level. Would love to talk more about that. #LION15

— Josh Stearns (@jcstearns) October 3, 2015

That’s problematic, considering how many people are getting news on their phones. Jesse Holcomb, an associate director at Pew Research (and recent Press Publish guest) presented on the state of local news, based off this Pew report from the spring.

“The local news experience for American audiences is mediated on the social web,” Holcomb said. “Younger news consumers are getting news about their community on their mobile phone and they’re doing so at rates just as high as other groups, if not higher.”

Local publishers’ reliance on banner advertising is “scary,” said Michele McLellan, the founder of Michele’s List, a database of local news sites. As she put it in a wrapup post on the conference:

There were encouraging, if fledgling, signs that some sites are looking beyond banner display advertising. Supplemental sources include crowd-funding, membership programs, sponsored content, and events. For now, however, banner display advertising is a mainstay…

That’s a vulnerability. If any source falters, companies with diverse revenue sources have others to fall back on. Online display advertising is taking a lot of hits these days, including ad blockers, concerns about impression fraud, and concerns that people simply ignore banner ads.

Some publishers are expanding in this area: Scott Brodbeck, founder of ARLNow and a network of hyperlocal sites in suburban Washington. D.C., described the new sponsored advertising methods he’s using.

.@scottbrodbeck has sponsored columns written by advertisers. Meant to be information on useful topic to readers, not a sales pitch. #LION15

— Ren LaForme (@itsren) October 3, 2015

And Joe Hyde, founder and publisher of the Texas-based San Angelo Live, created a custom ad size for his email newsletters — they’re 570×216, to match the size of an iPhone screen and a finger.

Site founders often develop for the desktop web first, with mobile-optimized or responsive sites sometimes an afterthought. Brian Wheeler, the executive director of Charlottesville Tomorrow, discussed three crowdfunding projects the site has done. Two, where people saw a “pain point,” were successful, but the third — a project that aimed to enhance Charlottesville Tomorrow’s mobile site to make it more similar to the Texas Tribune’s — failed, raising less than a third of its $15,000 goal.

“We thought it would be an easy success via Kickstarter,” Wheeler said. “It did not get anyone motivated to help us. It was too backroom, too infrastructure. People said, ‘I don’t think your site’s that bad on my phone.'”

You can find LION’s blog here, video of some of the presentations here, and the Twitter stream here. In addition, conference attendees took notes in shared Google Docs — here’s the combined coverage of the Friday and Saturday sessions.

A previous version of this post referred to Brian Wheeler as founder of Charlottesville Tomorrow. He is executive director.

GoLocal24 is a different kind of online startup. It’s for-profit, unlike so many of the city startups we’ve seen. It’s fueled solely by advertising revenue, while many other startups get no more than 20 percent of their income from ads and sponsorships. It claims a 20 percent profit margin, when a break-even balancing of foundation, grant, membership, events, and ad income vs. expenses serves as a wider model among its nonprofit peers.

Sure of its model, it’s now building out a small chain. Started in Providence in 2011, expanded to Worcester (50 miles west of Boston and 40 miles north of Providence) in 2012, GoLocal is busting out of its New England geography and going about as far as the continental U.S. allows, announcing today its third site in Portland, Oregon.

The Portland site — set to launch this spring — will double the number of GoLocal staffers. Portland will have six editorial staffers and four business-side ones, equaling the 10 total staffers putting out the Providence and Worcester sites. GoLocal could have picked Portland, Maine — much closer to home — but wants to announce itself as a national company and a national model, so it’s the Portland on the other side of the country that will do that. If Portland succeeds, Fenton and his investors plan more GoLocals — and have already bought dozens of URLs to enable that.

It’s a noteworthy move, coming on the heels of the near-final implosion of Patch this week. Like a vast stadium slated for demolition, the slo-mo demise of Patch — just recently the No. 1 hirer of U.S. journalists — has been painful to watch. Through 2013, AOL CEO Tim Armstrong dribbled out information on key questions: Which sites will stay, and which will go? To whom, and how? Or would they just close in the middle of the night some time this year? AOL had been remarkably unclear with its publics, its staff, and its readers about those questions. Then, the expected news yesterday: New majority Patch owner Hale Global would dispatch more than 75 percent of the remaining journalists. Now expect the new ownership to harvest as much of the traffic and brand value of the sites at as low a cost as possible, using local aggregation and low-cost, programmatic sales. What once the biggest experiment in local journalism will live on largely in ghost form.

