Stay away from the five major misunderstandings of direct promotion and don’t worry about lack of traffic!

　　I believe that most drivers have experienced a period of confusion during driving, especially many students who have systematically learned the theory of direct trains will repeatedly ask some basic questions in practice.

　　There are three reasons for this.

　　First of all, the basic theoretical knowledge is not solid enough. During the operation, you cannot combine it with the actual situation of your product or category, or you do not know what the purpose of driving is, which leads to a confusion during operation;

　　Secondly, I am not proficient in the practical application of driving and cannot find data problems, but I do not know where to make corresponding adjustments;

　　Third, the appearance does not have a thorough understanding of data, and the deeper it is the lack of ability to analyze the market level. After the data is feedback, the problem cannot be read.

　　From the communication with most classmates, I selected several typical questions, which are also answered by more than 90% of people who ignore but are particularly important when driving.

　　Hope it will make your thinking clearer. Secondly, I wrote down the practical methods of several important strategies of the express train in a fool-like and one-stop manner, so that students who are short of time and have no time to conduct in-depth research will immediately enter the practical operation step by step like reading the instruction manual.

　　Crowd combination errors and incomprehension of keywords

　　When doing market positioning analysis, you often don’t know a subcategory

　　Since the direct train joined the crowd, the weight has begun to move, and the proportion of the crowd weight has become higher and higher, and the crowd has given the keywords higher weight.

　　There are two benefits to putting people in:

　　First, it allows you to place it more accurately to the consumer group you want. The people here are more accurate than targeted and diamond exhibition groups. Second, you can obtain higher label weights.

　　When driving, you must open the selected crowd, and the best all clicks come from the selected crowd.

　　Although many people have opened up selected crowds, they have not selected crowds very well. First, the crowd is blindly grouped together, second, they don’t know how to continuously optimize crowds, and third, they ignore how the crowd bids for premiums, and their eyes are still focused on keywords.

　　There are generally four ways to select a store to target people:

　　1. Test the precise crowd of this store or this baby through custom groups (common methods)

　　2. Traffic vertical and horizontal - traffic - visitor analysis

　　3. Single product analysis - visitor analysis (single product visitor portrait)

　　4. Market Professional Edition - Buyer Group (View category visitors group portrait)

　　The first method is recommended to use, that is, the data test of the express train will be more accurate. The 2nd, 3rd, and 4th methods all have problems of incompleteness or data deviation, especially the portrait of the industry population. Babies at all levels of price in the industry are in the data.

　　If there are serious order brushing behaviors in the industry, the crowd portrait will be inaccurate.

　　So how to test the crowd through the express train? Before understanding the practical methods, there are two important points that you need to understand.

　　1. The relationship between population and keywords

　　The essence of a direct train is search, because the logic is the same. We place keywords (baby title), optimize creative ideas (main picture), keyword bids (baby title weight ranking position), buyers search for keywords, and the system displays them to buyers based on bid/rank position.

　　If your main image/creative picture is done well and the click rate is high enough, the buyer will click to enter to form a UV. The crowds of people in the express train are dependent on keywords. The background explained this:

　　The number of people covered here is an estimated value, which is related to the estimated population of people covered by the keywords added by your baby, as well as this filtering criteria.

　　The buyer has an account tag, and the unit price, gender, age, and average monthly consumption limit of the category in the custom group are all account tags.

　　After the account tag is associated with the keyword, a crowd pack for this keyword is formed.

　　For example, if you place the word "dress", when a buyer with a gender of a woman searches for a dress, his account will be labeled as a gender "woman" under the word "dress".

　　At the same time, the system will be included in the crowd bag with the word "dress". The crowd in the crowd bag is not static, and the update cycle is 72 hours. Moreover, the account tag will be associated with many keywords, such as the buyer of this gender. While searching for a dress, he will also be associated with the mask.

　　The system will constantly update the crowd of this group of people based on the buyer's search behavior within 72 hours. Therefore, the crowd will fluctuate and change. When you place fewer keywords, it will inevitably lead to the number of people covering less.

　　2. The relationship between crowd premium and keyword bidding

　　It is recommended to drive now and try to put all clicks into the crowd, which involves how the crowd will pay premium. The crowd's premium is above 100% and can basically ensure that the crowd is placed in the crowd. There is no mandatory requirement for the specific premium.

