How to do a good competitive product analysis? What are the common methods?

There are many articles on the Internet about writing product competitive analysis , but there are relatively few articles about operational competitive analysis, so today I would like to share with you my thoughts and experiences in analyzing competitive operations .

The various changes that occur in our APP every day are the best materials for operations learning, especially the changes in competing products, and there are many experiences that are directly related to ourselves that can be learned.

Observing and analyzing competitors can help us understand dynamic changes, market structure, and find segmentation opportunities; gain inspiration, absorb experience, and plan high-quality activities. When competitors come up with killer features or viral activities, we can also follow up quickly and avoid detours when activities are proven to be unsuccessful by competitors.

In order to write a good competitive product operation analysis report, you need to first know what the good standards are.

In my opinion, a good standard is mainly related to the specific purpose of writing a report. Whether the purpose of writing a report can be achieved is the only criterion for measuring whether it is a good report.

For example, your boss asks you to study the user operations of your competitors and write a report:

It could be to find something to learn from, it could be to check out competitors and prepare countermeasures, it could be used as reference data for financing plans, or it could be to find something for you to do because you haven’t been busy with work recently (囧). …

The above-mentioned different purposes require very different analytical reports. An excellent report that meets one purpose may have very low reference value when used for another purpose.

Clarify the purpose of analysis

What are the main purposes and focuses of common competitive product operation analysis?

1. Learn the advantages

You need to think independently, see the underlying essential reasons through superficial appearances, and ask yourself why several times.

2. Expand the market

Before entering this field, I want to see who the competitors are, and focus on judging the overall situation through multi-dimensional horizontal comparison. Potential competitors with overlapping products and user groups should also be taken into consideration.

3. Competitive Strategy

Focus on analyzing advantages and disadvantages, and the satisfaction of needs of segmented groups. The competitive strategy will generally start from the opponent's weaknesses and unmet segmented needs.

4. Preventive measures

The focus is on analyzing the rhythm and types of past operating activities to infer the possible trends of competitors, and this requires regular observation and updates.

5. Used to add points for job hunting

The analysis should be thorough in at least one dimension and may include some analysis and research at the product level, appropriately expressing optimism and love for the product and industry.

Now that the purpose of making this report is clear, we can proceed to more specific operations.

Determine the object of analysis

The method to determine the object of competitive product analysis is to be comprehensive first and then selective:

"First comprehensive" means to find as many competitors and potential competitors as possible in order to gain a comprehensive understanding of the industry. "Later selection" is to focus on key competitors.

If it is a more mainstream and common direct search keyword, more products will come out. Because mainstream products will carefully optimize ASO and try their best to make their products searchable for popular core keywords, it is relatively easy to find them in this aspect.

If it is a new field or industry, you can search on these new product discovery platforms. Product Thunt is the originator of new product exposure platform, where you can see various new products produced every day. Next, owned by 36Kr , is a follower of similar domestic platforms, and more domestic products will be released on this platform. Mind and today are platforms under iFanr and ITjuzi respectively. They are also good and can serve as a supplement.

In addition, we should also look at products that have overlapping user groups. Even if they are not direct competitors, they can be considered for analysis. Just like toothpaste and toothbrushes, they are not competitors, but their user groups are the same, and they are also similar in many marketing strategies and operating methods .

How can I find the products I need comprehensively?

【 Keyword search extension method】

In layman's terms, it means finding other related keywords through the results of a keyword search.

For example:

For example, I tried a field of infants and young children that I am not familiar with. After searching for "infants and young children", I found many related keywords, such as "early education", "prenatal education", "enlightenment", "baby", "baby", "cute baby";

By searching for "early childhood education", we obtained some keywords, such as "reading", "recognizing fruits", "learning numbers", "pinyin", "telling stories", etc.

By searching for "prenatal education", we also discovered keywords such as " mother and baby ", "mommy", "children's songs", "naming", "parent-child education", "complementary food", and "parenting".

By using a keyword, we can extract more relevant keywords from the information we searched, and then continue to search for information using the keywords and extract keywords again, until these keywords can form a relatively complete information map , so that we can have a relatively complete understanding of what is available in this field.

