Lies, lies and Snooth: I investigate

There’s been a raging debate going on in the wine internet world.

Paul Mabray of vintank has made some serous accusations about Snooth after being obviously antagonized by website traffic claims in a post from Philip James of Snooth.

Paul vs Philip

The end result of a wide ranging debate has not seen any agreement between the two. Although one particular issue raised resulting in Philip apologizing to Cellar Tracker, albeit under the threat of legal action.

Ignoring the personalities and focusing on the substance of what they’re saying is very illuminating. So let’s get into that.

The Posts

Here are the posts from Philip James and Paul Mabray blogs. Note that the many of the comments are also worth reading and come from other personalities in the wine industry.

• Philip - Powering Wine Online
• Paul - The TRUTH About Snooth Data
• Philip - Digging into the Details
• Paul – Is Snooth scraping data from CellarTracker?
• Philip - An Apology to CellarTracker

The accusations can be split into these categories:

• Misrepresentation
• Hidden objectives and SEO
• Poor user experience
• So-called Success Stories
• Legal Breach

Misrepresentation

The post that started this all was Snooth saying they were achieving extraordinary high website traffic figures making them the fifth largest content platform on the web.

Snooth used two independent research companies to make this point – comScore and Quantcast. Snooth syndicates data to more than 300 companies including Epicurious via an open API (which I’ll come back to). And given that their business model is all about driving impressions to boost advertising revenue this is an important statistic for them and their advertisers.

What I think really set Paul off was the sentence “We reach 10-12 million people per month“.

Paul points out that the Quantcast snooth report includes all generic traffic to myrecipes.com, epicurious.com, foodandwine.com, receipetips.com, various google sites etc.

He says that 90% of this traffic might go to the partner sites but not the wine sections of these sites and certainly not to Snooth (though acknowledges snooth is still very large with over a million visitors).

Indeed Paul does some investigation and concludes that it is hard to find Snooth’s widget, and, once found, is at the bottom of the page.

Philip disagrees that it’s not on most pages and uses Quantcast data to make his point.

Not being sure who was correct, I did a bit of research and …

I think both have some good points

Firstly I tried to find wines by just browsing around the recipes on myrecipes.com and couldn’t. Then I found the Wine Finder and sure enough there are snooth reviews.

Here’s the odd thing. I did the following internal site search using google.com. The searches were:

• site:myrecipes.com (88400 results)
• site:myrecipes snooth (35900 results)

This is a search commonly used to work out how many of a domain’s web pages has been indexed by Google. By adding the word “snooth” I hoped to be able to have a robust and quick way to work out who was correct.

At which point I thought Snooth was fine because 41% (35900/88400) of pages indexed included the word snooth.

Until I clicked on one of those snooth results and couldn’t find snooth or indeed any wines mentioned at all.

Now I’m wondering about Snooth, myrecipes, quantcast and even google! Perhaps google has cached / indexed old pages that use to have Snooth on them?

For the purposes of this post I can see a smoking gun but not absolute proof of wrongdoing.

What I’d want to see

• Snooth on the high traffic web pages of myrecipes
• Unique visitors per day and sources screenshot from Snooth’s Google Analytics reporting software

Hidden Objectives and SEO

The last part of the original Snooth post was a call by Philip for wineries to claim their data. Paul is a wine and internet technology expert (as is Philip) and has a real problem with how Snooth does this. He believes that this results in Snooth outperforming the winery on Google.

In essence the wineries “donate” (loaded word) their content to Snooth who uses this unique aggregated content to boost their Google ranking which generates traffic allowing them to charge higher advertising rates.

Note I believe Google puts the winery at the top of a simple query but if you add “buy” to the search phrase snooth can be at the top (as at Jan-11).

However Paul also points out that the way the API is constructed ensures that traffic is driven to Snooth not the winery.

