Common operating methods of information flow advertising!

Today we are going to talk about paying a high price. We mainly discuss headlines, and you can refer to other media to see if they are applicable.

"Bidding a high price" does not mean how high the absolute value of the bid is. For example, 1,000 yuan is a high price and 5 yuan is a low price. To bid high is to indicate a relatively high price.

Offering a high price is a sensitive operation in the information flow, because offering a high price means the risk of high costs, which may bring certain losses; but sometimes it is also a key operation. In some cases, if you do not offer a high price, your account may not be able to spend even 500 yuan. The operation of "offering a high price and then lowering it" may be the key to whether the account can increase its volume.

Regardless of whether you will use the high-priced operations or not, it is necessary to understand them.

1. Two ways to offer a high price

When people say a high price, there are two ways of thinking: one is that the bid represents the true expected cost; the other is that the bid does not represent the true expected cost. Let’s talk about the first situation first.

1. The bid price represents the true expected cost

The bid represents the real expected cost, and at the same time, a high price is offered, which means that the expected cost is to be increased. It is equivalent to requiring a conversion cost of 5 yuan before, but now it is allowed to be 6 yuan, and the bid can be increased.

Because the bid is the key factor in calculating the eCPM value, the higher the bid, the higher the eCPM value and the greater the volume that can be obtained, so increasing the bid is definitely a direct and effective way to promote consumption. Among similar products, if a product has a higher bid, it will have more advantages when it is launched. For example, for the same short video product, whoever has a higher cost will definitely have a larger volume.

However, cost requirements are related to factors such as product profitability and the scale of delivery, which cannot be determined by optimizers. Optimizers still have to live by sticking to costs, so let’s focus on the second type.

2. The bid price does not represent the true expected cost

The bid price does not represent the actual expected cost, so bidding high is an optimization method, and a very common optimization method. Offering a high price is usually accompanied by a price reduction, which means “offer a high price and then lower it.”

2. Lower the high bid

When should you bid higher and then lower it? In the absence of any other options.

The ideal situation would definitely be to bid directly according to the target cost, and then the plan can be implemented without having to go back and forth with bidding.

But many times new plans don't cost any money. Wait and wait, but no money is spent. Especially when the entire account has no money spent, I am really anxious. After trying to create plans, change materials, change targeting, change labels, etc., and there is no response, I have no choice but to raise the price.

By offering a high price, we can first plan to have a certain volume, see what the cost is, and then control the cost. Generally, the price is raised by 10%~20% based on the target cost. For example, if the target cost is 200, you can bid 220~240. If it is really not feasible, then increase the price gradually.

The principle here is "you can't just sit there and wait without spending money." You have to spend money first before you can see any results.

What is the principle behind this?

The principle is to correct the system's estimates . The formula for calculating eCPM is:

eCPM=bid×CTR×CVR×1000. The plan does not run volume because the eCPM value is low. There are three influencing factors for eCPM. If the plan settings remain unchanged, we can do nothing about CTR and CVR and can only increase eCPM by increasing bids .

If the actual CTR and CVR of your plan are good, but the system thinks it is not good, then it will not work; at this time, you increase the bid to increase the eCPM, let the system see the actual CTR and CVR, and correct the system's estimates. If the actual CTR and CVR are high, then the plan can be implemented even if the bid is lowered.

It should be noted that because the high price is used to correct the system's estimate, it cannot be invested on a large scale. It must be invested with a "small budget" (for example, 500, 1000, 2000). Otherwise, it is easy to lose sight of one and plan to spend tens of thousands, resulting in a big loss.

This is the principle of “starting with a high price and then lowering it”, which is equivalent to the manual version of “increasing volume with one click”.

Does that mean the actual CTR and CVR are definitely higher than the estimates? Of course not, there are definitely cases where the results are lower than expected. This is one of the risks of bidding high: if the plan still works when the bid is lowered, the plan is tested successfully; if the plan does not work when the bid is lowered, the plan is tested failed.

