Sadly, not every keyword you attempt to target in your AdWords accounts will work. The good news is, because PPC offers a fountain of immediate feedback in the way of data points, it’s possible to identify poor choices in your initial targeting and take swift action.
In this post, we’ll walk through the two major examples of keywords you’d be better off pausing, as well as some potential pitfalls to consider in thinking about actually shutting down these keywords.
Keywords that Spend Money but Don’t Convert
This is probably the most immediately obvious and intuitive example of a keyword that should be paused: keywords that are costing you money but not converting into a lead or a sale. But even within this classification of keyword, there’s a bit more nuance than meets the eye.
Let’s imagine we sell paid search software and every lead is worth $100 to us based on both the number of those leads that turn into sales, and our company’s margins. Obviously once a keyword spends something like a thousand dollars it seems pretty clear that keyword should be paused, but what if:
- The keyword is sending visitors to a poor landing page?
- The keyword could convert better with a different (lower) bid?
- The keyword has poor, untested ad text associated with it?
- The keyword has spent and not converted in the last 30 days, but generated a lot of business for us in the past at an acceptable cost?
- The keyword is an informational term that drives a lot of later stage conversions?
Any of the above could mean that you’re leaving money on the table in pausing a keyword, so before you decide that a keyword is something that spends but doesn’t convert, you want to look at:
- Different Date Ranges – Even though the term hasn’t performed in a 7- or 30-day window, it might have done well previously. If this is the case: what changed (bids, ad text, landing pages, something in your market/niche?)?
- Landing Page Effectiveness – How effective is the associated landing page overall? This is something you can drill into on the AdWords dimension tab , and what you want to confirm is that the landing page this keyword drives to performs well when paired with other keywords.
- Ad Text Efficacy – The same holds for ad copy – does the associated ad copy convert well from a CPA standpoint overall? This may well be a reason for a recent drop in effectiveness – maybe the ad copy you initially wrote has become stale and a new test would jolt your ad performance.
- Assists – Google’s new multi-channel funnel reports  allow you to drill down to see not only last click attribution but also whether a term is driving a lot of conversions by acting as an earlier touch in the conversion process, so you can avoid cutting off a term that is helping to drive conversions that are being attributed to other keywords or channels.
As you can see there are a lot of moving pieces to analyzing all of the pertinent data points before deciding to “kill” a keyword. This is a lot of work and analysis to perform if you’re looking keyword-by-keyword in even a mid-sized PPC account, so it’s probably more practical to either restrict this type of deep dive to higher-volume terms you’re looking to pause, or to apply automation to the process .
Keywords that Don’t Spend Much, Don’t Convert, and Have Low Quality Scores
“Hard costs” in a PPC campaign aren’t the only costs – keywords that underperform on other metrics can also have “soft costs” in the way of negative account-wide Quality Score impact.
While it’s true that low Quality Scores don’t mean you can’t advertise profitably on a keyword , there may be keywords that have low Quality Scores and click-through rates, don’t convert well, and slipped through the cracks of our analysis as we attempted to find good candidates for keywords that spend but don’t convert.
Extending the example above, let’s imagine we have a handful of keywords that are all driving very little spend but not converting – call it $10 per keyword over an extended period of time. In theory these keywords probably wouldn’t be “pause ready” as they haven’t yet demonstrated that they aren’t profitable (if conversions are worth $100 to us, they could spend five times the current amount for a conversion and be extremely profitable), but these keywords – particularly in aggregate – may be costing us money indirectly.
Among the factors in determining Quality Score is your historical, account-wide CTR and Quality Score, so having keywords in your account that aren’t helping drive your business directly (via conversions) and also hurt your account-wide CTR has a very real soft cost in that it drives down Quality Scores and subsequently drives up costs and makes it expensive to push out new keywords and ad groups.
But, not unlike with keywords that spend but don’t convert, these terms may also be suffering from low Quality Scores due to factors you could potentially correct, such as poor ad copy and keyword segmentation.
If you’re not sure of whether you should be pausing keywords to help improve Quality Scores, one good approach is to run your account through the AdWords Grader  to see how your Quality Scores are in general (you’ll see this as part of your report), and assuming your Quality Scores need work account-wide you can start to home in on keywords that are in groups with otherwise reasonable Quality Scores.
Developing a System for Pausing AdWords Keywords
Ultimately, over time and as you’re managing a campaign, you’ll want to develop a process that takes all of the above into account and leads you to the right AdWords keywords  to pause (keywords that have been given a fair shot to perform but are still dragging down your pay-per-click ROI and your account-wide Quality Scores).
Alternatively, WordStream just released a new feature – Pause Keyword Alerts  – that runs a lot of these calculations and analyses for you, taking care of the heavy lifting and bubbling up paused keyword candidates on a weekly basis.