Bronte Media

Scandal! Scandal! Scandal in the local ad world!

In a truly bizarre occurance, John Battelle, noticing that a months-old post on MerchantCircle was buzzing with disgruntled comments by small businesses claiming to have been harassed by an automated MerchantCircle voicemail alerting them to a negative review. He suggested it was the magestical nature of search and word of mouth, since that post was ranked number three in a Google search related to Merchant Circle.

Peter Krasilovsky, author of the excellent LocalOnliner blog, had also written about MerchantCircle in the past and received a similar inundation of comments claiming fraud. But he smelt a rat. All the comments are in a similar voice, referring to the word scam and always referencing previous comments. The IP addresses map to different ISPs, so if it were a false smear campaign, it was well enacted.

But were the automated calls even placed? Matt Mcgee, who authors a blog for small business marketers, puts an even more bizarre twist on the story. A company he works with, got one of the MerchantCircle automated voicemails saying their busniess had a negative review. Except they didn’t. No one had rated their business.

In Battelle’s latest post, the CEO of Merchant Circle Ben Smith all but admits to using the automated voicemails. “We have been experimenting with many different methods of informing merchants when their listing changes,” he says. Lamenting that “for a while the entire team of 9 personally contacted local merchants”.

So there you have it. Looks like automated technology went wrong, contacting too many businesses (a shame they didn’t screw up by telling people they had positive reviews) and then a strange comment spam/legitimate word of mouth merchant backlash.

And like sands through the hourglass, these are the days of our lives…

Today over at Homethinking, we launched RealEstateVoices, a social news site for real estate.

The site is based an open source project called Pligg, which replicates many of the features of sites like Digg and Reddit. Even though those sites have been growing like weeds, editorial direction is concentrated among a few that enjoy technology, weird things and not much else.

Other verticals are launching, from search to celebrities. But there isn’t a clear forum for real estate news, so we decided to start one.

The challenge seems to be getting across the 4-8 votes per story chasm. To that end, I joined with a few friends to act as community leaders and hope to add one or two more in the next few weeks. Hopefully we will.

To encourage contributors we are giving link love and may put an adsense banner on the site and then use the proceeds for a bar tab for the leaders at say one of the Inman conferences (Mike has done that in the past with Javablogs and it seemed to work well). If you have any other ideas, would love to hear them in the comments.

Heh Heh Richard Parsons said that Facebook and YouTube valuations were astronomical.

“Valuations that are put on those businesses that currently make no money are astronomical and you have to have a big leap of faith,” the newspaper quoted Parsons as saying in an interview in its Friday edition.”

Having partaken in the greatest merger of all time, between AOL and Time Warner, Richard is an expert in Internet valuations, so he should know.

Don’t Cry for Me Yahoo. One of the different aspects of the recent rise of online advertising is that the gains have been relatively concentrated in search based means.

Folks like CNET and iVillage haven’t grown that strongly compared to firms like Google. The riches have been concentrated amongst a relative few.

Performance of the online divisions of newspapers have masked a simple fact: most of the gains they have posted have been through their classifieds properties and not their flagship web sites with rich media banners.

That makes sense because classifieds are essentially three category search sites (real estate, automotive and recruitment).

But I was surprised to see, while sifting through Verizon’s earnings release for Q2, that “In the second quarter, VIS’ domestic online directory and search service, SuperPages.com, achieved revenue growth of 12.2 percent compared with the second quarter of 2005, and Internet yellow pages searches increased 143.9 percent over the same period.”

12 percent is pathetic compared to the growth of other search based firms, especially in the face of such impressive query growth. Superpages.com advertisers do not pay on a performance basis for the main part so query growth is not as tied to revenue as it is at other firms like Google.

In the face of the Yahoo postulation, it is interesting that a pure search based business like Superpages.com is also struggling for quality growth.

Barry Diller, my favorite non-Murdoch-surnamed media executive, has rightly changed course from building yet-another-ad-network-slash-server to letting Google, MSN and Yahoo whore themselves off recklessly. Not one of the three will grow their revenue in any substantial way from Ask. Rather they will offer ever higher percentage shares and more onerous guarantees because the value in the search world has always gone to the firm who owns the consumer relationship not the advertiser relationship.

For Yahoo and Google that wont really matter that much because they make so much on their own properties that Ask-type deals are cream on the top. MSN, for whatever reason, thinks it should have its own ad network.

Diller has recognized that all his efforts must go into growing consumer usage of Ask, itself no small task. At least he will have given himself the best chance of success.

Following hot on the heels of the Who’s Calling sale, eStara has itself been acquired for a similarly priced $48m by Art Technology Group. The transaction includes $2m in cash and the rest stock.

While the two firms both operate in the click to call world and have both dipped their toes (even it is mainly from a PR standpoint) in the pay per call advertising world, there are significant differences. Who’s Calling mainly helps offline advertisers like car dealers figure out their return on investment, whereas eStara is more known for helping online merchants with their customer service.

ATG is a provider of commerce tools that helps retailers figure out why consumers abandon their shopping carts among other things and helps them deliver a better user experience.

EStara has been growing fast in recent times, on track to over double their revenue this year. From the Reuters wrap: “EStara reported revenue of $6.5 million in the first six months of 2006 and $7.4 million for full-year 2005, the statement said.”

I have heard rumors that there is more financing/m&a activity on its way for click to call, call tracking and pay per call firms that will bring consolidation to the industry. It makes sense as the call tracking component becomes a commodity and other types of advertising firms and analytics vendors look to measure and invoice using the phone call as the billing event. Stay tuned.

