This is too exciting. This operation analysis strategy is really on point.

Faced with data fluctuations, operators often feel at a loss as to where to start. This article will explore in depth how to unravel the complex data, find the root cause of the problem, and propose effective solutions. Whether you are a novice or an experienced operator, mastering these analysis methods will make your operations work more comfortable.

Students who work in operations often need to look at data, but they are often confused when looking at the data: "I have looked at a bunch of data, so what? So what should I do?" Especially when the indicators are falling, they only look at the "year-on-year decline, month-on-month decline" and don't know how to find solutions to the problem from the data, which makes them anxious.

Today I will systematically share with you how to find operational solutions from data. The article is quite long, so remember to like and follow it, and come back to read it slowly.

01 Problem Analysis

Too many factors may affect user decisions. The division of labor in large companies is very detailed and often completed by different groups, such as:

• Promotional copy → Content operation
• Product portfolio → Product operation
• Promotions → Event Operations
• APP operation → product operation
• User Rewards → User Operations

When the indicators drop, I have to figure out immediately whether it is a problem with my department or someone else's, so that I can decide whether to work alone or hold a meeting to "align".

Although the process of small factories is not so complicated, it is still very important to find the key issues, especially whether to give additional promotional resources. The leaders of small factories are very tight on budgets, and they will definitely have opinions if they only rely on continuous investment in activities. However, if they do not invest resources, it is likely that there will be no effect.

Therefore, finding the key impact of the problem and distinguishing the operational focus are the key to analysis. In different scenarios, the operation methods are slightly different.

02 Content Operation Scenario

In content operations, the main evaluation indicators are generally: article reading number/video playback volume, forwarding/like number, sales/registration/company WeChat number. However, the factors that affect these indicators are not all controllable by content operations. Especially for sales/registration conversion content, product quality, promotional activities, registration rewards, etc. will directly affect the results.

Therefore, when doing content operation analysis, it is necessary to distinguish clearly:

• The goal of posting is: increase followers/increase reading/sell goods
• The article itself: topic/length/writing style
• Article CTA (Call to Action): Rewards/Offers

As shown in the following figure:

In this way, when analyzing the effects, you can first distinguish the types. Content that increases followers/reading volume often does not require cooperation from other departments and can be done by yourself. Therefore, you can just look at what works best and take advantage of hot topics and attract attention like entertainment self-media.

However, the sales/conversion categories will definitely be affected by other departments. At this time, you can compare:

• Popularity of different products
• The attractiveness of different rewards
• User participation at different operation complexity

After the analysis, you can provide suggestions to other departments. Whether it is product operation, user operation, or event operation, content is needed as a means of promotion. Content reading/conversion effect analysis can directly remind them: your product selection/event planning does not seem to be that good, pay attention to adjustments!

03 Product Operation Scenario

In commodity operation, sales volume, gross profit, and inventory turnover rate are generally used as the main evaluation indicators. Commodity purchase, sales, and inventory management are directly related to sales, so they are easily affected by user operations and activity operations. What type of coupons are given to users, what products are promoted, and what categories are promoted will all affect the regular sales trend of commodities.

It is also very likely that excessive promotion of a certain product in the short term will lead to false prosperity or excessive shortages, disrupting the normal operating rhythm.

Therefore, when doing product operation analysis, you can:

• Clear product positioning: hot-selling, high-volume, and profitable products
• Refer to the sales trend of the same type and price range to draw the sales forecast curve
• According to the sales forecast curve, distinguish the listing, hot selling and delisting stages
• Record the product's participation in activities and observe the differences in trends under different activity intensities

As shown in the following figure:

In this way, when analyzing the effect, first distinguish at which stage the problem occurs, namely, listing/hot selling/delisting, and then see whether it is interfered by activities/promotions/promotions. Common problem combinations:

• There were no activities to support the listing period, and the growth did not meet expectations.
• Participating in too many activities during the launch period, forcing sales to increase and resulting in excessive replenishment
• In order to attract users, the profit margin was excessively discounted, and the gross profit margin was out of control.
• Only popular products are used to attract users, no cross-selling is done, and other products are not sold well
• The selection of items is not good, and the products are not attractive, so other operation teams can only issue coupons.

These issues are what are truly suitable for all departments to sit down and discuss together. Avoiding mindless discounts/coupons can greatly improve the performance of each department.

04 User Operation Scenario

In user operation, the main evaluation indicators are generally: the number of new users, the number of active users, the conversion rate, the number of new VIPs, etc. User operation is the most complex scenario because it is related to multiple factors such as advertising, promotional content, and promotional activities, and it is the easiest to get entangled in various factors.

Therefore, to analyze the user situation, we must first clearly distinguish the user situation

• New user/Old user
• Which promotion channels do new users come from?
• New user activity conversion/natural conversion
• Old users log in naturally/pull back
• Old users are divided into high, medium and low consumption levels
• Whether old users use VIP benefits

As shown in the following figure:

After sorting out the information, when there is a problem with user conversion, first check whether the user problem comes from new users or old users. If the problem is more serious with new users, check the advertising promotion efficiency. If the problem is more serious with old users, first check whether there is a problem with the activity rate, determine whether a large-scale recall is needed, and then check whether the problem is at a high, medium or low level, and think about countermeasures based on consumer demand.

Here are some hidden issues to watch out for:

1. The poor conversion rate of old users is a legacy problem caused by the poor quality of a certain batch of new users.
2. There are too many subsidies for new users, resulting in a high proportion of "fleecing", and they continue to fetch more after becoming old users.
3. Over-reliance on activities. The proportion of subsidies among new and old users is very high. Without subsidies, they can't survive.
4. Over-reliance on advertising, low natural login rate among old users, and high promotion costs

In essence, these are the sequelae of over-reliance on discounts and coupons. For operations, short-term spending on growth VS improving product strength for long-term retention is a classic topic. Of course, it is necessary to first label the content, products, and user situations, and do basic analysis and attribution before dealing with more responsible issues.