In midsummer, as Patch’s direction was becoming clearer, we saw the usual spate of posts on the impossibility of “hyperlocal.” We can conclude that it’s the hype part of hyperlocal that’s been overplayed. Local is local — we know it when we see it, and we know when it’s well-done and when it isn’t.

GoLocal’s decidedly contrarian turn on local is worth understanding in the Patch context. Providence is a big, old city. Its metro area stretches into Massachusetts and touches eight counties. What happens in the Elmhurst neighborhood is different than what happens in the Elmwood neighborhood — yet the residents of both care about both their neighborhoods and the wider metro area, its crimes, schools, and only-in-Providence brand of politics. GoLocal emphasizes the city-wide issues — knowing that’s what (duh) brings in the engagement of audiences. It’s really just an application of old newspaper principles to digital local: It’s both city and neighborhood news that stokes readership. It’s worth stating that principle, given that Patch’s belief that hyperlocal was enough to build a business might have been its most fatal flaw.

Merrill Brown, a veteran and smart observer of local back to his early days plying online at MSNBC.com, is now a board member and minor investor in GoLocal. Over the holidays, he wrote a piece for Forbes comparing Patch and GoLocal. His conclusion and his GoLocal pitch:

Markets are loaded with advertisers eager to find solutions beyond the failing, ineffective local newspaper and the overpriced television station. Hopefully, entrepreneurs and investors will take away from the Patch mess that there’s a better way to serve audiences in local markets, one with far more potential than [a] random collection of suburbs.

It’s worth noting the kinds of communities that GoLocal has picked to operate in. While Portland is in economic recovery, with jobless rates falling below 7 percent; Providence’s is more than 9 percent, one of the highest in the country. While Fenton says that GoLocal looks at a diverse set of data points in picking its locations, another fact stands out: Each of its first three locations is home to struggling dailies.

The Providence Journal has been in a long tailspin of reader, ad, and staff loss and has been recently put on the market by owner A.H. Belo. The Worcester Telegram & Gazette, sold by The New York Times Co. to John Henry in the Boston Globe deal, has survived recent years in better shape than the ProJo, but it too is now up for sale. (Dean Starkman, a Columbia Journalism Review editor, former Providence Journal reporter, and now regular contributor to GoLocal, details the Telegram story here and the ProJo’s here.) In Portland, The Oregonian — by far the largest daily in the state — is an Advance paper, which means it’s retrenching on print, its community footprint receding along with it.

It’s not quite a vacuum, but it’s a vulnerability. Low-cost, high-energy startups, like GoLocal, can exploit the weakness.

What is GoLocal’s playbook?

• Advertising is the core business model. All of GoLocal’s revenues are ad-based. Fenton, a lead salesman as well as publisher, sells instinctively. A 50-year-old ad agency veteran, he knows Providence and has wide relationships in the business community.

  Let’s distinguish here between GoLocal’s local opportunity as compared to the national one — the one that’s spurring on a burst of new national news sites (“The newsonomics of why everyone seems to be starting a news site”). For GoLocal, it’s local institutions that are the bedrock of the business model — education, health, advocacy. GoLocal works with local advertisers; 30 percent of the revenue is sponsorship-based. All selling is optimized by technology, raising average CPMs well beyond what run-of-site programmatic advertising will fetch. Fenton says ad revenues have doubled over the last year and he expects a similar doubling this year.

  We see others pursuing similar strategies. Scott Brodbeck’s ARLNow too focuses on advertising and is profitable. While defining itself as “hyperlocal” (“In suburban D.C. a network of hyperlodal news sites expands and bets on local advertising”), it’s worth noting that Arlington is a city of 220,000 — far bigger than the average Patch hyperlocal of roughly 50,000 people.

