Two scenarios for Zhihu content to promote sales

Yesterday a client asked me why my popular content could not gain traction after investing in “Zhi+”?

There are many reasons why the volume cannot be increased. The most intuitive reasons are filtering, whether the click-through rate is too low, the bid is too low, or the topic is too narrow. However, his click-through rate is very high and the bid is not low, so why can't the volume be increased?

I’ll keep you in suspense here for now. I’ll explain the logic of Zhihu’s recommendation and search scenarios again, and then reveal the answer to this case.

Users come to Zhihu for three main purposes: to kill time, to find answers, and to watch the fun. They mainly correspond to the three major scenarios of recommendation, search, and hot list.

I have previously published an article titled "Information Flow VS Search, Anyone Doing Content Marketing Needs to Understand These Two Scenarios", in which I made some general comparisons. You can directly see the original comparison table, so I will not go into details here.

Here we focus on e-commerce sales. The main difference is of course in the logic of information transmission (recommendation is "goods finding people" and search is "people finding goods"). This determines the core things of Zhihu content sales: product selection, topic selection, and content (the last two links are path and interaction. If we focus on e-commerce, these two links are not much different).

Product Selection

The search scenario is more suitable for standard products, products with low user education costs, and some products with high average order value.

The information flow scenario is more suitable for non-standard products, fast-moving consumer goods that require user education and have low unit prices (mainly 50 to 300 yuan).

An intuitive metaphor is that search is more like JD.com. When you have a clear demand, you search, find the best result through comparison, and convert it into a transaction.

Therefore, the search is biased towards shelf logic, the SKUs must be rich, and the long-tail traffic must be continuous, but the chance of a single SKU going viral is small. Because there are only so many people who have this clear demand every day, if you want more people to have this active demand, you have to promote your products. Or your product itself is a hot-seller (iPhone, Xiaomi).

For example, in the case that I often share at the Zhi+ Client Salon, the GMV of a single article by Zhihu user MaxSam recommending laptops reached 50 million, which is inseparable from the long-term circulation of its high-quality content in the search scenario (long-tail traffic is a unique advantage of Zhihu content).

Standard products like 3C digital products are not likely to explode in recommended scenarios. Because if I don’t need to change my computer or laptop, it would be difficult to quickly convert my impulse purchases through articles. Even if conversion occurs, high-priced products may be completed a long time after viewing the content, and may not necessarily be attributed to this exposure or diversion. Therefore, for non-e-commerce products with high unit price (which I usually call cautious consumer goods, such as education, insurance, automobiles, etc.), I recommend using the private domain conversion path (add WeChat first and chat slowly).

As for the recommendation scenario, the most popular term recently is "interest e-commerce", which is actually to explore and meet universal needs through the method of "goods finding people".

Therefore, the recommendation scenario is more like Pinduoduo, which is the logic of explosive products. If the search scenario is to meet 10,000 needs of 10,000 people, then the recommendation scenario is to meet 1 need of 10,000 people.

Therefore, among the two major tools for promoting products through content on Zhihu, "Good Product Recommendations" and "Zhi+", Good Product recommendation has a larger proportion of JD products, while Zhi+ has a larger proportion of Taobao products. Because good things are natural traffic, we naturally have to cater to the users' active needs, and standard products (3C digital products, etc.) are popular. Zhi+ is a combination of paid traffic and natural traffic, so it can actively reach more potential users, and the gross profit margin of non-standard products (beauty and personal care, food and beverages, etc.) is higher.

Once you understand the scene logic, you can match product selection with the scene. If you are a brand and your products are standard products, then it would be good to increase the commission for good products (through JD Alliance, Taobao Alliance, etc.) to encourage more UGC to bring in goods (but this needs to be long-term, as search traffic is long-tail). Then, for content that has good sales effects, you can invest in knowledge + rights enhancement to accelerate it. If your products are non-standard, then you must invest in Zhi+ in order to quickly increase sales. On the other hand, if you are a creator who sells goods, you can match different goods based on different types of topics and content.

