Aside from technology, this article tries to explore from the perspective of products, how to design a basic automatic price comparison product, what stages are there, and what should be output at each stage.
For e-commerce companies, the price comparison system is like underwear. The difference is not obvious when you wear it, but if you don’t wear it, it feels empty and unreliable. After spending time and effort to compare prices, buying is a sense of security. After all, it is the nature of consumers to shop around, and companies cannot lose the starting line for pricing.
E-commerce also pays attention to a scale effect, so the automation of price comparison is inevitable. The core technology of price comparison is crawling, and the core process is comparison.
1. In-depth understanding of the price comparison object
As the saying goes: “Know yourself and the enemy, you will never die in a hundred battles.” The product manager must take the lead, starting from the most primitive manual methods, and using all feasible means to understand the objects that need to be compared.
PC station, M station, APP, open all, download all, get everything ready.
First, take a bird’s eye view of the other party’s product system and content organization system from 10,000 meters above. Then observe the other party’s new rhythm, store organization, special stage rotation, and event organization methods. Summarize each contact route on the business details page. Browse from the category menu? Entry from the search list? From the labeling system? From the special & event page? From the brand store? Are there other routes that can be reached?
Again, what information can be obtained from the business details page? Title, price, promotion, shipping, reviews, pictures, specifications, categories, and other fields. What are the differences between PC station, M station and APP?
Finally, the output of this stage is a summary of a 3-dimensional cube. The first dimension is the optimal route to reach, and the criterion is that the shortest route can reach the most SKUs. Simply put, it is the efficiency of the data. The second dimension is all the field information that can be obtained on the business detail page on each platform. The third dimension is each price comparison object (don’t tell me just want to compare one family). With this cube, the strategic focus of the comparison becomes clear.
Some people may say that I only want to see if the other party has the corresponding field when I need it. Users can think so, products can’t. Product capabilities must be based on vision. A deep understanding of the price comparison target is the only way to expand your horizons.
2. Investigate competing products, summarize the logic, and establish a system
There are also many mature cases for reference in similar price comparison systems. There are 2C and 2B. 2B’s products are the treasures of various e-commerce companies, with their own unique logic, and are not within the scope of basic product discussion.
But if you have been in some well-known companies, it will be of great help to the establishment of product logic. I just give you one example of 2C, Huihui Shopping Assistant. The difference between 2C and 2B is whether there is aggregation or cross-section analysis, and whether to observe the market from the perspective of the merchant.
So the question is, what we are doing here is a 2B product without prior experience, what should we do? At this time, we need to think about the role of price comparison products in the upstream and downstream. Is it a decision or a reference? Is it recommended or controlled? Is it automatic or manual? How to settle the data? What is the content and frequency of precipitation? What business value can this data generate for me in the future? How to deal with unknown products? How to continue the logic of price comparison when it does not match with your own SKU? Who are the potential users of the product? From what angles will they use the product? What is the shape of the evolution of the product at each stage? Web query? Report summary? Supply chain integration optimization? Procurement decision? Demand Forecast? What is the ultimate goal of price comparison?
Interview as many people as possible, interview as high-level managers as possible, and break the unknown boundaries as much as possible.
Finally, the output of this stage is a complete product logical structure diagram, which mainly answers three questions: how data flows, how humans and machines interact, and how values are reflected. This is very important, because this is the product manager’s own understanding of the product and the world view of the product.
3. Fully collect operational requirements
The logical skeleton of the product established in the previous stage should be supplemented with flesh and blood at this stage.
Operation (actually including commodity operation and procurement and sales) is not only the originator of the original demand, but also the ultimate and most frequent user. Most of the value of price-comparative products is realized through the work of operation.
The product manager needs to go to the operation team, go to the daily actual work, and work with the operation members for a period of time to explore the pain points and scenarios of the product to solve the problem from the most realistic perspective.
Product managers summarize the needs of operations as functional points and need to avoid three major pitfalls: Don’t take users’ false needs seriously, don’t imagine non-existent needs, and don’t promise low-value needs.
