No worries, playing devil's advocate is a great exercise because it forces all of us to think and go over interesting and useful strategies. Sometimes it's downright necessary because things are not always black or white – often it depends. For instance, in some very niche industries, you might introduce broad match a lot sooner because of the lack of search volume and data. So broad match keywords can help you discover what people are searching for related to your product or service.

I double checked what Google says about quality score as I had not heard that QS was only calculated on an exact match search of the keywords in a campaign. Here's what Google says on their site: “A Quality Score is calculated every time your keyword matches a search query — that is, every time your keyword has the potential to trigger an ad.” So even the performance of irrelevant queries triggered by broad match (and phrase) get included in the QS calculation.

I was able to find that for first page bid estimates however, “the estimate approximates the cost-per-click (CPC) bid needed for your ad to reach the first page of Google search results when a search query exactly matches your keyword. The estimate is based on the Quality Score and current advertiser competition for that keyword.”

So it appears that for first page bid estimates, Google only takes into account the quality score of the exact match of the keywords in a campaign. Good to know.

I'm just playing devil's advocate and tend to agree you both: that it's better to start very focused and have all match types bidding at the same level until data is collected to prove otherwise. I think there are many possible routes and methods to achieve paid search success, and as long as they are valid and logical, I respect all of them.

The one main reason why you'll want to avoid bidding lower on broad match initially is because of quality score. Since the floodgates are usually open on broad match keywords, it's likely you'll get more impressions. So if your ad is in a lower position because of a lower bid, your CTR is more likely to suffer and you may see your QS fall quickly.

I know that Google claims to normalize the CTR by ad position so in theory ads in a lower position are not penalized as much for having a lower CTR, but I prefer not to take chances by leaving this in the hands of Google.

Also you'll want to find out sooner rather than later if certain broad match keywords are worth keeping or not. This is more difficult to do if you lower bids to reduce the amount of clicks. You'll have a harder time telling why CTR is lower on broad match vs the other match types. Is it because there are a lot of irrelevant queries that Google is showing your ads for or is it simply because you've chosen to reduce your bids and your ads might be in too low a position?

I would tend to start all match types on the same bids, like Xurxo, but my point is that if you know broad match will generally perform worse that exact and phrase, why not use this experience and start their bids a bit lower? If it's common knowledge that broad match opens the floodgates and throws all sorts of traffic at you, why risk starting broad match at the same bids as exact and phrase?

I agree that some broad match keywords will perform very well, and agree that testing is key, but isn't it better to start low then increase bids on those keywords that are working well, rather than starting high and reducing bids on those that aren't?

Of course, data by match type might be skewed, but surely a more cautious approach with broad match is better than ignoring broad match completely at the start?

I have to agree with Christian that bid tactics are better made based on ROI rather than match type because there is no guarantee that an exact or phrase match keyword will outperform a broad match or vise versa without testing.

Bid stacking might skew the data and lead to the wrong conclusions based on match types. For this reason I recommend setting the bid at the same level initially and then adjusting based on actual performance.

It's a pleasure to share. I look forward sharing more and learning even more. Writing good ads could be a good subject to touch on.

I have subscribed to the bog and will participate when I can.

Chris

PS – Thanks for the reply. I'm very excited to talking in person.

This is an interesting twist on the strategy – where you introduce phrase match only after you've put in exact match to jimmy up the quality score a little more. Increasing the bids initially is also something we do to stimulate CTR early on to in turn help boost the quality score so I'd recommend that too – good point.

For brevity's sake I didn't touch on writing killer text ads, but you're absolutely right, without them a campaign is sitting dead in the water with lots of potential through proper keyword selection and match types, but no action because of poorly written ads.

Thanks for taking the time to share your insights, I look forward to future comments of yours!

P.S. I got your email and we too are looking forward to meeting you.