9 Things You Might Not Know about DKI

At its core, Dynamic Keyword Insertion (DKI – basics here) is an efficiency tool. As opposed to sitting around all day doing manual keyword stuffing into your ads, this little tool will do it for you automatically! Plus, it’s a great source of buzzwords – for instance, “We used the DKI code to dynamically increase ad relevancy to users in order to improve our CTR.”

Because DKI is such a big part of PPC, I think it’s important to explore its lesser-known ins and outs. Hopefully by reading, you’ll find a few things you didn’t know about DKI

1. It works anywhere in your ad.

In the display URL: DKI always works in the display url. Google separates tokens with plus signs.

In the destination URL: I’m not sure why you’d use DKI instead of ValueTrack parameters, but DKI works just fine in the destination URL.

2. It doesn’t work outside of your ad, such as with sitelinks.

Unfortunately, there is no way to increase relevancy of your of sitelinks with DKI.

3. It comes in 6 uppercase/lowercase flavors!

Type: “KEYWord” How it looks: First token uppercase, trailing tokens proper case.

Type: “Keyword” How it looks: First token proper case, trailing tokens lowercase.

Type: “keyword” How it looks: All tokens lowercase.

Type: “KeyWord” How it looks: All tokens propercase.

Type: “KeyWORD” How it looks: Last token uppercase, all previous tokens proper case.

Type: “KEYWORD” How it looks: All tokens Uppercase.*

(*Note: this is the casing you’ve requested, not necessarily the casing Google will give you. In this example, I requested all uppercase….Google, however, output proper case. The same will occur if you request nonexistent DKI variations such as keyWORD, keyWord, or KEYword)

4. It works swell with modified broad match.

Google is smart enough not to also insert your plus signs when using modified broad match.

5. It’s a privilege, not a right. 

If your quality score is too low, Google won’t give you the benefit of DKI, especially on

6. It was the original extended headline option.

Back in the day (way back when – I’m talking six-plus months ago in our fast-moving world), the only way to get an extended headline was with DKI. If the query matches your exact match keyword, Google will give you up to 30 characters in your headline. (To save you the trouble of counting, the image below has 27 characters in the original headline.)

7. You can use it for repetition.

Google doesn’t allow repetition of words unless you request an exception. With DKI, however, no exception request is needed!

8. You can use it in GDN.

DKI works in the GDN. The manifestations are different; in GDN, Google decides which keyword in your ad group to insert for you based on the content of the placement. Google may also decide to show your default text.

9. It doesn’t help your Quality Score.

I’ve read a lot of DKI-related articles that claim it helps your QS. The claim is, DKI tends to improve CTR, so the logical conclusion is that DKI improves QS. My response is that it’s an unfair claim, because if you’re using DKI, your ad group probably has several different keywords that don’t have much to do with each other. So, that means your static text is probably much less relevant to a given keyword in that ad group. If you serve a static ad that’s keyword-stuffed against a DKI ad, however, DKI loses its edge. So, if anything, the claim should be, ‘relevancy improves CTR, which improves Quality Score,’ not ‘DKI improves Quality Score.’

Additionally, Google may in fact knock you for using DKI! Much in the way Google normalizes CTR by position, Google may expect better CTR of an ad that uses DKI vs. one that doesn’t.  Currently I’m testing this hypothesis by having 2 ads duke it out; they’re exactly the same except that one uses DKI and the other doesn’t (there is one exact match keyword in the ad group, and the keyword matches the default text). So, if over time the avg. position for the DKI ad is worse and/or the CPC higher, that’ll indicate a lower QS relationship between the keyword and DKI ad. Stay tuned.


Going back to my original statement, DKI is ultimately an efficiency tool, not an optimization tool. If you’re hard-pressed for time, dealing with long-tail keywords, or simply have better things to do, DKI is a great option. If you really want to have the best PPC account structure (i.e. small, well-targeted ad groups), however, DKI isn’t going provide any benefit over static ads. If you add in the fact that with DKI you lose some control over your messaging (in GDN, or if your QS is too low, or if you have unique keyword casing), I would say DKI doesn’t help improve your AdWords numbers compared to old-fashioned static ads. So I’ll end with some wisdom that you certainly did know if you’ve read much about DKI: proceed with caution!


Mike Nelson
Mike Nelson is a senior account manager at PPC Associates, an SEM agency with offices in the Bay Area and downtown Chicago.

Comments are closed.

One thought on “9 Things You Might Not Know about DKI