There are three ways to meet demand: improve reality, lower expectations, and shift demand. Improving the reality has the highest cost and requires development and participation. It has a long cycle and large investment. It is the most compelling means to use after expectations cannot be met, and the goal cannot be achieved through other means.
The output of this stage is an ordered list of requirements, and the product manager needs to sort by comprehensive cost and value.
4. Detailed design of the front and back ends of the product
2C products play user experience, 2B products play business logic. In enterprise application scenarios, products need to pay more attention to the logical order of the business. At this time, reasonable control of the back-end organization and context is much more important than the front-end design. Integrate the outputs of stages 1, 2, and 3 to assemble a complete functional structure of the product. This is a cyclical and progressive process.
First determine the core of the product, including the minimum feature set that can maintain the normal operation of the product, and a list of requirements that would rather be hacked than you can. Then there are enhancements, and the outermost layer is the user experience. The 2B product has an arrogant place, that is, as long as the experience is not too bad, it will not affect the use.
2B products make money and make money is king, feelings are the least important, things that can work in black and white, don’t waste a minute to decorate it. If the product’s understanding of the business is not in place, the final outcome is: if the operation is not satisfied, no one uses the product; if the development is not satisfied, no one uses the things made.
How does Sankei evaluate himself? Look at the satisfaction indicators of these two groups. The output of this stage is a carefully designed list of product features.
5. Estimate the workload in advance with the development classmates
If the product is to be implemented, it is necessary to take care of the development gang. It can be understood as respect, or it can be understood as coaxing. No matter what means, letting developers work actively and happily is the final result.
The first and most important step is to disassemble the product goals in place and sort out the approximate workload of each function point one by one with development. With this data, the product manager knows very well which part of the magnificent idea in his mind can be realized by the poor scheduling resources in his hand.
The output of this stage is the approximate time-consuming estimate of the senior development on the function points, but note that this is not the final time, nor the commitment to the realization of the function. It is likely that you will have to ask a few more people to get a reliable answer. The product does not need to do this step, but it must be psychologically prepared to be questioned and challenged.
6. If you want to refine the product, you must first come from the palace
In the face of time constraints, cutting function points, there is nothing to say. It is better to cut yourself than to be cut by others. The process of self-cutting is also a process of reflection, a process of summarizing, and finally a process of sublimation of self-values.
The output of this stage is a castrated version of the function list, that is, the requirements document that has reached preliminary agreement with multiple parties. Unlike ordinary castration, the leftover part is the most valuable part.
7. Review is just a process
The first 6 steps are good, and the requirements review is just a process.
8. After the development started, the work has just begun
Generally speaking, when a product enters the development stage, there are three follow-ups: follow the progress, follow the quality, and follow the function.
2B’s products also have three follow-ups: change with functional objects, change with business process, and shift with value focus.
Comparing products with similar business and strong coupling products, there are three other things: capturing data with competitors’ prices, comparing data with self-operating prices, and using and adopting data with users.
The total of the above is nine followers, and the nine products of our day are based on our products. Data is God’s handwriting. These data need to be followed every time, every day, every week, and every month. If there is any error, the problem should be discovered in time and the development students should be urged to correct it.
The most valuable asset of 2B products is data, not function. The accumulation and precipitation of these data will also bring a more reliable basis for the improvement of the next version of the product, and bring more powerful support for a higher level of vision.
9. What is the value of the product manager?
- More than just porters for operational needs
- Establish your own product logic system, business is the skeleton, function is flesh and blood, and value is spirit
- Give full play to leverage, and try to think about how to use the least cost to maximize output
- Leverage each other with development, and achieve each other
- If there is a problem with the product and it is not discovered in the first time, it is time to review and reflect. How to find out? Let the developers provide reports and use the products continuously. Can’t stop for a moment!
One last sentence
Believe that everyone has done their best, rather than questioning. Even if the facts do not match, thinking this way will make you feel better.