Copy homework! Teach you how to build an operations analysis system step by step

The analysis of a single operational action is certainly important, but what really determines success or failure is the comprehensive evaluation of the strategy combination. This article will explore in depth how to go from a single activity to a systematic analysis, and teach you step by step how to build a comprehensive and efficient operational analysis system.

From analyzing a single operational action to building a system, it is an essential ability for a senior data analyst. Today I will share it systematically.

Single campaign vs. strategy combination

If we only analyze one activity, we will follow a four-step approach:

Step 1: Set goals (improve XX key indicators)

Step 2: Find the gap (see the current situation → target gap)

Step 3: Look at the process (analyze the execution process to find problems)

Step 4: Write suggestions (provide optimization suggestions for execution issues)

For example, to analyze a user acquisition optimization activity, we will do it step by step in four steps as shown below:

But! Trouble comes when it comes to analyzing a strategy portfolio.

1. Discounting coupons while offering discounts on products makes it difficult to attribute the effect
2. Each activity seems to be effective, but the overall market is not rising much.
3. Overall, more and more resources are consumed, and operating costs are getting higher and higher

Therefore, the so-called operational strategy analysis is to avoid each department working independently. Focusing on the same goal, we should sort out the resources that can be invested and then evaluate the results in a unified manner. This will maximize the overall benefits of the company. Compared with the evaluation of a single activity, strategy analysis is a more advanced approach, which is divided into four steps.

Step 1: Clarify strategic direction

The first step is to sort out clear goals, which is a prerequisite to avoid working alone.

In general operations, traditional companies will take products as their main goal and build strategies around this year's new products/traditional popular products; Internet companies will take users as their main goal, first perform user stratification, clarify the focus of user growth, and then look at the details.

If you choose products as your main goal, you can start with the layout of four major categories (as shown below):

If you choose users as the main goal, first determine the overall strategy, then determine the goals for each layer (as shown below):

This step is very important.

With the guidance of a big strategy, we can then clearly classify various small activities and evaluate their effects.

Many makeshift teams are stuck at this step, without big goals, which leads to various chaos in business. Companies with high management levels will implement the overall goals from top to bottom, and subsequent operations will be smooth.

Step 2: Collect information uniformly

To collect the following information at the same time:

1. How many activities are going on?
2. Which department does each activity belong to?
3. What goals does each activity serve?
4. Which groups are targeted by each campaign?
5. What rewards are allocated for each activity?

In short, the five elements of goal-department-object-product-reward must be aligned, so that a true strategic combination centered around one goal can be formed.

Also, if there is overlap between strategies, if offers are stacked on top of each other, or if campaigns are competing for users, this can be discovered at this stage.

This can help you avoid being fleeced in advance, and in case a certain user group is ignored, you can also discover the problem in advance (as shown in the figure below).

Step 3: Focus on overall impact

Once the event actually starts, each operator of the event will focus on their own business.

Therefore, at the strategic analysis level, it is necessary not only to monitor the performance of a single activity, but also to have a comprehensive understanding.

First calculate a few large numbers (as shown below):

1. How much of the target population is covered by each activity?
2. Summarize the actual rewards received by a single person and the actual amount spent on each goal for each activity
3. Can the overall goal be achieved under the current activity group?
4. Which goal was achieved well and which was poorly achieved?

When outputting suggestions, give priority to the overall market. This will reflect the sense of strategy, rather than being stuck in one or two specific activities.

Overall, the layout can be as follows:

In summary:

1) If the overall benefits are good and resources are abundant, explore more unknown areas

2) The overall benefits are average and resources are tight, so we focus on products with high investment-to-production ratio

3) If the overall efficiency is poor, we should solve it specifically and cut off the lagging ones first

Doing so can also provide a good answer to leaders' questions: "Why are individual activities doing well, but the overall market is not rising?"

If several activities are superimposed and the users affected are actually of the same category, then the effect of multiple activities complementing each other will not be achieved.

In short, you need to put the activities together to see the effect.

Step 4: Optimize item by item

In addition to looking at the overall results, there are also many interesting discoveries around the sub-goals (as shown below):

Imagine if:

• Special activities to awaken users attract more dormant users than the end-of-quarter clearance activities.
• The coupons used to stimulate new users to place their first order are not as much as the new users attracted to spend by the popular products.
• No one uses the VIP exclusive discounts, everyone is buying new products on sale

This means that these special activities are not as effective as overall promotions.

Either the rewards are not strong enough, or they simply fail to grasp user needs and just use some meaningless coupons to get by.

After discovering these problems, you can optimize the specific activity settings.

Furthermore, if a user repeatedly participates in an activity, it means that the user is an obvious discount lover. If a user is not interested in an activity, it may be demand-driven/brand-driven.

This grouping and labeling can also promote refined strategy formulation and benefit the overall strategic direction (as shown in the figure below).

summary

From the above, we can see that if we want to achieve true operational strategy analysis, we must first liberate operational work from isolated, single-point activities and look at the problem from a second-level perspective.

Only by integrating various proactive actions and clearly understanding the strategic combination surrounding a goal can we achieve strategic level analysis.