In addition, searching through Sogou WeChat and Moments are both good ways to supplement information. Sometimes you can be surprised to find things that cannot be found through ordinary search engines .

After a comprehensive competitive analysis, you still have to go back to the initial object of competitive analysis to make your choice. If the purpose is to learn from the advantages, you generally choose several leading products among the competitors. If the purpose is to expand the market, you need to select several more for horizontal comparison.

Collect industry information

Check out some analytical articles in this field to learn how experts in this field view the development of the industry, obtain some high-quality opinions and insights, expand your perspective on the problem, and gain some industry data and development forecasts.

This kind of information is generally more common in science and technology media. The following are recommended commonly used websites:

Collect relevant data

Data is generally a company's core secrets and will never be shared in its entirety at any time. So for many people, obtaining data on competing products seems to be a very difficult task.

Although we cannot obtain some core data directly, we can also obtain some related data indirectly.

1. Measure statistics and capture data yourself

By actually experiencing the product and recording the changes in product data, you can infer some of the product data for the entire platform.

For example, if it is a community product, the most core data should be the number of posts, views, replies, online time, etc. The first three data can be manually counted directly through the product, taking 24 hours as a unit, counting the data for a period of time (such as 14 days), and performing average calculations, etc.

Some data can be captured in batches, especially for web products. Some data is explicitly visible, but manual statistics are troublesome. You can ask your development colleagues to write a script to run the statistics.

2. Make inferences using known data

There are some data that can be used to make inferences and estimates by utilizing the relationships between the data.

For example, if we want to know the download volume of a competitor's APP, we can check the download data of some Android markets , such as App Store , 360, Baidu, and Huawei App Market . Assuming that the download ratios of different mobile phones are basically the same, plus the download volume of our own products in different app stores , we can infer the download volume of this competitor's product in app markets that do not have public download data (Apple, Xiaomi).

If you do not have known data for the corresponding product, you can use the mobile phone market share in public reports to make inferences. For example, the mobile terminal model distribution data of WeChat can be used to infer the overall download volume of competing products.

3. Obtain public data reports and article reports

This part can be found in professional data reports from research institutions, such as iResearch, Analysys, 199it , Talkingdata, Penguin Think Tank, izhike, etc. Although most of them require payment, there are also a lot of valuable free content for reference.

If the company is larger, it can still consider paying to buy some data, such as that from Appannie, where you can see more direct data, which is much more accurate than using speculative methods and changes dynamically.

4. Mixing into the seed user group of competing products

The users in the seed user group are relatively active. They can take the initiative to chat with other users, understand users' ideas and needs, and it is easy to communicate with the other party's staff. By mixing in the seed group, they can also learn a lot of the latest developments, data and other information of competitors bit by bit.

Note:

1. Nowadays, there are a lot of data manipulation activities, and a lot of public data is inaccurate. Don’t believe them all, and always be skeptical of the data! We should also be cautious about the conclusions we draw and use other angles to cross-validate if possible.
2. Don’t be too subjective in your judgment. Try to get to the heart of the problem as much as possible and ask yourself several times why. Really? What other reasons are there? Is there any evidence to support this?

Select analysis dimension

Products and operations are closely integrated, and operations need to be based on products. Therefore, I believe that when doing operational competitive product analysis, it is inseparable from the understanding and analysis of the product.

Therefore, when choosing analysis dimensions, it is sometimes necessary to add the product dimension.

Below I list some dimensions of comprehensive, product, and operation categories:

In actual analysis, combined with the purpose of competitive product analysis, select several (not all!!!) as analysis dimensions.

Different dimensions have different importance in different types of products. How do you know which dimensions are more important?

Dimensions related to the core processes of the product will be more important. For example, for e-commerce products, browsing and ordering are the core processes. Therefore, dimensions related to browsing and ordering, such as UV, bounce rate, order rate, SKU, and GMV will definitely be more important in e-commerce products and need to be the focus of analysis.

