Advertising with Apartments.com CTR discrepancy...scam?
January 24, 2012 1:56 PM Subscribe
I took out an advertising package with Apartments.com. This is no cheap proposition, at $2,000 for 500,000 impressions. I wasn't expecting much out of this, but was hoping for at least the 0.05% the apartments.com rep claimed was their average, considering that our company is in the moving industry and has some impressive accolades under our belt all mentioned in our ad.
50,000 impressions in and my apartments.com contact claims there've been 18 clicks (0.03% CTR). However I only see 6 originating from apartments.com, 2 of which were just me testing out the ad.
They say they don't know the reason for the discrepancy. However, they also don't seem to know much of anything. For example, they said all GA referrals would reportedly be from apartments.com, however, the 6 I did find were actually from ad.doubleclick.com. I had to drill down to see they were from a apts.com directory therein to confirm.
Any ideas on solving this discrepancy? I don't see any other referral entries in google analytics that look like they could be from apartments.com. Any chance the reps are being dishonest on the success of this campaign, or even how many impressions they're actually posting? How in the world could one actually verify an advertising company is truly delivering on what they promised?
Feel free to give me any and all advice and thoughts you think of, relevant or not. I'm always interested to hear what the users on metafilter have to say. A very intelligent bunch, and very generous with your knowledge!
thanks.
They say they don't know the reason for the discrepancy. However, they also don't seem to know much of anything. For example, they said all GA referrals would reportedly be from apartments.com, however, the 6 I did find were actually from ad.doubleclick.com. I had to drill down to see they were from a apts.com directory therein to confirm.
Any ideas on solving this discrepancy? I don't see any other referral entries in google analytics that look like they could be from apartments.com. Any chance the reps are being dishonest on the success of this campaign, or even how many impressions they're actually posting? How in the world could one actually verify an advertising company is truly delivering on what they promised?
Feel free to give me any and all advice and thoughts you think of, relevant or not. I'm always interested to hear what the users on metafilter have to say. A very intelligent bunch, and very generous with your knowledge!
thanks.
Did they not allow you to include campaign-specific tracking code? Something like this. Then you could verify whether they had shown the ads on other sites.
I'd concur, in any case, that it was a very expensive ad buy. Too few impressions for too much money. For example, based on what I've spent over the last three years on Facebook ads, you could have had more than a million impressions for the same money on Facebook and, if your targeting was good, much better results.
posted by Mo Nickels at 2:48 PM on January 24, 2012
I'd concur, in any case, that it was a very expensive ad buy. Too few impressions for too much money. For example, based on what I've spent over the last three years on Facebook ads, you could have had more than a million impressions for the same money on Facebook and, if your targeting was good, much better results.
posted by Mo Nickels at 2:48 PM on January 24, 2012
If apartments.com is just reselling doubleclick ads, maybe it'd be better just to go straight to the source next time.
posted by rhizome at 3:36 PM on January 24, 2012
posted by rhizome at 3:36 PM on January 24, 2012
This is a common question I get from clients, as I work on the analytics side. Here's a good google doc I send them to as a starter for this question.
What it really comes down to is that click is not a visit and ad servers/web analytics packages are using different methodologies so its hard to compare apples to apples. Discrepancies arrise because of these two reasons. Also, keep in mind that an analytics package can't always correctly identify the source of the traffic on its own which will cause further issues. Since the click numbers you're dealing with are so small, its hard to see if this is because of a real problem or if its within the standard range of error.
To me, the bigger issue is that you're using branding tactics for direct response goals. CPM buys (of which the $4 CPM you're getting here isn't that expensive, but is more expensive than a DSP can get you and less expensive than a pre-roll on Hulu) are typically used to get impressions in front of an audience, not necessarily to induce a direct response. If clicks/conversions are your goal you should be utilizing a CPC/CPA buy.
Like nerdcore mentions, you need to be:
1 - tracking impressions/clicks yourself (via a 3rd party ad server like doubleclick or mediamind)
2 - applying conversion tracking (whether via campaign codes or conversion pixels)
3 - calculating CPA and figuring out what sites/placements/tactics are above/below your ideal acquisition cost.
