Google Content Match - Goldmine or Minefield?

Content match is a minefield that many publishers never survive.

A few basic thoughts before we get started:

  • If you are not tracking content match separately from search network, please don’t wonder why it is not working for you. yes it is more work, and yes it is worth it. Whoever told you internet marketing was easy was wrong.
  • Yahoo content match is for the brave or foolish. Just turn it off. Too much fraud, too much indifference from yahoo. Save yourself the trouble.
  • MSN is about to launch content, it is on by default. Watch out.
  • Content match can be a goldmine if treated properly. If you just turn it on and hope for the best, then you get what you deserve. Hope is a pretty poor marketing strategy.

    If you stop to think about it, Content and Search have almost nothing in common. Search is someone telling the engine that they want to learn more about a topic and you are paying the engine to feature you. Content on the other hand is someone aimlessly wandering the web, the article they are reading might not even be about your product. They might be interested in what you have to sell, but they were not pro actively seeking that information out.

    In some ways content is like radio advertsing, you broadcast an ad out to hundreds of thousands of people, and hope that you interest a few of them. Search is more like the phone book, the user has made the decision to look for information and you want to be there.

    So how to best monetize content…Lets start at the beginning.

    Since we are now tracking each keyword individually in search (we are tracking each keyword individually aren’t we?), do not do that in Content. Each adgroup should have anywhere from 1 to 20 keywords for content. The keywords should be tightly grouped around a central theme. Your highest volume search keywords, should probably have their own adgroup with just that keyword. I also tend to take my top 10 or so and make every combination of those keywords in 2 keyword groupings. (So keyword A plus Keyword B, Keyword A + Keyword C etc.)

    I then take major keywords and do lots of variations of them, I take a list of keywords and put them in front of or behind my main keyword. I add plurals, I add tenses. Anything that might get used by an author I want to have in the adgroup. So Buy, Own, Get, Service, Order etc plus the keyword all go in the same adgroup.

    I also usually do a group or groups for compeititors or brand names that are often seen together. For mortgage, I would make sure to have e-loan, lending tree, bankrate and other competitors keywords that might get mentioned in an article about mortgage in an adgroup. Also think a little outside the box, just because you might not want to bid on the keyword for paid search, does not mean you do not want that keyword for content.

    Lets stick with mortgage for now, you would almost certainly negative match the keyword “Foreclosure” from any mortgage campaign (if not, you should…), because someone who types foreclosure may not be a candidate for your products. But someone who is reading a news article about foreclosures, might not be in that circumstance themselves, instead they might be in the market to buy, or already own and worried about how this might affect future interest rates etc. Both those might be a candidate for a mortgage lead.

    Think outside the box for complementary products for the person you might be targeting. Selling baby products, bid on words like Ultrasound, baby names, baby shower ideas, maternity clothes etc. If you can figure out what else people you want to be your customers might be interested in, then you can put a targeted ad in their face at a time when they might be more interested in your product.

    Now that you have dozens of keyword groups. You need to write ad copy. Someone reading a search ad is totally different than someone seeing your content ad (tattoo that on your left hand, it is very important) so they will respond differently. In search telling them you have a low price, free shipping, and that you will pay them to take your product may be necessary, because that is what others are promising at the same time. In Content, because it is more of a serendipity purchase, you want to do more on the selling front. Tell them why your product is so great, give them a message about why they need the product. Content readers might not have even known your product was available, if you hit them with guaranteed lowest price, then they might think “hmm, i wonder who else sells this product, I bet I can get it cheaper”. You are probably the only one selling lime green widgets on that page, use that to your advantage.

    Similarly, your well optimized and tested website (that should be white…) that converts great for paid search might just need a little tweaking. Afterall, your visitors are different (have i said that before?) and content visitors probably want a little more information about the product, probably need to be sold a little more. They might want to bookmark your page or ask for emailed information. Leaks like these will lower paid search conversion rate, but will in fact improve content conversions.

    Content also allows you to control your distribution. If your product is not available in Canada, get rid of those sites that target them. Type in Canadian Blue Widgets, someone will be running an arbitrage campaign pointing to blue widgets and you will get clicks that cannot convert. Negative match that URL and save yourself money. What about other arbitrage? Some work, some dont. Just use your log files to see where you are getting traffic that is not converting (use CPM targetted to buy more from places that convert great…) and negative match those sites.

    Bids. I usually bid less for content than I do for search, but be aware of your position. By moving up a position in content you will get roughly double the impressions. The higher the position, the more impressions you get. Think about the pages that have just one ad, or only have a single 3 ad adsense block. If you are in 4th position, your ad will never show on that page. Bid aggressively and you will reap the rewards.

    Lastly, unlike in search content campaigns must be updated frequently. Every time I have ever had content figured out and was getting great clicks and great conversions and then stopped tweaking it, the content campaigns would just die a slow agonizing death. Nothing that was obvious until you looked at it over a months long period. Each time, just making minor changes, adding a period, changing a period to an exclamation, changing a word, anything to make it fresh and new would re-energize the account.

    This is by no means a definitive review of content, i can already think of 4-5 other hints/tips. This is just a starting point. Also unlike paid search, what works for one vertical or campaign does not necessarily work for another. Test test test.

    Tell others about this post:
    • StumbleUpon
    • del.icio.us
    • Digg
    • Sphinn
    • Facebook
    • Google
    • Reddit
    • Slashdot
    • TwitThis

    5 Responses to “Google Content Match - Goldmine or Minefield?”

    1. Dave Davis Says:

      Another great base.

      I agree about the Yahoo! content network. 10c minimum is crazy considering people are selling Yahoo accounts and the targeting is brutal.

      We like to run a short (few weeks) content network campaign, Trawl through the referrer data and see what sites are converting best, then site target those campaigns and play around with positioning. Then of course TEST TEST and TEST a little more.

    2. GeorgeB Says:

      How is MSN “about to” launch content match without a publisher network?

    3. Paul Benetis Says:

      loved this post
      thanks for sharing such good info

    4. Uber Affiliate Marketing Guide | UberAffiliate - We Hate 9-5 Jobs As Much As You Do Says:

      [...] Google Content - Goldmine or Minefield? [...]

    5. >Diorex snaps Diorex.com and relaunches his blog! Says:

      [...] Google Content Match - Goldmine or Minefield? [...]

    Leave a Reply