　　For example:

　　If you place a "dress" bid 1 yuan, assuming the ranking is 20th, the crowd has a premium of 300%, and the crowd is "female + 18-24 years old", then the keyword bid under this group becomes 1* (1+300%)=4. If you want to rank 3, then when people outside this group search for "dress" your display ranking will be 20th. When the buyer's label meets "female + 18-24 years old", the keyword bid becomes 4 yuan, ranking third, and is given priority to the people listed. Therefore, selecting the crowd can help the delivery of the express train more accurately and efficiently.

　　3. How to combine people

　　There is no explanation for the store customization group, Taobao high-quality group, and promotional group. You can go to the backend of the direct train to see the specific instructions, and focus on the custom group.

　　Customized groups usually have:

　　Category unit price

　　gender

　　age

　　Average monthly consumption amount

　　Among them, the population packages of gender, age and average monthly consumption quota are basically the same. The unit price of the category refers to the average purchase price of visitors under the category in the past 180 days. They are the customers who have been sold, but other categories include searching for unsold and searching for customers who have been sold.

　　When combining, gender can cover all other categories of attributes. When testing the population, they are combined in the early stage by "gender +1", and later, they can be further combined and optimized according to their respective performance.

　　Click rate: The optimization direction is unclear

　　There is no need to mention the importance of click-through rate. Everyone should know that through click-through, click-through rate largely determines the size of the planned weight. The intuitive manifestation is that the high click-through rate can reduce the PPPC, making driving more efficient and cost-effective.

　　Therefore, the excellent click-through rate is more than 1.5 times the industry click-through rate. Usually, the click-through rate can be optimized is as follows:

　　1. Keywords: Three principles of word selection, high click index, good correlation, and high click rate. Of course, it would be even better if the competition is small. This is the so-called temple. During the early selection of words, the word with high click-through rate is artificially selected to help the entire plan to achieve a higher click-through rate. During the optimization period, the words that are not performing well can be optimized and left with good performance words.

　　2. Crowd: After the crowd passes the correct combination, after a cycle of testing, data will be feedback, and the desired crowd and excellent click-through rate will be gradually optimized.

　　3. Placement area: After one cycle of delivery, sort the region according to the click rate according to the direct train report, and gradually optimize the areas with poor click rate.

　　4. Delivery time period: Generally speaking, the click rate is relatively high during the time period with less competition, and it can also be delivered according to the recommended industry time period of the system. Most time periods are used to control keyword bidding discounts, control conversions and delivery volumes

　　5. Creative quality: Optimize the click-through rate of the creative concept and the relevance of the creative title. It is recommended to fill 4 creative concepts, which helps to overlap the relevance and weights of the creative title.

　　In the early stage of driving, when you are weighting on quality, everyone knows that you need to optimize the click rate. The click rate we are talking about here refers to the plan/baby click rate, but the plan click rate actually corresponds to the creative idea click rate.

　　Many students will ask me questions like this: On the second day of the score-up cycle, I have optimized the keywords, crowds, and regions of poor performance on one side according to the teacher's guidance, but found that the car could not burn out, or the click rate did not change significantly.

　　Combining the above content and this question, 99% of people ignore the first and most important step in optimizing click-through rate, which is to optimize the creative concept! Not keywords, regions, and populations.

　　When you optimize keywords, crowds and regions with poor click-through rate performance, fewer words, fewer people, and fewer displays, of course, there will be no burning out!

　　When black cars were popular, the purpose of various rampage and pornographic pictures was to achieve high click-through rates! Everyone forgets that optimizing creative ideas is the fundamental way to optimize click-through rates. Keywords, regions and populations do need to be optimized, but I think these are all fine-tuning of later strategies.

　　So everyone must remember: be good at click-through rates and optimize creative ideas! Optimize creative ideas! Optimize creative ideas! A good creative idea to solve all problems with the express train! Yes, all the questions.

　　ROI: Error interpretation of ROI roi=output/input

　　Here is a roi profit. The guarantee formula: when roi≥1/profit margin, it means that the roi at this moment is guaranteed or profitable.

　　For example, the profit margin is 50%. When your roi can be greater than or equal to 2, the car in this account will be profitable or guaranteed.

　　However, there is nothing wrong with understanding this ROI from a theoretical perspective. You do need to achieve conversion rates when driving, and you cannot drive a car that costs money. However, under the current industry competition, there are really not many categories of cars that want to ensure a guaranteed capital.