Topic:

Based on the above "warehouse matching" logic, once the product selection is determined, the scene is determined. So how do you deliver the content to the target scene?

It’s very simple, it mainly depends on the selection of topic. The topic selection determines the crowd portrait, the size of the traffic pool, and the scenario.

Let’s go back to the case at the beginning of the article. Why is it that content with a high click-through rate cannot generate sales? I took a look at his content, and the titles are basically the same:

"** Model XXXX Actual Experience | XXXX Recommendation | Brand A/Brand B/Brand C/Brand D/…"

"2021 popular XXX/YYY actual test, teach you how to choose! With brand A/brand B/brand C/brand D/... recommendation"

Among them, XXX and YYY are category words, and the following ABCD are the mainstream brand words under this category.

Isn’t this very similar to the title of MaxSam’s recommended laptops mentioned earlier? This type of cross-review list-type title has a very high click-through rate in search scenarios (generally over 10%, and sometimes as high as 20%), because it clearly provides a result that is neutral, objective, and credible.

However, if it is a recommendation scenario, because the title information is already very clear, it is equivalent to filtering the crowd through the title. Unless someone really needs it, they will not click in. This is the same as investing in information flow ads. The more open the title or even the question, the higher the click-through rate, but the target audience is also wider, so a low CPC is required; the more precise the title, the lower the click-through rate, but the target audience is more accurate, so a high CPC bid is required.

The problem with this case is that the click-through rate of the order display is a comprehensive a posteriori click-through rate (i.e., the existing result data). It is pulled up because the click-through rate in the search scenario is very high. In fact, the click-through rate in the recommendation scenario with greater traffic is very low, so the volume cannot be measured.

It is very simple to solve this problem. Just change the title to a more open and popular question in the recommended scenario (of course, if you change the title directly, it will easily affect the search inclusion, so it is recommended to write a new article).

For topic selection, you can refer to my previous knowledge + operation mind map.

The direction of topic selection also determines the content indicators that need to be paid attention to in the early stage. For example, the topics for recommendation scenarios should pursue popular ones, so in the early stage, more attention should be paid to the content click-through rate, interaction rate (such as the like-to-read ratio), etc., and the bids can also be relatively aggressive to grab high-quality traffic. When it comes to search scenarios, the amount of content is more important, and attention should be paid to search inclusion and ranking.

content:

Although both are long pictures and texts, the content of the two scenes themselves is very different.

Content-carrying format:

The recommendation scenario is more suitable for answers, and the search scenario is more suitable for articles (if you don’t understand the content format of Zhihu, please make up for it on your own). Find popular questions and write answers. This is more likely to go viral in recommendation scenarios, and the number of articles will be relatively small.

In the search scenario, articles are more controllable and there is more room for title optimization. Another point is that the questions that can be written to answer are limited, but the number of articles is unlimited and can be mapped to various long-tail search terms (those who have played SEO will understand).

Content style:

The search scenario needs to provide deterministic results to meet the user's comparison needs, so it is more suitable for horizontal review list recommendations. The structure is general, specific and general, with a conclusion and table of contents at the beginning, multiple products arranged in the middle, and a summary at the end. In terms of product ranking, those with high conversion rates are placed first.

As for recommendation scenarios (interest-based e-commerce), you need to plant grass first, so the content should first attract readers based on the universal needs of users (such as image improvement, quality of life improvement, etc.), and then explain it through multiple dimensions such as principles, effects, cases, and experiences. Simply arranging products is only suitable for categories with extremely low average order value (such as snacks and clothing priced under 100 yuan).

So to summarize, articles in search scenarios should be oriented towards cost-effectiveness, and answers in recommendation scenarios should be oriented towards a better life, good price VS good product.

Author: Yuan Chao

Source: Yuan Chao