According to Astronomer Royal Martin Rees just six numbers are needed to finely tune our universe. Search Engine Marketing requires more!

In the early days of paid search, campaigns were judged predominantly on traffic metrics, impressions, clicks, CTR, CPC etc. As Paid Search evolved and budgets expanded, it was vital that search marketers and advertisers developed strategies. Consequently keyword bids became reliant on whether they converted or not, the problem was that recognition for orders/actions was given to the keywords with the last click (mainly brand terms). Currently it is viewed that the cost per order model is dated as it only factors in last click orders. The next step for paid search is to move towards the contibution per keyword model, which offers greater insight into every keywords contribution to the company’s bottom line.

Nowadays, it is possible to follow the search ‘path to purchase’ comprehensively and accurately, due to technological developments. This enables credit to be shared out to all keywords being queried before a purchase. This is crucial as some keywords are better at generating assists than last click orders, such as the generic keyword ‘catalogue’, which will generate more assists than direct orders. Technology is also helping make these optimisations automated, however, the technology must be flexible, since every advertiser has different expectations and must be finely tuned in numerous techniques.

Paid search is currently moving into further complex methods of optimising accounts in order to get the most profit out of each specific keyword. Every advertiser has particular target audience’s and product margins to adhere too, which need to be taken into consideration at a keyword level. Factors to be considered before buying a keyword and the level of the bid include: Day of Week, Time of Day, Geographical location, Match Type, Order History, Assist History, Creative and most importantly Profit.

Brands can have different awareness in different geographical locations, which means different strategies and optimisations for the varying reponses. Day of the week and time of day are important as some keywords can be used for researching and buying products for differing time periods. Some keywords drive sales through awareness, so assists play a vital role in an account’s overall performance and should therefore be considered in the algorithm. Moreover the creative with the best conversion rate for a specific keyword can be part of the optimisation as well.

The challenge for automated bid management tools is to bring all these factors and more into the algorithm, but be flexible and versatile when it comes to modifications to new strategies.  The aim of campaigns can change, and often contain multiple purposes, which may shift based on internal company politics. Figuratively there needs to be dials connected to a flexible algorithm that can be moved based on strategy. Additionally the algorithm needs to have room for predictive changes. For example it needs to be ready for surges in traffic and orders due to seasonality, weather or offline awareness.