Keyword Portfolio Modelling
A number of PPC agencies use a variation of the Portfolio Theory model to optimise paid search campaigns but Branded3 are the first agency in the UK to adopt a similar approach to natural search.
Using the same fundamental principles that Wall Street & City traders use to predict risk vs. return across multi billion pound funds we can predict the return on investment for your keywords before we even start working on a campaign.
Understanding the mathematical theory behind Portfolio Theory is complex, and yet the basic principle is quite straightforward. We draw up a list of target keywords and work out the expected ROI for each based on the following variables:
- Total monthly search volume for the keyword
- Estimated traffic for a particular ranking
- Estimated cost to achieve that positioning
- Conversion rate for each keyword (based on current PPC and SEO data)
Our bespoke software plugs in to clients Google Analytics accounts and conducts continual analysis to find and discover new keyword opportunities allowing our SEO team to spend time on execution rather than time consuming research and analysis. We check daily rankings across over 200,000 unique keywords and match these rankings with keyword search volume data to model current and potential traffic to a very strong degree of accuracy – in tests we can predict traffic across tens of thousands of keywords to within 5-10% accuracy almost every time.
The software can check rankings and potential traffic/sales across every single keyword that has driven traffic to your site via SEO or PPC over the past few years – we regularly analyse keyword portfolios comprising over 10m keywords.
These calculations are carried out initially during the planning stage to determine the estimated cost and ROI for each keyword, allowing us to maximise the efficiency of the campaign by actively targeting the most profitable keywords and keyword groups first.
Keyword level reports
We predict the traffic, sales, cost and ROI at every top 10 ranking position for each keyword in your portfolio.
Subset level reports
The same as keyword level reports, but for a subset of keywords. For example, an office supplier might want to estimate ROI across a campaign involving all keywords related to office chairs.
Overall reports
The overall report can help predict the ‘sweet spot’ for your campaign and help decide the overall strategy for maximising ROI across all keywords.
Continual monitoring
We also run the algorithm continually throughout the campaign. This allows us to highlight new opportunities within hours, or even minutes. Our systems continually monitor rankings, click-through data, conversion rates, search results and keyword search volumes for thousands of keywords every hour to ensure that we are armed with as much information as possible about your portfolio.
The biggest complication in predicting keyword ROI is the ever-changing search landscape. Advertising campaigns and social media can cause certain keywords to spike dramatically within minutes. Click-through rates on a 3rd place ranking can drop by 30% if Google chooses to add some news or video results.
Even conversion rates vary dramatically as clients change prices and add special offers to increase sales.
