Net Profit Per Day

Bid Testing And Budgets – Optimizing Spend Without Waste

Add Your Comments

 

Synopsis – If you are involved in paid search, you are certainly familiar with the need to bid appropriately. To secure the position you want for the price you want to pay, you cannot just bid once and forget about it. It is crucial that you stay on top of your bids and constantly assess whether your strategy is working as you planned. Key to doing so is to implement some form of bid testing.

In this article, Rob Cooley discusses the different types of bid testing and how to find a balance between the need to test bids and the cost involved in doing so, which is dependent upon your goals as well as your budget. With some concrete advice on how to make the most of the bid testing process, Rob provides the small business owner with the knowledge needed to get organized and ready to implement such a program.

Bid Testing And Budgets — Optimizing Spend Without Waste

Bid testing has long been a sensitive subject in the world of paid-search marketing – a necessary evil in the most extreme of opinions. Yet the practice continues to live on in various forms and variations with little by way of standardization or best-practice sharing between marketing organizations, online advertising platforms, and bid management solutions. Obviously the competitive nature of keyword valuation makes the sharing of practices and information challenging, as companies vie for high-value keywords and all want to spend the least amount of money and get the highest return over their competitors.

Despite Cost, Bid Testing Is Necessary

Having an additional cost center in your marketing organization means increasing scrutiny on already-constrained budgets. However:

  • Normal bid maintenance is a necessity;
  • Ongoing bid maintenance is key to remaining competitive, as values shift with increasing speed while search marketing opportunities continue to expand;
  • Without ongoing bid maintenance through testing, businesses may find themselves overpaying for a word that has lost monetary value or being unknowingly outbid by their competitors.

It’s true, though, that even under the most stringent of bid testing programs, a business can lose some level of money or efficiency at least temporarily.

Options In Bid Testing

The choice of a bid testing methodology plays a huge factor in the types of positive or negative results you will see in creating a new program. It is important to assess what your company is trying to attain and what sort of risks are reasonably absorbed. If looking to quickly identify where a keyword sees increased conversions or value per click, then position-based bid testing is a simple, straightforward option to pursue. However, because position-based bid testing answers the question: “What do I have to pay for position X and what is the effect on conversion rate or value per click,” people who practice this method often experience dramatic increases in cost with little-to-no increase in revenue.

Countering position-based bid testing are current-bid–based and value-based bid testing methodologies, which offer lower risk to a business. The value-based model tests against the average observed value of a keyword rather than on arbitrary points, while current-bid-based testing focuses on small adjustments, up and down, in the bid of a keyword using the current bid as a baseline. While both are low-risk options, current-bid-based testing does not offer the same returns as value-based because its conservative approach of incremental changes usually results in financial improvements that are just as conservative.

Value-based testing balances aggressive and conservative testing by using a keyword’s average conversion value as a starting point for testing. In doing so, companies test for the word’s sweet spot by moving, up and down, around a bid that matches the keyword’s average value per click. Value-based testing also allows companies to use predictive modeling for instances where there is no conversion or value data, making initial testing strategies on new words more successful while learning from previous activities. Practitioners of value-based bid testing are able to drive improved financial performance with very little risk to paid search budgets.

Finding A Balance

Making bid testing a common practice (without consuming the budget and thereby resulting in constant monitoring by accounting or your CMO) can be tricky, but it is a feasible system. This means evaluating your goals and aligning your bid testing strategy to them in a holistic way. Ask yourself:

  • Are you looking to maximize conversions?
  • Is achieving optimal profit most imperative?
  • What sort of return on ad spend (ROAS) is ideal?
  • Does cost-per-conversion play a big factor?

Nearly every paid search marketer would rightfully say all are important. But the reality in bid testing is that achieving the optimal for all of these, all the time, is incredibly time-consuming and expensive, if not nearly impossible. Because of this, it is imperative to weigh which one or two objectives are most important, focus on them, and balance that focus with the other factors of bid testing needing to be addressed.

Once this balance is struck, assess how much upward or downward testing you can conduct with relation to what level of financial risk the business and budget is able to cope with. Ultimately, bid testing comes back to the basic math displayed above: bid a keyword up incrementally and try to recover that with a performance gain, or bid downward incrementally and try to maintain existing performance levels.
Net Profit Per Day

Systematically Approaching A Bid Testing Strategy

Part of bid testing’s negative stigma within marketing organizations and businesses stems from the fact that best practices on the subject are kept internal, creating a wide variety of methodologies with little third-party corroboration. A paid search marketer could be running a bid testing program at his agency or company knowing it is costing excessive amounts of money, but thinking that is just the way it is and not knowing any other way.

Like any other business functions, bid testing starts with determining a high-level goal within the marketing organization overall. Many choose to simply do bid testing with a shotgun approach, bidding up one day and down the next trying to increase efficiency or raise financial value blindly. However, assessing this high-level goal is relatively easy in determining your strategy. Essentially, two addressable categories exist: generate more revenue, profits, or conversions or lower costs while maintaining acceptable keyword performance levels.

With an overarching strategy in mind, the next step is to define metrics which you can tie back to your high-level goal and measure your program’s effectiveness. For instance, if improving efficiency is key, then measuring relative clickthrough rate will provide a benchmark for increased efficiency and cost savings.

Pivotal to defining your metrics is also defining constraints on your bid testing program to ensure budgets, business needs, and returns are not derailed by constant testing. With these two items established, reflecting your new goals and metrics against your existing results is necessary to ensure you are neither overreaching nor undershooting.

Test Small, Test Frequently, And Remember Your Business

It is easy to run a bid testing program all year long that isn’t tied to specific business milestones like sales, holidays, or buy cycles. However, this would render your program largely useless in driving the company forward. To this end, keep in mind your company’s objectives, buyer behaviors, and, of course, the amount of money that can realistically be used without harming the overall operation. In this way, you can adjust your bid testing frequency more accurately and more effectively for the business rather than “just be cause.”

In this same vein, an often-overlooked best practice in bid testing strategy is constant evaluation and adjustment of the program. If you are seeing a positive trend in your program, then it is acceptable to be more aggressive to see how much more value or revenue you can achieve before seeing diminishing returns. Conversely, if you are seeing negative results, you will have a work-back framework in the above steps to evaluate where misalignment occurred and correct your course quickly.

No matter the results above, the best course of action (once you have outlined how you would like to do bid testing) is to always test in small bid increments and at higher frequency. By maintaining small upward or downward bids at a faster pace, businesses will be more agile in optimizing their work, particularly in seasonal industries like travel or retail. Most importantly, by using smaller bid increments, the effects of a negative test result will have less of an impact on your budget and provide a much more rapid learning process to the paid search marketer on how they need to adjust.

Image: Bidding by Shutterstock

About the Author

Robert Cooley, PhD, is chief technology officer for OptiMine Software. He is known for his groundbreaking work in web mining, with over 15 years of experience managing projects and analyzing data, with a focus on ecommerce, CRM, and unstructured text. OptiMine develops technology that dramatically improves online advertising performance and solves complex pricing problems.

Add Your Comments

  • (will not be published)