From Baidu Takeaway, which once occupied 33% of the market share, was swallowed up by Ele.me, to Wen Xin’s first-mover advantage and became the laughingstock of the industry, to the collapse of Jiyue Automobile in the field of intelligent driving, this giant that laid out many cutting-edge tracks in the early days has repeatedly missed opportunities. Recently, Robin Li blamed the problem on “lack of focus” in an internal letter, but the root cause behind it may be more complex. This article analyzes the deep-seated problems of Baidu from an internal perspective: the “Truman’s world” built by the CEO isolates the real user experience, the department wall leads to internal friction and inefficient collaboration, the lack of employee motivation and the failure of the evaluation system make the executive level tired of coping, and the distortion of information makes strategic decisions deviate from reality.
Although I haven’t been in Baidu for a long time, if you ask me which company has the highest status as a programmer, I will definitely say Baidu, because the product manager will really cry and “kneel” to let us change our needs!
But in recent years, Baidu’s development has been very strange, and there have been a lot of jokes, and finally it has come to an end: get up early in the morning to catch a late market? “reputation”.
Recently, at the quarterly executive meeting, Robin Li, the founder of Baidu, finally couldn’t help but give an internal speech with the theme of “Seeking Truth and Pragmatism”, which was very rare and systematically reflected on Baidu’s organizational culture, and also reviewed the gains and losses of the business.
About Baidu getting up early in the morning and catching up with a late market is nothing more than these things:
- Wen Xin Yiyan grasped the first-mover advantage and stretched all the way, almost becoming the laughing stock of the industry;
- Baidu takeaway once reached a market share of 33%, and then was directly eaten by Ele.me;
- Baidu intelligent driving also got up very early, both the track and the early stage of evaluation were very good, and last year’s Jiyue car collapse incident caused a lot of uproar;
Things are all right, but the factory director seems to have summarized the reason as unfocused in the end, which may be wrong again…
How to say it? Maybe the factory director is different from the world we see!
1. Our world is different
This is not a joke, the factory director may really be different from the world we see! As far as I know, in some companies, various departments will make in-depth warnings about the CEO’s account!
Let’s take a practical example: a listed boss’s hometown is in Chengdu, and the video feedback is very stuck every time during the Chinese New Year.
B-end product manager’s ability model and learning improvement
The first challenge faced by B-end product managers is how to correctly analyze and diagnose business problems. This is also the most difficult part, product design knowledge is basically not helpful for this part of the work, if you want to do a good job in business analysis and diagnosis, you must have a solid …
View details >
The architect has already deployed a complete monitoring link for the CEO’s account, from the CEO opening the APP, to when he clicked what button, what video he watched, and what operations he did, almost like screen recording!!
On this basis, they quickly discovered that there was a problem with a CDN node in Sichuan, and the result here was: the boss just sent a message in the group, and the problem disappeared the next day, but other people’s problems disappeared or not, I don’t know…
As far as I know, some large factory departments will customize the development of APP functions for the CEO (no matter how complex the technology is), and it is very likely that the boss sees it differently from other users: when users complain about a lot of advertisements, the CEO is very puzzled that there are no advertisements, and when users complain that the video is very stuck, the CEO beauty swipes flying…
How do you get your boss to empathize with users? How do you want your boss to complain about the product? His experience is really good!
Our world is different, and this world is the world given to CEO Truman by middle and senior cadres!
2. Department wall
In the previous time, we did AI+ management 2B when entrepreneurship failed, and we summarized: the success of an enterprise is difficult to summarize, because either you don’t know why you became one, but if you don’t manage well, it will definitely drag down the entire team!
Baidu is a company that loses to a large extent in execution, and from a personal micro point of view, they have too many elites and smart people!
I remember that in 23 years, we were doing a medical AI project, and at that time, in addition to the self-developed model (LLama), we were also trying various large domestic models, including Wenxin, Zhipu, etc.
Baidu’s pre-sales is actually quite powerful, in order to promote Wenxin, directly transferred two T8 senior engineers to run over and debug with us every day, and also achieved some results.
Because I was from Baidu before, I quickly got closer to the engineers, so I asked about their Wenxin iteration, and the feedback from the executive level was actually a little pessimistic, which probably means:
Although Wenxin is the main one internally, there are still many teams in the company that are training with LLama by themselves, and even some teams are very troublesome to sort out the data.
Then Baidu seemed to send a “star” team to discuss strategic cooperation with us, and our boss only asked one question and announced the termination of the cooperation: we are doing medical AI, you are doing general models, Baidu Health is also doing medical AI applications, from the layout point of view, will they continue to do it?
The leader on the opposite side seems to be not low in rank, but when he says it, he is directly cold: Of course they will do it, how can we control them…
This small thing actually reflects a huge problem: Baidu is really not unified, their businesses are fighting with each other, and this situation seems to be very common?
Baidu’s recruitment culture is very fond of elites, such as qujing, “elites” like qujing are particularly active in fighting internally, after all, “the outside must first settle inside”, so there are too many things to fight insiders and laymen.