Memail me if you need any more help QA-ing or just want to talk more about this.
posted by jourman2 at 5:46 PM on January 24, 2012 [1 favorite]
What it really comes down to is that click is not a visit and ad servers/web analytics packages are using different methodologies so its hard to compare apples to apples. Discrepancies arrise because of these two reasons. Also, keep in mind that an analytics package can't always correctly identify the source of the traffic on its own which will cause further issues. Since the click numbers you're dealing with are so small, its hard to see if this is because of a real problem or if its within the standard range of error.
To me, the bigger issue is that you're using branding tactics for direct response goals. CPM buys (of which the $4 CPM you're getting here isn't that expensive, but is more expensive than a DSP can get you and less expensive than a pre-roll on Hulu) are typically used to get impressions in front of an audience, not necessarily to induce a direct response. If clicks/conversions are your goal you should be utilizing a CPC/CPA buy.
Like nerdcore mentions, you need to be:
1 - tracking impressions/clicks yourself (via a 3rd party ad server like doubleclick or mediamind)
2 - applying conversion tracking (whether via campaign codes or conversion pixels)
3 - calculating CPA and figuring out what sites/placements/tactics are above/below your ideal acquisition cost.
Memail me if you need any more help QA-ing or just want to talk more about this.
posted by jourman2 at 5:46 PM on January 24, 2012 [1 favorite]
If apartments.com is just reselling doubleclick ads, maybe it'd be better just to go straight to the source next time.
If it's coming from doubleclick, then I think they're just using Google's DFP Small Business, the free, hosted version that they've introduced since they bought doubleclick.
posted by stavrosthewonderchicken at 12:16 AM on January 27, 2012 [1 favorite]
If it's coming from doubleclick, then I think they're just using Google's DFP Small Business, the free, hosted version that they've introduced since they bought doubleclick.
posted by stavrosthewonderchicken at 12:16 AM on January 27, 2012 [1 favorite]
Response by poster: Thanks for the responses everyone. My apartments.com contact apparently saw my post on metafilter and is now being a bit more helpful in reconciling the discrepancy. She posits it's because many visitors aren't spending enough time on my site to even register. Seems feasible, but would indicate the vast majority of traffic I'm receiving, both registered and unregistered, is just accidental clicks. Amazing the extent to which this impression advertising doesn't seem to work these days. I really thought a moving company like ours would do well with apartments.com, but guess not.
Anyway, going forward I'll reroute this advertising money into SEO. Might experiment with facebook ads too, though I'm not too optimstic.
posted by grammalvsu at 3:01 PM on February 3, 2012
Anyway, going forward I'll reroute this advertising money into SEO. Might experiment with facebook ads too, though I'm not too optimstic.
posted by grammalvsu at 3:01 PM on February 3, 2012
This thread is closed to new comments.
From a business perspective purchasing the advertising is merely one step in a larger chain of events. A ton of business owners assume that just paying for ad space is going to get them results. Hopefully you've already figured out a lot of this stuff and I'm just being redundant, but I just want to cover some bases:
- Do you have $2,000 to spend on this type of advertising?
- Have you already spent an equal or greater amount on SEO, and optimizing your website to convert the leads you get into buyers?
- What other avenues are you using to generate leads?
- How successful are they?
- How much do they cost?
- How old are most of the people who rent from you?
- Do they fit in the target audience of searchers on apartments.com?
- How much is one "sale" worth to you?
- How many sales do you need to make in order to pay your advertising costs?
Before you buy advertising, you need to spend the time to really understand your audience so that you know for sure whether you're allocating your resources in the best way. You also need to make sure that the message is moving people to click on your ad. If your ads aren't converting:
- Can you A | B test your messaging?
As an aside, this is a small sample, but recently a mefite was looking for alternatives to craigslist for apartment hunting and only 2 folks suggested apartments.com.
posted by nerdcore at 2:17 PM on January 24, 2012 [1 favorite]