Click Through Rate (CTR) Definition
Click-Through Rate (CTR) is an essential concept in search engine marketing, pay-per-click ad campaigns or email marketing campaigns that is the percentage of people who upon seeing an ad or viewing an email click on the ad or link.
A Little More on What is the Clickthrough Rate
CTR (Click-Through Rate) is a metric of performance that is expressed as a percentage measuring the number of times an email, organic search result, or ad gets clicked relative to the number of impressions or times it’s been viewed. Most often it is used to gauge the effectiveness of paid search, email and display marketing campaigns. It can also show the performance of meta titles and meta descriptions, subject lines, and ad copy.
Typical examples of where CTR gets measured include:
- An element on a website such as a headline, image, button, etc.
- A link in an email or call-to-action button
- A Facebook display ad
- An organic result on a search engine results page (SERP)
- A landing page link
- A pay-per-click (PPC) ad on a SERP page
A CTR gets determined by dividing the number of all the impressions by the amount of total clicks times 100. For example, if an advertisement for a blender generated 100 clicks and 8000 impressions, the Click Through Rate is 0.80 percent:
CTR = 8000 : 100 x 100 = 0.8%
The CTR is the telltale indicator of marketing campaign success. With the calculation of how many people end up clicking on the ad after viewing it, it shows the weakness or strength and quality of the keywords, imagery, ad copy, and positioning. The quickest way to boost conversions that lead to sales is to improve CTR. To determine if email marketing campaigns, SEO (Search Engine Optimization), or display and paid search advertising are doing a good job, a company can compare their Click Through Rate to their industry averages.
References for Click Through Rate
References for Click Through Rate
- Web-scale bayesian click–through rate prediction for sponsored search advertising in microsoft’s bing search engine, Graepel, T., Candela, J. Q., Borchert, T., & Herbrich, R. (2010). Omnipress. The authors outline a new prediction algorithm for click-through rate based on Bayes’ theorem. It follows a model that traces discrete or input features of real-value to probability. The algorithm keeps with Gaussian or normal distribution beliefs. It is used for paid search on Microsoft’s Bing search engine.
- Predicting clicks: estimating the click–through rate for new ads, Richardson, M., Dominowska, E., & Ragno, R. (2007, May). In Proceedings of the 16th international conference on World Wide Web (pp. 521-530). ACM. A significant component of the web browsing experience is search engine advertising. The likelihood of a user viewing and clicking on an ad depends critically on the advertisements chosen and the order in which it’s viewed. Revenue depends on the ad ranking. It’s important to know the click-through rate of ads. Repeatedly displayed ads can be measured empirically, but something different must be used for new advertisements. The authors demonstrate how features of advertisers, terms, and ads can develop a model for predicting CTR for new ads. The model increases user satisfaction and revenue.
- Predicting click–through rate using keyword clusters, Regelson, M., & Fain, D. (2006, January). In Proceedings of the Second Workshop on Sponsored Search Auctions (Vol. 9623, pp. 1-6). Sponsored search providers, search engine optimizers, and advertisers have interest in click-through rate. The CTR is a tool for predicting revenue, evaluating cost, and determining the effectiveness of an advertisement. Click history provides material information for investors to predict future click-through rate. However, there isn’t always sufficient historical data. The authors hypothesize that specific terms can instinctively have different probability of getting a sponsored click. They predict the level of CTR for term differences in likelihood. Click history might still not be sufficient for estimating term level click-through rate, so the authors suggest including groups of related terms for less frequent ones. The authors discover that they could get more accurate estimates of CTR by using click history information aggregated by clusters of words.
- Spatio-temporal models for estimating click–through rate, Agarwal, D., Chen, B. C., & Elango, P. (2009, April). In Proceedings of the 18th international conference on World wide web (pp. 21-30). ACM. The authors use a Gamma-Poisson model to track the CTR of articles on one site. They combine data from related sites. A new model is presented to estimate the click-through rate within the context of recommendations for content. The model is demonstrated on a Today module that is commonly published on Yahoo! Front Page. Experiments are done to test the model in various scenarios. They yield encouraging outcomes.
- The impact of content and design elements on banner advertising click–through rates, Lohtia, R., Donthu, N., & Hershberger, E. K. (2003). Journal of advertising Research, 43(4), 410-418. This study examines the effect of design and content elements of banner ads on click-through rate, using data from 8,725 actual banner ads. The study is the first to investigate the impact of banner advertisements using CTR and the first one to look at the differences between business-to-business (B2B) and business-to-customer (B2C) banner ads. The design elements studied include animation, color, and interactivity. The content elements were emotional appeals and incentives. The results suggest these elements work differently for B2C and B2B banner ads.
- Internet advertising effectiveness: the effect of design on click–through rates for banner ads, Robinson, H., Wysocka, A., & Hand, C. (2007). International Journal of Advertising, 26(4), 527-541. Online advertising has seen tremendous growth since it was introduced in 1994. This study looks at the effect of seven characteristics of 209 banner advertisements on the impact of online advertising using a multiple regression model. The industry chosen for the sample was gaming. The results were widely in line with previous studies showing that the creative characteristics most effective in banner advertisements included the absence of promotional incentives, larger size, and information about casino games. Animation and action phrase was among the aspects that had no impact on click-through rate. Surprisingly, higher CTR correlated with long phrases on the banners.
- Learning the click–through rate for rare/new ads from similar ads, Dave, K. S., & Varma, V. (2010, July). In Proceedings of the 33rd international ACM SIGIR conference on Research and development in information retrieval (pp. 897-898). ACM. Search engine advertisements are typically ranked depending on their CTR. Therefore, accurate predictions of click-through rate are of the most significance for maximizing search engine revenue. The authors present a model that obtains data of CTR for semantically related new or rare ads.
- Click–through rate estimation for rare events in online advertising, Wang, X., Li, W., Cui, Y., Zhang, R., & Mao, J. (2011). In Online Multimedia Advertising: Techniques and Technologies(pp. 1-12). IGI Global. Click-through rate is one of the most educational metrics used in business for budgeting and evaluation of performance. Due to sparsity in data, predicting CTR for rare events is challenging. The authors come up with methods and models by taking advantage of the natural ranking of data to smoothen estimation of CTR.
- Click fraud resistant methods for learning click–through rates, Immorlica, N., Jain, K., Mahdian, M., & Talwar, K. (2005, December). In International Workshop on Internet and Network Economics (pp. 34-45). Springer, Berlin, Heidelberg. In pay-per-click online advertising, advertisers are paying for ads only when somebody clicks on them. There are many advantages, but the system is susceptible to click fraud, which is when a service provider or advertiser creates clicks on advertisements intending on increasing the payment of the advertiser. The authors have learned of click-based algorithms which are in a way unaffected by click fraud.
- Using boosted trees for click–through rate prediction for sponsored search, Trofimov, I., Kornetova, A., & Topinskiy, V. (2012, August). In Proceedings of the Sixth International Workshop on Data Mining for Online Advertising and Internet Economy (p. 2). ACM. The authors outline a new method using MatrixNet to solve the prediction problem of click-through rate. The problem is significant because, in order for the search engine to know which ads to display, the prediction of CTR is influential. It has an impact on the search engine’s revenue and user experience. The authors explain various issues like feature importance, training data set size, evaluating and tuning MatrixNet algorithm, and performance.