In fact, every company has a department wall, and Baidu has a department wall, but how to prevent the basic goals of the department from being mutually exclusive at the company level during major strategies is a big problem.
3. Where is the driving force?
When I was still at Baidu, it was very difficult to take over the big search business, and their product experience was not acceptable to people.
First of all, the company’s most important business documents are extremely badly written, and I need to ask more than 5 people to access traffic!
I still remember asking A a question at that time, A asked me to find B, B asked me to find C, C asked me to find D, D asked me to find E, and finally E couldn’t solve it and let me find A!
ITM worked from morning to night, and I was back to square one!
I really didn’t want to find A as it was, so I first changed a female nickname on Baidu Hi, and then changed to a small fresh beauty avatar, and then asked A, and finally I didn’t know if it was an avatar problem or a circle of knowledge increased, anyway, the problem was solved…
Another time (before changing the avatar) was also the access to the company’s core business search, because their documentation is really bad, and there are many language requirements, I really can’t help but complain: What if the accessor doesn’t know that language, do you consider adding the language type?
Who knew that the other party would overturn me with one sentence: I won’t ask you to go to the tutor, I have no obligation to teach you!
My go: This is their attitude towards their product experience…
In fact, looking back, I can understand that defender, because he does all the dirty work, and he can’t get a promotion or salary increase if he does it well, and he doesn’t have a big pot if he does it badly, so he looks like a resentful woman, and he has no driving force…
On the other hand, there is a returnee master in the team and a Beijing eldest sister (particularly rich!). I heard that he often goes to work and hangs out outside, once the department boss let the meeting and couldn’t find anyone, he lied that he was visiting a customer, and ran from the cinema to the company in 10 minutes, which is really a dick…
There are many people like this in Baidu, but just look at the resume panel, they are very good at it…
4. Information distortion and evaluation failure
In fact, many large companies have similar problems, the boss lives in Truman’s world, and he doesn’t know how the opinions of various departments are pulled, let alone how the low-level executive employees get by.
Information distortion causes the boss to see everything well, and in turn misjudges and gives the wrong strategy, one wrong step, one wrong step;
If I am surrounded by people who are cunning and slippery, then I will only do work worthy of the salary.
But the failure of this evaluation is by no means a problem at the grassroots level! It points directly to the core lesion of the company’s “elite system”: the distortion of the upward channel and the misallocation of resources.
Here you can compare the diagnosis of large company diseases that broke out in Xiaohongshu before, which is very similar to Baidu’s situation: bureaucracy is prevalent!
The leader himself will not end up, he will only draw a big picture, use macro narrative logic to push the front line, and the final result is that you fool me and I perfunctory your model.
If they are keen on “analyzing the intentions of the upper level word by word” and turn a blind eye to the “important issues” that really affect user experience and business promotion, then the product must be very difficult to use!
And the most fatal problem here may be: the promotion of upward management, the resignation of those who work hard.
From this point of view, the factory director attributes the problem to “lack of focus”, which may have seen the appearance, but the deeper reason is the underlying mechanism that supports “focus”: the upward channel has failed.
If the problem of information transmission and evaluation failure is not solved, the strategic focus will still not be implemented.
5. Conclusion
What Baidu needs may not be a simple focus, but a deep reflection: where should resources go?
What is the company’s core strategy? What are the key actions that support these strategies, such as opening up collaboration, optimizing core experiences, and tackling key technologies? The evaluation system must be strongly bound to these key actions.
Writing bad product documentation leads to high collaboration costs? Then make document quality and collaboration efficiency one of the core indicators of promotion and assessment! Has a significant contribution to breaking down the departmental wall? The reward should be much higher than holding one’s own territory.
Secondly, how to measure true value? It is necessary to break the simple mapping of “rank = value”.
Is it possible to explore a more refined valuation? For example, tiering projects/tasks so that solving complex problems, taking high risks, and creating significant user/business value are rewarded to match?
Let the value of those who do “infrastructure” and “gnaw hard bones” be seen, measured, and rewarded, rather than just letting those who do “highlight” projects eat meat.
And then break the “Truman world”? The evaluation system needs a mechanism to encourage reporting worries, exposing problems, and solving real problems.
For example, set up a channel to reward “problem terminators” to protect the voices that raise sharp questions, and take “user/front-line pain point resolution rate” rather than “boss satisfaction” as an important assessment indicator. Make the information pipeline more direct and transparent.
In fact, all the above problems seem to be things, but in fact, they are human problems, as long as the person is right, everything is right, if the entropy increase can not be solved, then find a way to break the system:
Choose the right person, form a new system, and the head will end up by himself and do a good job in business, instead of evaluating behind his back, so that the truth cannot be seen.
I believe that the factory director’s speech on “Seeking Truth and Pragmatism” is just the beginning, but “truth” and “reality” will not automatically emerge, and it requires a profound revolution on “how to measure value” and “how to allocate resources”.
What Baidu needs is not only business focus, but also a reconstruction of an information transmission system and evaluation system that can truly benefit those who are “truth-seeking and pragmatic”, let energy flow to strategic points, and effectively curb entropy growth.
Finally, I hope Baidu is getting better and better!
