• Home
• PPC
• Interactive Metrics Tree

Interactive metrics tree

The Jellyfish metrics tree allows our PPC managers to clearly understand which metrics make up the campaign CPA and how they relate to one other:

 

---

In practice, the metrics tree can help to optimise your campaigns in a number of ways:

• Forecasting – using known and estimated data, create various performance scenarios to guide the development of your PPC campaign
• Insight – adjust metrics based on historical data to clarify ‘what if’ outcomes
• Reporting – deliver a simple, insightful and actionable view of complex PPC campaign data

Let’s focus on just one of these activities – providing campaign insight - and walk through the metrics tree in practice. Using sample data, we have the following key metrics to apply to the tree:

 

TIP – when you enter the data, hit tab or click on the next field to update the tool.

Based on these metrics, we are currently delivering 79 acquisitions at a £84.62 CPA. Let’s now assume that my campaign target is to significantly increase our conversion volume, while maintaining current CPA. In other words, more sales, for the same spend. What are my options?

Our first action is to raise bid caps across the campaign, to deliver the following effects:

More impressions

• Keywords that were on the second page of listings may now be on the first page
• Keywords that were inactive may be reactivated and therefore generate impressions
• Keywords on broad match have an increased quality score and therefore the reach of the broad match is extended

CTR increases

• Increased bid caps pushed ads up the listings making them more conspicuous

CPC increases

• Increased bid caps allow the networks to charge more per click

Conversion drops

• The higher the listings, the more click-happy searchers become (as a rule), increasing the likelihood of generating unqualified and non-converting traffic.

To see this in action, click on the small right hand arrow:

 

 

By increasing the bid cap to hit a £2.77 CPC, I have increased volume of sales to 95:

 

 

However, CPA has now increased to £103.61 – too much! Our next action is to review the keyword portfolio by adding in negatives. Possible effects:

Impressions drop

• All search phrases containing the negative search terms will no longer trigger our ads thus reducing the overall number of impressions.

CTR increases

• A natural symptom of not showing your ads for irrelevant search terms. The ads that do show are more relevant and the ratio of searchers who will be interested and therefore will click through will be higher.

CPC drops

• The higher the CTR the higher the quality score. The higher the quality score the less the networks will charge per click.

Conversion increases

• A natural symptom of not showing your ads for irrelevant search terms is again the fact that people who do actually land on the site are more likely to convert.

To model this, click on the small right hand arrow:

 

By increasing negative keywords, I have decreased CPA to £96.23 (still too much, but heading in the right direction), and my volume of sales is 98 – looking good!

My next action is to use tail keywords as seeds to generate further tail keywords. Possible effects:

Impressions increase

• Even though in theory broad match should capture all search terms which contain the search terms included in a campaign, it does depend on the quality score how broad the matching will extend. Therefore by identifying the broader terms and submitting them to the networks, the net is cast out even further, thus increasing the number of search terms that will trigger ads.

CTR increases

• The further down the tail the keywords are, the more relevant we can make the ad copy and therefore achieve a higher CTR.

CPC drops

• The further down the tail the keywords are, the less competitive and expensive they become.
• The higher the CTR the higher the quality score. The higher the quality score the less the networks will charge per click.

Conversion increases

• The further down the tail the keywords are, the more relevant you can make the landing URLs. The more relevant the landing URLs are, the more likely a visitor is going to convert.

To model this, click on the small right hand arrow:

 

The CPA has been brought back into line…..in fact, it has been reduced, and volume of acquisitions is increasing (NOTE don’t forget that our original metrics were 79 acquisitions at a £84.62 CPA).

 

Great news, we can now start to raise bid caps again, because we still want more volume! By increasing bid cap to £2.86, our volume has increased again, but CPA has jumped up:

 

Now we can focus on conversion rates, and specifically on landing page performance. Let’s see what an improvement in conversion rate achieves. We’re assuming that we can test some new landing page copy, so the metric to adjust is landing page relevancy. Click on the small right hand arrow:

 

 

By adjusting the conversion rate to 3.11%, I can achieve my target CPA:

 

 

And my sales have more than doubled to 165.

Is it feasible to increase your conversion rate? In most cases, yes; we generally find that a structured landing page testing and conversion rate optimisation plan can deliver very significant gains in campaign efficiency.

Summary

As you can imagine, there are numerous ways that you can use the metrics tree to view and review the core metrics that make up overall campaign CPAs. In general, bear in mind that viewing metrics in isolation is a mistake; as you can see from this walk through, adjusting specifics usually impacts on other areas of the campaign. That’s why it is important to focus on the most important goals and KPIs, usually your CPA target. Feel free to experiment, or contact us if you’d like any further information.