Google tv is easy, but bad for a startup. Converting brand new users from TV to the web when they don't know who you are is VERY hard.
I spent about $15K with my last company on google tv ads. It was a sports site and we got on espn etc. I poured over server logs and google analytics and didn't see any value from these ads - no spikes or noticeable conversions whatsoever.
CPM was super cheap so we got a TON of impressions.
People don't usually rush from their TV to their computer when they see an interesting URL, though, so you really don't know what effects the TV ad had on your site.
If they don't go straight to their comp then they usually won't remember - with a new brand that is.
And despite what our habits may be most people don't have a laptop open while watching TV.
It was not scientific, but we saw ZERO uptick in sign ups while running millions of TV impressions. I would never run TV ads trying to get traffic for an unknown web startup again.
Very true. TV advertising is more geared towards evoking recall at a later time, since for most products the purchase (or conversion) decision isn't immediate (after all, the viewer is in the middle of watching something). This works well for someone like Pepsi, since they want to be remembered the next time you're at the supermarket, but not so well for something that would demand an immediate response.
People watch TV with laptops in their laps and iphones in their pockets. I don't know if people write down the domain and visit it later. Just a thought. I wonder what the research suggests.
Mostly targeting by time and show (which is demographic by nature). There was some geographic stuff but it was pretty limited. You can create an account and browse without spending any money on it.
The 1,000 visits is pretty good at $1.3 CPC, but I bet it'd be a lot smaller for regular ads. This was the equivalent of a "click here' banner ad...you might get a click...but it's just people curious about what it is you have to sell.
Someone should do the same test for a regular ad, i.e. this is our product come buy it at oursite.com. I bet the number of hits will plummet.
Oh and I bet the production values matter a lot, their ad was actually good, anything a regular person comes up with, will probably be junk.(i.e. the equivalent of those dealership ads)
The cost of creating the ad should be included in the CPC calculation. Then, to limit it's impact on CPC, you have to run the ad a lot more so the production cost as a percentage of the CPC drops further. You need a lot of money for a decent TV campaign.
But I guess the point of the video was to show how easy it is to setup a TV campaign. In that regard it really seemed easy. It was surprising that it didn't require a large minimum to get air time.
If you are just targeting locally, say going into individual markets at a time. It can be surprisingly cheap to run a TV campaign.
You can go through the local network stations or the actual cable providers. Both will help you put together the ad (cheap if not free, quality depends on the idea, ie horrible car ads)
You can also pick your channels and general time frames that you want your ad to air in. So if you are targeting parents who own their home, you could select HGN, DIY, etc. and set the time frame to between 5 pm(afterwork) to 12 am(pre-sleep).
Same with radio in terms of dealing directly with the radio stations. You can write your copy, tell them you want one of the on air personalities to read it to some music. In a couple hours you can have a few different ads to choose from. It is surprisingly cheap to dominate any given radio station if you are in the right media market(not Boston or the likes).
When it comes to internet startups, the real problem with TV, Radio, or offline advertising in general is the bad conversion and lack of metrics like we are used to in the online world.
We are already used to single digit(or less) conversion rates for online advertising campaigns. You can imagine how much more that drops when your ads are reaching people who might not even be by a computer and are expecting them to remember your brand/product next time they are.
As for metrics, the easiest thing you can do is have control and variable zip codes. See how registration/purchases fair for zip codes outside of the market you are doing media buys in(but still relatively close geographically), compared to the rates for the area that you are advertising in.
The $100 mentioned is the minimum cost to get an ad on Google TV, not what Slate spent (though that doesn't mean the $1300 doesn't include the cost of the ad).
That's a good point. I didn't think the $1,300 could include the production cost of ad mostly because I figured creating the ad would be much grater. I can't begin to guess what it would cost to make the ad so maybe $1,300 did include production costs but I doubt it. I think they would make a point of mentioning that.
* 7 times during Glenn Beck episodes (starting at 2:49am)
* 54 times total on 4 networks
* 1.3MM people reached
* 1000+ visitors to ad website
Not sure how it would play out for conversion but cost of acquisition seems high, of course, the branding you get out of a network TV ad probably has its own benefits (especially if your core audience is 18-35 males awake at 3am. uhh.)
You're paying about $1.30 per visitor at that rate. This could work well demonstrating an actual product positioned to the paranoid crowd. I think that the ad might have done better if it was done a little differently. If there was an implied benefit of action, they might have seen visitor numbers closer to 1% of viewers.
A guy I know in the radio business said that, in the dotcom boom years, start-ups were paying 3x what they should have for airtime simply because they didn't know better. Google's platforms at least take away the fear that an ad salesman is quoting you 5x and telling you what a good deal you're getting.
Startups paid that much because they didn't buy lots of air time and they went directly to networks (paid "retail") instead of going through a media planning and buying agency. Most startups couldn't afford to go to major media planning and buying agencies because they mostly had very high monthly commitments and were owned by the major ad agency holding companies.
Nowadays with Google, it's much easier and much cheaper to get your ads on TV, but it's still much more expensive than going through media buying agencies.
Really now? I've always found negotiation works better, as long as you're willing to take some time to do it right.
As an example, AWS's one simple price structure for bandwidth is way above what you'll get from most other providers given some negotiation. Of course, your negotiations with other providers will start higher, but you can get em down substantially.
I don't claim that Google's TV ad buying system is optimal, but it's likely a better deal than the one offered-up to an ignorant buyer.
I see a parallel to the fixed-price used car lots. While a good negotiator can get a better deal elsewhere, an uninformed buyer can be somewhat assured that he's not getting ripped off.
I've always thought we should have a "best of the week" category for some stories.
If this isn't the best of the last week, it's easily in the top ten.
Lots of little details and nit-picking that should be done, but bottom line is that nationally-advertising on TV is not the impossible thing it used to be for young startups.
Another solution: An account, shared by multiple HN users that posts short comments on stuff that seems to merit "best of" consideration. Then you just check the comments feed for the account. Thoughts?
TV advertising is a completely different animal than the advertising Google is used to selling.
Ad "frequency" - eg how many times users see a given ad - is the most important metric for TV.
Without a large budget commitment it's extremely difficult to gauge the success or failure of a given campaign – even if you get a handful of visitors. That’s why you see brand advertisers continue to fill up the majority of TV spots.
Nearly everybody in this space has had a lot of trouble getting traction, internet advertisers don’t have the headspace for TV, and TV isn’t suited to small budget / small frequency campaigns.
Is this how those ads with the special urls get posted at the morning hours? my.dom/tv38 etc. I always wondered how these small sites/affiliate advertisers managed to afford commercials. I suppose they have decent conversion rates or I wouldn't see them every night. An interesting service overall though. I might toss a few hundred dollars at it one of these days.
I spent about $15K with my last company on google tv ads. It was a sports site and we got on espn etc. I poured over server logs and google analytics and didn't see any value from these ads - no spikes or noticeable conversions whatsoever.
CPM was super cheap so we got a TON of impressions.
We had a pretty decent ad too I think: http://www.youtube.com/watch?v=lUfTjtOQ8fM
It ended up being more of a vanity thing of having our ad on TV - total waste of time and money for us.