• Content is broadly defined. It’s the occasional big local news break, good data-driven enterprise, trend pieces, lots of one- and two-source quick reads, press releases, and listicles. You can move from much depth to little in the move of a mouse, much more like the BuzzFeed or Business Insider model than that of a local newspaper. More than 50 contributors fill the site with content, from those with impressive credentials like Starkman to a quite young staff learning the ropes, making up for their inexperience with energy.
• Content is cheaply produced. GoLocal’s small profit is hard won. It squeezes the local penny, with its full-time staffers get paid no more than a third of what we see national journalists getting in all those highly publicized startups. Fenton’s arithmetic is compelling: He says he can produce content at 1/16th the cost per story as compared to a daily paper. Given its low compensation, it’s seen a fair amount of turnover in its brief history.
• Social is juice. It’s in the swim, with “180,000 social media relationships.” “If you break big stories, you get tips,” says Fenton. GoLocal works all the social levers for sharing, and consequently enjoys traffic, despite its youth, that seems to equal a third to a half of the ProJo’s.

What about GoLocal’s journalism? It’s a startup in every sense of the word. Knowledgable observers credit it with some agenda-setting journalism and lots of engaging coverage — and fault it for incompletely sourced stories that may over-promise. Check out its home page. It’s in constant motion. As a true startup founder, ad leader, and effective editor-in-chief, Josh Fenton runs a work-in-progress. That’s fair to expect from a two-plus-year-old startup. As the company expands, one big test is to come: How much more experienced and stable editorial leadership does it hire on?

In Portland, it will face a bigger challenge. Fenton has regularly visited Portland, doing his homework, aware that it’s a far different audience (Portlandia is only partly satiric) and market than his northeastern. It’s also more journalistically diverse, with one of the country’s best urban weeklies in Willamette Week, in addition to The Oregonian, other weeklies, active TV stations, and an innovative public media network in Oregon Public Broadcasting. To try to meet that challenge, GoLocal’s Portland operation will have the same number of full-time journalists (six) as do Providence and Worcester combined.

It’s important to consider where we’re at in local media models. They’re struggling. Just this week, the Knight Foundation put millions more into helping build the sustainability of local online news. That grant follows up a comprehensive and to-the-point Knight report that showed, unsurprisingly, that online startups need to “create space for business capacity.” As well meaning as it is, it sounds like a Republican punchline to a joke about a Democrat’s anti-business leanings.

We can ask about for-profit vs. nonprofit local news models, but our question isn’t whether one is better than the other. I’m agnostic on that question: More good journalism, however funded, is better than less.

That’s one reason the GoLocal model is an important one. Nobody had to tell Josh Fenton about the need to build business capacity; he’s an ad guy who knew he was running a business before day one.

We see what looks like two kinds of local ships headed in opposite directions: the (mostly) nonprofits largely run by dedicated, community-minded editors who are trying to figure out the world of business because they have to — and the for-profit model that operates like any other small business. Undoubtedly, the nonprofits are better staffed editorially — more and more experienced journalists — than GoLocal, which carefully spends each dime to build its business along with its news-gathering capacity.

That’s where the Fentons fit, and why we need more of them in the mix. Editors-turned-publishers can learn the nuances of digital business — but advertising natives like Fenton bring a far different experience and perspective — in addition to all those relationships — to local media creation.

Fenton is a believer in profit-seeking businesses. Though he doesn’t disparage the nonprofit local model, he believes that going for-profit provides an edge of sink-or-swim discipline to the endeavor. I’ve heard similar sentiments from other news startups choosing for-profit status, as they make the case that for-profits selling advertising are regarded very differently by businesspeople that non-profits. Both points — the discipline edge and commercial acceptability — are debatable, but perhaps vital as we look at the reconstruction of local news in our time.

In an ideal world, we’d see a better meeting ground. We need serious business natives juiced about the building of local digital news companies, working in tandem with tenacious editors and reporters out to serve communities. Getting the models right — sooner rather than later — is essential. It’s no accident that GoLocal has gone into the under-served cities it has. The critical question about local journalism is replacement: how to replace the vast staffs, beats, and inches of reporting that have simply disappeared since 2007. Replacement will never be (and shouldn’t) one-to-one, but volume — and the quality of that volume — matters across the country.

Photo by Christopher Michel used under a Creative Commons license.