We often like to have all our bases covered… to tick all the boxes and know where all the dials are pointing.
But this brand of thoroughness isn’t always the wisest approach.
When it comes to analysis and optimisation,
- It pays to prioritise our best (and worst) performing traffic segments
- It pays to prioritise our time
- and likewise, depth over breadth is a useful mantra when considering which metrics and dimensions to concern ourselves with
So below are three of the big ones that keep drifting into our line of site…. and don’t need to be there.
1. Competition
This is often one of the first things clients ask about.
And for us PPC managers – there is a lot to gain from taking an initial interest in the competitive landscape – to see things like:
- What recurring themes show up in competitors’ ad text
- What service-related phrases are used on their sites?
- Which brands might we include if we try a ‘competitors’ campaign (rarely worthwhile as that is…)
But as an ongoing focus, competition has a lot less value than it’s often credited with.
Why you might want to look at it
To diagnose a change in CPC (and for this – use the new predefined Auction Insights report, rather than the very limited, traditional Auction Insights)
Why you don't need to look at it
Because what matters is the level of success in your own account, and the potential to improve it either by expansion or refinement.
The actionable insights into this all live within a few standard metrics (for performance: cost / conversion value and a few related metrics will do… and for potential: IS lost to rank and budget tell you what you need to know).
Any changes you see in Auction Insights are only (actionably) interesting as far as they actually impact those metrics.
In practice, it may be satisfying to investigate a metric change in Auction Insights and return with a plausible explanation… but you won’t be doing much differently as a result of that explanation, that you wouldn’t be led to do by the key metrics themselves.
2. Quality Score
Many a sleepless night has been fuelled by stubbornly low Quality Scores, and charts like this that purport to show disastrous CPCs at the lower end of QS.

(This one – still widely used – is taken from a correlation analysis in 2013, which doesn’t begin to separate correlation from causation.)
There’s no need for that lost sleep.
Why you might want to look at it
Quality Score is not really the factor governing ad rank or influencing CPCs… It’s a snaphot-window onto the more intricate relevance score which is recalculated for every auction, and actually does weigh into Ad Rank.
We won’t go into that here (see lesson 3.5 in the Google Ads Level Up course if you would like to take an in-depth look at)
However, it is the closest thing we have to that relevance score, so we’ll take it as a proxy – which is what it’s for, after all.
QS then does have an impact on both your CPCs and deliverability… and yes, this can come into play in a painful way at the lower extremes.
Also, as a (however flawed) measure of the relevance from search term > keyword > ad text > landing page, it’s also a rough guide to how well your whole setup is paving the user’s journey from search to purchase.
As Google themselves put it…

Why you don't need to look at it
- It’s clunky and unresponsive. Even making the ’right’ moves often won’t result in the kind of QS improvements you could reasonably expect to see (in all three of the sub-metrics). From this point of view, not paying it too much attention comes under the ‘wisdom to accept those things we can’t control’ category.
- It’s not always aligned with profitability. Ads often come in the lower-CTR but higher-CVR variety. Favour them if you want to optimise for return on ad spend… but not if you want to optimise for Quality Score
- Middling and even low-Quality Score keywords are often perfectly compatible with profitable activity… Focusing on those core, bottom-line metrics will serve you better than the tail-chasing that often results from a focus on QS and its components
3. CPC
OK, here’s a controversial one…
And I’ll admit it now, I don’t exactly recommend ignoring your CPCs (naughty clickbaiter!)
But I do spend a lot of time explaining why CPC should not be treated as a KPI.
Why you might want to look at it
When the whole PPC system is viewed in its component parts (you pay for clicks, clicks bring you conversions…) – how much you pay per click obviously matters.
It’s also fair to say that ‘at a constant conversion rate’ how much you pay per click obviously matters…
Why you don't need to look at it
…But that qualification – ‘at a constant conversion rate’ – is completely unrealistic.
If keyword A has a higher CPC than keyword B, but also a higher conversion rate – and the latter difference outweighs the former – we will favour keyword A every time.
And within each keyword, different search terms, audiences, demographic groups, devices etc are all subject to the same evaluation.
In practice, that trade-off happens frequently. Market forces produce more competitive auctions for more valuable searches / ad inventory.
(Take your brand terms out and check the CPC for all your converting keywords vs all your non-converters…)
Remember when we could easily see the user locations of clicks from outside our targeted countries (when the advanced location setting for only ‘users in’ our targeted locations hadn’t been set)?
Low-CPC, low converting clicks from the other side of the world, anyone?
Search Partner traffic is another straightforward example.
It typically comes at lower CPC; lower CVR; it’s sometimes worthwhile… but almost always has a higher CPA and lower ROAS than Google Search traffic.
To put this into a wider context, watch Rand Fishkin’s excellent explanation of the depressing ‘Ad Slurry’ principle… how ad platforms monetise lower-quality inventory with increasing aggregation and obfuscation (not naming any names… 🧐)
So in summary
- Any relevant impact of the competition is already baked into those metrics
- Quality Score is an unreliable guide to the genuine health and value of your setup. You want it to be high, but it’s rarely worth pursuing beyond those actions you would be taking anyway, based on your bottom-line metrics alone
- CPC ÷ CVR matters (it needs to come in under your value-per-conversion)… but then we call that CPA. CPC in isolation gives you nothing to work with for evaluating your clicks.