Paul has a conflict of interest here (that he acknowledges). His client, cruvee.com, has an alternative system whose API links back to wineries. Now Paul is a winery champion whereas I’m more in the retailer camp so I’m not so worried about that, however I assume that the same content a wine retailer puts in there also goes back to snooth not the retailer.

And this is a fundamental disagreement

Paul thinks Snooth is pretending to help wineries whereas really it’s just helping Snooth. He thinks Snooth should be more transparent in what they are doing.

The counter point is that Facebook, Expedia, Yelp, and Trip Advisor do the same thing. It’s not dishonest it’s just they way content aggregators work. Paul counters this by saying they are all about engagement whereas Snooth is really just a comparison shopping engine (CSE). He says CSEs are bad for wineries because they can’t compete on price due to channel conflict.

I actually agree that Snooth is primarily a CSE but I also think Philip honestly thinks he’s created an aggregator – one with reviews, prices and ways to purchase. More of a Yelp than a shopping.com.

My thoughts

• I don’t think Philip’s being dishonest
• Snooth could be a little more transparent with their business objectives.

Poor User Experience

Vintank also claims Snooth offers a poor customer experience.

An example is given where Snooth only lists two of Bolen Family Estates many wines, and says no wines are held in stock at Bolen or at Californian merchants. As Bolen is a vintank client, Paul can confirm that this is not true. There are a number of similar examples given in the comments section by other people.

Philip acknowledges difficulties in wine data management

That he’s been working on it for years and it is a tricky area. It’s a reasonable, humble and upfront admission. But, well, it’s still an admission that they have product problems.

What I’d like to see

• A bug and feature roadmap. Be open about the issues, we all know that software is not perfect! Most software companies have moved to this. Magento is one that springs to mind (bugs and features), salesforce.com is another (performance issues).

So-called Success Stories and an Insight

Philip then finishes with some examples of companies that have been very successful: Zinfadel Advertisers and Producers, and Massican. Both obviously were very successful snooth based campaigns.

However Paul points out that the success was due to emailing the snooth list not from being snooth website listing per se.

The accusation here is that snooth takes winery data, gets the SEO benefit, boosts traffic, which increases the customer email list. And that the email list is where the real power of snooth is.

Insight

It keeps coming back to the old direct marketing principle of the list, list and list being the three most important things of DM. Just the way to build that list has changed.

What I’d like to see

• A success story of a winery doing all the right things with its listing and being able to directly attribute significant sales to snooth.com not from snooth emails.

A Legal Breach

Lastly Snooth has been accused of taking (“scraping”) data from Cellar Tracker.

One of the comments in the vintank post suggested this so Paul looked into it. And sure enough the user tags were been driven by Cellar Tracker’s reviews.

Philip immediately looked into it, found the accusation to be correct, apologized and explained that it was a coding error, and has started to delete all this data.

Apparently Snooth had an agreement to take reviews from Cellar Tracker back in 2007. However this ended on Cellar Tracker’s request. According to Philip the software that was taking reviews was indeed cancelled. However Snooth’s user tags were continuing to use Cellar Tracker data.

Now I’ve made mistakes and I’m reluctant to get too judgmental but, Philip, you have really screwed up. As the CEO you can’t delegate that to being a just programmer error. The responsibility was yours.

My thoughts

• You’ve done all the right things by humbly apologizing.
• I think the loss of face is punishment enough
• But this has legal action written all over it. Good luck, though currently Eric at Cellar Tracker seems very reasonable in his approach.

In Summary

• Misrepresentation: Unsure but smells bad
• Hidden objectives and SEO: I don’t think Philip’s being dishonest. Snooth could be a little more transparent.
• Poor user experience: Be open about the issues, create a bug and feature roadmap.
• So-called Success Stories: show direct evidence of listing driving success not emails
• Legal Breach: good luck.

Disclosure: I personally do not know either nor have a business arrangement with vintank or Snooth, though I have used Snooth when representing merchants and used vintank reports for my own thinking. I am connected to them in LinkedIn and Twitter.