　　So, can’t the ROI drive a car that loses money?

　　Obviously not! In addition to ROI, there are many other strategies. In terms of output, ROI is indeed a consideration indicator.

　　But in the long run, can the direct train increase the weight of the baby? Is it cost-effective to exchange for short-term investment for long-term hits? When measuring roi, you can’t just look at profits and losses. Try it from another perspective? Check out this data

　　Judging from the data, it is a good performance. The click conversion rate of the express train is twice that of the industry. This model has only 5% conversion on the search side of Taobao, and the average customer price is only 55, but the ppc is 1.7. It is already very good that the ROI can achieve 1.97, thanks to the optimization of the express train conversion.

　　However, if you calculate profit and loss according to this roi, it is still a loss! Because the profit margin is 30%, the net profit is 148565*30%=44569, and the investment cost is 75555. This car has a monthly loss of 30,986, and the number of clicks it brings is 46,074, and the loss is more than 30,000 in a month. Moreover, the data of this car has no room for optimization, and it has basically been optimized to the extreme, because the average customer price cannot be changed at once. The ppc is already 3 cents lower than the industry's 2 yuan, and the conversion rate has been twice as high as search. So can this car be driven?

　　From another perspective, I lost 30,986 yuan in a month, and the clicks I brought was 46,074. On average, my click costs were 30,986/46,074=0.67 yuan. Instead of looking at the data report feedback ppc, one click of 1.7 yuan,

　　At the same time, this express train helped me maintain and stabilize my search traffic. Secondly, 10% of these more than 46,000 visitors became my customers, and more than one person bought other products in my store, and most of them became potential customers of my store.

　　In fact, from the perspective of 0.67 yuan, my customer acquisition cost is not high, and it has driven the related sales of my entire store. What these more than 46,000 people brought not the 14w+ transactions seen in the report, but more. So it will be more relieved to look at these data by jumping out.

　　Of course, everyone has different understandings of ROI, and it must be determined in combination with its own products and industry PPPC. If the store is a single hit product and has no connection, it is not a good thing for ROI to lose long-term losses.

　　Misconceptions about how low ppc should be

　　I don’t know when it started, but everyone thinks that the PPPCs that drive must be very low. Major training institutions also place extremely low PPPCs to watch for the people who don’t know the truth. Everyone also regards driving extremely low PPPCs as their ultimate goal of driving and showing off.

　　But my understanding is that the PPPC for driving a ox must not be high (not high does not mean it is particularly low), and the PPPC does not necessarily mean driving a ox.

　　Please understand the purpose and essence of driving the express train.

　　The ultimate goal of driving a direct train must be conversion!

　　Whether it is a long-term goal or a short-term goal, driving without making a deal is a hooligan!

　　I have also driven a black car for a long time and have seen cases of successful driving a black car, but after thinking about it, is it really the car that was finally successful? No! It is because the product itself has the genes of popular products. It’s just that the owner is promoting it. It’s like you want to go to a destination, and you can walk faster than others, but it’s faster after getting in the car.

　　There is another problem that cannot be ignored: industry average PPPC, every category and even every word will have its own data performance: traffic analysis.

　　If the industry ppc is 2 yuan and you have to reach 50 cents, it would be a bit contrary to market rules.

　　Of course, you can trick the system into forcibly making the ppc to 50 cents by using a black car, but it will soon be monitored by the system until it is blocked.

　　I have done a test, driving a black car, and if the PPPC reaches less than 70% of the industry, it will be monitored by the system, and the black room will be closed for 30 days, and the second time will be 60 days. So what level does PPCP calculate OK?

　　In the case shared above, the industry average PPPC is 2 yuan, and we operated at 1.67 yuan, which is already very excellent because the click-through rate is very good. There are also a lot of releases, but ppc is just like that. Can you continue to drag down? Of course, but it doesn't make sense.

　　The data has achieved what I want. It is already lower than the industry. It is already excellent to optimize ppcs to the industry average or 70% below the industry. Those ppcs that rely on black cars to make low-costs are not only tasteless but also a pity to abandon.

　　Under normal circumstances, the click-through rate can be optimized to 1.5 times that of the industry, and the PPC can be reduced to about 70% of the industry. At this time, the release can be liberalized according to the later strategy.

　　Let’s summarize it in the end. When driving, you can change your perspective, grasp the essence of affairs, and grasp the key indicators. A macro-view of problems and data helps not stick to data.