For example: an advertising plan bids 30 yuan, and the estimated CTR and CVR are low, so there is no volume; the bid is raised to 40, and it costs 2,000 yuan, and the CTR and CVR are slightly higher. If the plan itself is feasible, the bid can be reduced from 40 yuan to 30 yuan and it can still run; if the plan itself is not feasible, then the plan will have no volume if the bid is reduced.

A high price is often accompanied by high costs. Next, let’s look at the cost issue.

3. Cost issues after offering a high price

Let us first make it clear that the principle of cost control for conversion bidding is: conversion cost ≤ bid . The system will not understand that you have a real expected cost in addition to your bid, nor will it help you optimize according to your real expected cost.

Therefore, it is normal for the conversion cost to be below the bid. It may be lower than your actual expected cost, or it may be higher than your expected cost. At the same time, the system has a principle: try to make a plan run at the lowest cost it can achieve. So we see a lot of cases where the cost of conversion is much lower than the bid. So can the high price achieve the expected cost? We look at this in two situations.

Let’s first talk about the situation where costs meet expectations. If you can achieve your expectations, there is no need to lower your bid. Just keep bidding. This situation is rare, the cost is low, the plan is very competitive (the bid is high, so the eCPM value is very high), it is simply perfect.

Like the following two plans. The conversion goal of both plans is installation completion, so the Android installation completion cost is the conversion cost.

The conversion cost is the same, which is more than 5 yuan. The difference between bidding 40 yuan and bidding 6 yuan in a day is huge.

But it also has a problem. The problem is that the conversion cost may rise at any time. Take the first plan above as an example. Now the conversion cost seems to be more than 5 yuan, but it is very likely to rise to 10 yuan, because your bid is 40. As long as it is lower than the bid, the system will consider that the requirements are met.

And when costs are high, it is difficult for optimizers to control them. Because there is a difference between the conversion cost and the bid (let’s call this difference the “gap”), it is impossible to achieve the effect of “lowering the bid will reduce the cost”. It may even happen that the cost increases after the price is lowered.

For example, in the following example: reducing the bid from 40 to 35, the installation completion cost increased from 5.8 to 6.2. So how can we control costs? Unless you lower your bid to the true expected cost, you won't be able to achieve it.

Because the system's task is to "make the conversion cost lower than the bid", within the range of being already lower than the bid, the cost will not be controlled by the "straight line" of the bid - if the bid drops by 5 yuan, the cost will also drop by 5 yuan.

And this is a very common situation. It is relatively rare to offer a high price and have the cost meet expectations. In most cases, a high price is offered and the conversion cost is lower than the bid but higher than your expectations. For example, if your expectation is 1 yuan, the bid is 10 yuan, and the system runs out of money, it is 3 yuan. What should I do at this time?

It's difficult. But we have to do it even if it is difficult.

Always remember that our marketing goal is to gain volume while keeping costs reasonable. If the conversion cost cannot meet the requirements, then the price will have to be reduced. Reduce until your target cost is met. Don’t worry about “whether price reduction will harm the plan”. Paying high prices to gain volume is a short-term strategy that ultimately serves long-term costs. If the "short-term cost" is too high, or even if you can only survive by making high-priced plans, then you must give up and stop the loss in time.

There are also two simple tips in the price reduction process.

4. Tips for price reduction

In the process of price reduction, it is generally believed that " reducing prices while spending money " is better. The plan will easily fail if the price is gradually reduced as consumption increases or if it is reduced too much at one time. (A “dead” plan means no money will be spent, and a “live” plan means money can be spent.) You can refer to it and reduce the bid to the expected cost in 2 to 3 times, 1 to 2 times a day, and reduce the bid by about 10% each time . But we also have to consider the cost of testing. We can’t spend too much money when bidding high, so we can’t lower it too slowly.