『Experience Method』

To do analysis, start with experience. When experiencing a product or activity, you can look at it from the perspective of a novice user, a loyal user, and a product designer. Go through the entire process and feel the changes in your mentality during the entire guidance process, including confusion, impulse, irritability, anger, joy, and temptation.

Having a basic understanding of the product through overall experience is the basis for applying all methods.

Start operational analysis

"Comparison Method"

The most important analysis method is the comparison method, because the differences found after comparison can best reveal valuable information.

The comparison can be between product A and product B, or between the same product at different times.

There is a trick to comparing different products, which is to use a table to list the different products and different dimensions, filling them in one by one to avoid omissions, and finally forming a table that looks very intuitive.

Here is a diagram for reference:

(The picture is taken from an analysis report made by the NetEase team a long time ago)

If the table above is expressed in pure text, you can imagine how tiring it would be for both the writer and the reader.

When comparing the same product at different time stages, it is important to consider the background and product characteristics of different time periods and think deeply about the specific reasons for the changes.

I would also like to remind you of a very important operation method, which is to draw a flow chart. When analyzing a product, you must draw a product flow chart. When analyzing operational activities, you must draw an operational flow chart. By drawing out the flow chart, you can see many differences that are not visible on the surface.

If you don’t believe it, you can try to draw the ordering process of three products from JD.com , Meituan and Vipshop , and you will find a lot of unexpected gains.

For the comparison of the same product at different times, let me give you an example of my daily analysis. For example, I want to know which articles on Huxiu’s official account are very popular and which ones are not read in the past month. This will help me understand the preferences of technology reading enthusiasts, which will help me write articles that users like more in the next step.

This is what I did. First, I counted the titles , reading counts, and like counts of 206 articles on Huxiu.com over the past month. Then I looked at the top 20 articles with the highest reading counts and the 10 articles with the lowest reading counts, and analyzed and studied the reasons for the differences.

(Photo: The top 20 articles on Huxiu.com from April 17 to May 17)

[No picture here]

(Figure: The 11 articles with the worst reading numbers on Huxiu.com from April 17 to May 17)

Due to space limitations, I will not go into detail here. A simple analysis of the titles and reading numbers alone can yield two conclusions:

1. Reports and analyses on hot topics such as Papi Jiang , adult products exhibition, Durex Air Cannon, Baidu Putian System, live streaming , etc. are likely to bring higher attention and traffic . What inspired me is that if I have the opportunity to write some in-depth articles based on hot topics, they will have considerable spread. 2. Six of the 10 lowest-read articles are advertisements.

The above analysis is just based on the relationship between the title and the number of readings. If you open each article and conduct the corresponding analysis, you will find many valuable discoveries.

"Testing method"

By performing certain pre-planned actions for testing, valuable conclusions can be obtained based on actual feedback.

For example, if you want to know the content preferences of users of a community product competitor, and want to understand whether the users of this product like practical information , chicken soup, beauties, current affairs or gossip, you can prepare 3 copies of content in each category in advance, select 3 time periods every day, post them at the scheduled time every day, and then draw conclusions based on user feedback.

By doing this, there will be at least 3 copies of each type of content, eliminating some differences in content quality. They will be published in all three time periods, eliminating the differences between time periods. The final effect will be relatively accurate.

The figure below is a content delivery schedule based on the above plan requirements. Content is released according to the scheduled time and the reading volume is counted. In this way, the conclusions drawn based on actual feedback will be more reliable.

Through the final calculated reading volume, it was found that this user group prefers chicken soup and beauties, and is not interested in current affairs.

"Inference"

Based on the existing data and information, certain inferences are made to draw conclusions.

For example, if you want to study the effect of a certain event on a certain public account, before this event, the average number of headline readings of the public account was 1,000 (more than 7 days), and after a certain event, the average number of headline readings increased by 200 (more than 7 days). Based on the public account opening rate generally being around 10%, it can be roughly estimated that the number of fans increased by about 2,000 people due to this event. In reality, it should be less, because the opening rate of new fans will be higher, and there will also be growth brought by non-events during the estimated period, so a rough estimate is that the number of fans increased by 1,000-2,000 people.