Lyn Chitow Oakes, SVP Marketing of Jingle Networks posted a very thoughtful analysis refuting some of my points in a comment to my last post on the economics of free 411. So I thought I would re-run the analysis and revise my assumptions based on her feedback (a.k.a. correct my error-ridden napkin analysis)

On the cost side, Lyn says that a blended average of automated and operator assisted calls brings the cost down to 15 cents per call not 18 cents. So our theoretical 500,000 calls per day cost $75,000 per day not $90,000.

One the revenue side, she said “Our average pay per connect is much higher than $2, and we typically wouldn’t do a deal as low as $2 per connect.”

OK that’s fine. Firstly, there may be a misunderstnading in the definition. Mine was a net figure. That is after the ad network, in this case Ingenio, takes out its 30-50% cut. On that figure she said “Without going into specifics, this figure is not very close to the typical revenue sharing percentages we are seeing.” With such a broad range I don’t see any reason why it could not be close but let’s give the benefit of the doubt and say 70/30.

Lyn hints that “A connect fee of $8 with a response rate of 7% produces an average revenue per call of 56 cents. One call like that pays for a couple of calls with no revenue.”

Now, although on the understandably optimistic side, this is likely an endpoint, rather than where we are at the moment. So we need something north of $2 and something less than $8. This is one I don’t think I should budge too much on. I created a dummy ad on Ingenio for the hotels category on a national level. The bids current range from $3.03 for 10th place to $3.78 for 1st. That ad is only shown in Orlando, but let’s take it for our purposes as being the national average (free 411 services only show the first ad).

So $3.78 * 70% = $2.67 net to Jingle. Travel is a category well suited to 411 because it’s not too high (mortgage) and not too low value (restaurants) that it can be taken with as an approximate average.

Now the biggest hole Lyn poked in my analysis was the white pages versus yellow pages lookups. I said 50-50. She said “According to Pierz Group, residential lookups are 20% of the total; among our callers it is around 12-15%.” I am probably (err definitely) dead wrong on this. Let’s dial the assumption down to 15%.

So the final revenue figure becomes 500,000 calls * 85% yellow pages * 7% connect rate * $2.67 net revenue = $79k

So the second time around produces a break even analysis. Clearly Jingle isn’t break even at the moment, but the point I guess is that they could very easily be.

The drivers are clearly the connect rate and the revenue, which depend on advertiser adoption and quality/relevancy of the ads. Those two critical factors are in the hands of a partner at the moment, so perhaps we’ll see Jingle start selling directly to try and control more of their destiny in the near future (again baseless guess using common sense).

Either way, thank you so much Lyn for posting such a detailed response to my post and rickety analysis, and I’d be happy to continue writing about free 411 if you’d like to keep in touch with how Jingle progresses over time (niki dot scevak at gmail dot com).

Greg Sterling has an excellent post on the free 411 industry. Greg summarizes a recent sponsored comScore study about free 411 as well as pointing out the biggest problem: burn rate. And also interestingly he notes that “inFreeDa (1800-411-Metro) has discontinued service”.

The economics of free 411 are not for the faint of heart.

It costs around 18 cents to service a call at the moment. That’s pretty good considering the $1-2 that the phone companies get from consumers but advertising is a little harder to make work.

For instance, white pages queries (name lookups) cost just as much as yellow page queries (more commercial in nature) but make up roughly 50-60% of queries (have that on pretty decent authority).

Let’s be generous and say that half are yellow page type queries. Of those around 10-15% are for restaurant type queries which will be extremely hard to monetize. That 15% assumption is based upon category data from the YP association [warning PDF].

Greg says Jingle is doing around 500,000 daily calls. So 85% of 250,000 is 212,500.

The ‘click through rate’ for the pay per call advertising is 15% at the high end, but 5-10% is more common. Let’s assume 7%.

So that means that 15,000 call conversions. Let’s say the average price per call is around $2 net. That’s $30k/day in revenue.

For the costs, (500,000 calls * $0.18), it’s $90k/day.

The pay per call metric is obviously the most rubbery. Ingenio is providing the ads at the moment and are likely getting around 30-50% of the revenue. So the advertiser would be paying $3-4 per call.

That’s a blind assumption. If advertisers were paying, on adverage, $12 a call, then the business would be at break even, according to the assumptions above. But if it follows the above assumptions, they’d be losing $60k/day.

No wonder VCs have kicked in $30m. This to me seems another YouTube type business, and one where a VCs can really help because of the significant cash drain in the startup phase. Over time it could be a great business but as Greg notes, history is littered with failed 411 startup attempts.

If you’d prefer a cogent analysis of the rise of Myspace over the Page 6 version of events, I’d highly recommend you read Nisan Gavvay of Startup Review’s timely piece on how MySpace became popular.

In short it wasn’t music, it was the fact that the servers stayed up when Friendsters were down. They didn’t show friend chains or let people from the Phillipines sign up because it taxed the servers too much.

They tried to spam their parent companies email lists but it didn’t really work. But it gave them a start and the site spread virally from there.

Also, interestingly the ‘overnight’ success didn’t really start gaining traction until 6-9 months after launching.

Either way, go and read the piece. Certainly poses interesting questions on the endevors of Blue Lithium and Oversee, who are now launching consumer facing startups on the back of similar direct marketing prowess.