There is another trick when lowering prices: the "gap" between the general bid and the conversion cost can be considered as a safe space for price reduction. If you lower the price within the range of the "gap", the plan will generally not fail because the plan can already reach this cost. If you lower the bid to below the conversion cost, there is a greater probability that the plan will not run after the price reduction.

It is difficult to reduce prices because it is easy to spend no money after reducing prices. But only by lowering the bid to your expected cost can you effectively control costs. So you have to surrender no matter what. If you die, the test fails, just accept it.

If I have to summarize something, it is that if you can offer a low price, try not to offer a high price. When you must offer a high price, raise it a little bit at a time. Don't offer it too high all at once. The higher the price, the harder it is to lower it, and lowering the price is too painful.

Let’s look at a failed application case of “starting at a high price and then lowering it”.

5. A failed case

This is a second-tier e-commerce account, and the expected cost is 35. The account had not been spending any money, so I raised the price of a plan that I thought was more promising, from 35 to 40, 50, 60... all the way to 88, and finally I spent money. Let’s take a look at how it runs.

The delivery data of this plan:

Operation log:

This plan started on the 9th, with a budget of 1,000 and a bid of 88. The budget was hit on that day and the conversion cost was 50. The cost of 50 is much lower than the bid of 88, but it is still higher than my expectation. I hope it can reach 35. It is difficult to control costs by bidding at this time, but it is not possible not to control them. Selling like this will result in a loss, so I plan to forcibly reduce the bid from 88 to 35.

In order to avoid the plan from being killed if the price was reduced too much, I first lowered the bid to the existing cost of 50, and reduced the budget from 1000 to 3000, letting it spend money while reducing the price, to see if I could get it down.

However, after the price reduction, the planned conversion rate dropped. The conversion cost increased significantly the next day, from 50 to 71. The planned spending speed was also very slow, only 200 yuan a day. If you still want to save it at this time, you should raise the bid to around 75, so that the bid is higher than the existing cost. However, I think the price reduction is too large. It is difficult to reduce it from 88 to 35, so I didn't raise the price and just let it go.

On the third day, the plan performed even worse, with the cost reaching 116. It was probably because the model became messy after the price reduction, and because there was a large gap between the initial conversion cost and the bid, it was difficult to compensate, so it was shut down. Test failed.

There are risks in offering a high price. If the price is lowered, the plan may fail. You need to choose carefully.

Summary of key points

1. There are two ways to bid high: one is that the bid represents the real expected cost , and the customer acquisition level is increased by bidding high; the other is that the bid does not represent the real expected cost , but is only used as an optimization method, and the bid must be lowered after it is bid high;

2. "Offer a high price and then lower it" is a common operation in the information flow. After trying to build a plan, change materials, change targeting, change labels, etc., but there is no effect, you can try to raise the price. You can't just wait and see without spending money;

3. The rhythm of using this method is " no money-increase price-spend money-lower price ". Set a small budget when raising the price, and release the budget after lowering the price;

4. Its principle is to obtain real delivery data by increasing bids, thereby correcting the system's estimates. It is often used to test new materials and new plans. If the plan itself works, you can make it work by offering a high price and then lowering the price. If the plan itself does not work, the plan will die after the price is lowered.

5. The cost problem and corresponding strategy after offering a high price can be summarized as the following picture:

6. When lowering prices, it is worth paying attention to the “gap” between the conversion cost and the bid . This “gap” can be considered as a safe space, and the price reduction plan generally will not fail within this space.

It is suggested to “reduce prices while spending money” and reduce prices gradually. You can refer to it and reduce the bid to the expected cost in 2 to 3 times, 1 to 2 times a day, and reduce the bid by about 10% each time .

The above is for this sharing.

Related reading:

High-quality landing page optimization techniques for information flow!

The most comprehensive guide to Kuaishou information flow advertising

An inventory of information flow video advertising formats!

Your information flow ads are not converting well? 2 big problems!

Tips for planning landing pages for information flow ads!

Author: Aunt Ning

Source: Sanlitun Information Flow