January 10, 2012 at 12:05“Not a single conversion ...” or tips for analyzing large contextual campaigns
Recently, we often encounter an analysis of the effectiveness of contextual advertising of large online stores with many product categories. This post is dedicated to just two simple tips that will help you more accurately evaluate the effectiveness of each contextual advertising campaign for such cases.
1. Some contextual advertising specialists tend to compare the advertising effectiveness of each product group by its general indicators. For example, now the conversion indicator, already beloved by everyone, is used very often for this, a little less often, the comparison occurs at the price for achieving the goal (or purchase). In this case, reports of the following type are found in the reports: “Conversion of campaign A 0.5%, the price for achieving the goal is 100 rubles; campaign conversion B 1%, the price for achieving the goal is 30 rubles. We recommend lowering the cost-per-click in campaign A and using the freed budget in campaign B ”. The conclusion is quite understandable and, in the general case, correct. The only thing is that if Campaign A advertises yacht sales, and campaign B - paddles for inflatable boats? Products come in completely different price categories and in no case should they be compared in terms of conversion and price for achieving the goal.
The example, of course, is greatly exaggerated, but this does not change the essence. If several product groups are advertised, it is necessary for each of them to calculate the optimal price that you are willing to spend for attracting one buyer in this product category. For yachts this price will, of course, be high, but for oars low. Further, these prices must be considered together with the results of the campaign and, based on the deviation of the desired price for the client from the real one, to draw conclusions about the redistribution of budgets between campaigns and changes in CPCs.
2. The second advice is also related to the features of the product groups themselves. It is well known that different products have different sales cycles. Therefore, for example, cell phones are bought quickly enough - if the price and delivery methods are satisfactory, and jewelry is often bought after consulting with his wife, girlfriends of his wife or with someone else. In the end, for the store there is no difference now the purchase will occur or a day later, but for the conversion source the difference can be significant. The visitor can come to the site again to complete the purchase in any way convenient for him - from the browser bookmark, through a search by the name of the store, through the address bar of the browser. Moreover, in reality, the “merit” of this sale can be attributed to the initial entrance to the site, which could happen in particular due to contextual advertising.
Thus, when comparing conversions and prices for achieving the goal of campaigns for various goods (even from the same price category), it is necessary to take into account goods with “deferred demand”, the indicators of which can be significantly worse than the others. The multi-channel sequence tool in Google Analytics, which we wrote about earlier, is perfect for this.
Consider an example:
In the “Associated Conversions” report, we examined contextual advertising indicators by campaign. It can be clearly seen that for the category “Products for children”, the ratio of associated conversions to last-click conversions is very different from the average. This means that it is for this group of products that the purchase after repeated return to the site is more characteristic. This may be the reason for the low conversion rate of this campaign, although in reality, in addition to last-click conversions, it resulted in significantly more associated conversions than other campaigns. This means that the effectiveness of this campaign is higher than what its conversion rate shows us.
1. Some contextual advertising specialists tend to compare the advertising effectiveness of each product group by its general indicators. For example, now the conversion indicator, already beloved by everyone, is used very often for this, a little less often, the comparison occurs at the price for achieving the goal (or purchase). In this case, reports of the following type are found in the reports: “Conversion of campaign A 0.5%, the price for achieving the goal is 100 rubles; campaign conversion B 1%, the price for achieving the goal is 30 rubles. We recommend lowering the cost-per-click in campaign A and using the freed budget in campaign B ”. The conclusion is quite understandable and, in the general case, correct. The only thing is that if Campaign A advertises yacht sales, and campaign B - paddles for inflatable boats? Products come in completely different price categories and in no case should they be compared in terms of conversion and price for achieving the goal.
The example, of course, is greatly exaggerated, but this does not change the essence. If several product groups are advertised, it is necessary for each of them to calculate the optimal price that you are willing to spend for attracting one buyer in this product category. For yachts this price will, of course, be high, but for oars low. Further, these prices must be considered together with the results of the campaign and, based on the deviation of the desired price for the client from the real one, to draw conclusions about the redistribution of budgets between campaigns and changes in CPCs.
2. The second advice is also related to the features of the product groups themselves. It is well known that different products have different sales cycles. Therefore, for example, cell phones are bought quickly enough - if the price and delivery methods are satisfactory, and jewelry is often bought after consulting with his wife, girlfriends of his wife or with someone else. In the end, for the store there is no difference now the purchase will occur or a day later, but for the conversion source the difference can be significant. The visitor can come to the site again to complete the purchase in any way convenient for him - from the browser bookmark, through a search by the name of the store, through the address bar of the browser. Moreover, in reality, the “merit” of this sale can be attributed to the initial entrance to the site, which could happen in particular due to contextual advertising.
Thus, when comparing conversions and prices for achieving the goal of campaigns for various goods (even from the same price category), it is necessary to take into account goods with “deferred demand”, the indicators of which can be significantly worse than the others. The multi-channel sequence tool in Google Analytics, which we wrote about earlier, is perfect for this.
Consider an example:
In the “Associated Conversions” report, we examined contextual advertising indicators by campaign. It can be clearly seen that for the category “Products for children”, the ratio of associated conversions to last-click conversions is very different from the average. This means that it is for this group of products that the purchase after repeated return to the site is more characteristic. This may be the reason for the low conversion rate of this campaign, although in reality, in addition to last-click conversions, it resulted in significantly more associated conversions than other campaigns. This means that the effectiveness of this campaign is higher than what its conversion rate shows us.