Then we saw that the competitor was doing this activity and gave away a 999 yuan mobile phone and 10 movie tickets, which cost about 1,500 yuan. In other words, the average cost of increasing fans for this activity is 0.75-1.5 yuan per person. If you think this cost is acceptable, you can change the gimmick to imitate your competitor, and at the same time do more optimization of conversion rate in various aspects. In this way, you can get a good event planning through reference and optimization.

What should I do if I can’t find some past event information?

Many activities in APPs are mainly promoted through Push or Banner. After the activity ends, it is difficult to find the entrance to this information. Here is a little trick. Generally, more important operational activities will be released on the official account with the same name as the APP. By browsing the historical messages of the official account, you can obtain a lot of past activity operation information.

『Apply tools and models』

Commonly used analysis tools for analyzing problems: 5W2H analysis method, cross-quadrant analysis method, MECE principle, etc.; commonly used methods in market and strategy, such as SWOT analysis, Porter's Five Forces Analysis Model, Porter's Value Chain Analysis Model, SCP analysis model, McKinsey Matrix, etc.

These tool models can be searched, and you can learn them if you are interested.

Tips for writing a good report

1. The report reading experience should be good:

You should fully consider the experience of the person reading the report so that he or she can clearly understand your views and arguments. Generally, before we read a long document carefully, we will quickly scan it first. For a document with a poor reading experience, we will subconsciously think that the quality of this document will not be very high.

There are several specific points to note for a good reading experience:

• a. There should be a table of contents, clear chapter structure, standardized paragraphs, and labels such as "1.1, 1.2".
• b. The layout is clear, the paragraph and line spacing are moderate, and the font size has a sense of hierarchy.
• c. Use both text and pictures, use more charts, and add screenshots for parts that are difficult to describe.

2. There must be a conclusion:

An analysis report is not a novel, it must have a conclusion! You can't let others read your long research paper and then have to rack their brains to guess the conclusion you want to draw.

Even if the object of analysis is relatively complex and it is currently impossible to accurately judge or draw inferences, there is still a "conclusion", that is, "the situation is complicated and requires further study."

The conclusion of the analysis can generally include two parts: one is the analysis and summary of the research object, and the other is the suggestions for one's own actions.

3. Don’t be too subjective in your opinions and descriptions

Analysis that is too subjective is difficult for people to believe and accept. Even if you have subjective feelings, try to analyze what objective reasons caused them.

For example: "I like this interface more" is more subjective. Can you analyze why you like it? Is it more concise? Is the spacing appropriate? Better colors? Cool effect? “

The ordering process is pretty good. What makes it good? Faster loading? Fewer steps? More in line with existing habits? More payment methods supported?

If the subjective description is more complex, it can be replaced by scoring.

For example, if you use words to describe the experience of a certain module, there will be "okay", "very good", "great", "not bad", "awesome", and "superb". It is difficult to make a comparison. Is "awesome" better or "superb" better?

In this case, you can use a score to measure this aspect, such as scoring the experience from 0 to 10 points, so that you can make a more direct comparison between different products relatively objectively.

4. Suggestions for submitting job application reports

If the structure is clear, more words and pages generally mean that you have invested a lot, which will first win you some extra points in terms of attitude.

Writing an overall competitive product analysis is less risky than analyzing only the job-seeking product, because if you only analyze the job-seeking product, it is unlikely to surpass the interviewer in cognition, and highlights are not easy to appear, but flaws are easy to show. It will be easier to stand out if you write it together with the competing products.

It is best to generate an online link for the report, and then use a short link + QR code tool to generate a QR code, which can be embedded in the resume for easy opening by HR.

Final words

In fact, it is not easy to finish reading this article in its entirety, and it is even more difficult to actually start writing it. It may even be painful. This is normal, because after all, you don’t know enough about this industry and you need to consume a lot of brain cells to study and think.

But precisely because of this, writing a competitive product analysis report is a great opportunity to learn and grow. People who don’t want to write or are too lazy to write will gradually fall behind. On the contrary, people who can calm down and write will gradually have differentiated competitiveness in the workplace.

Source: zhaohan