Amazon Keyword Search: How do Amazon Keywords?

Key words for many Amazon sellers, even if they are like flies, they will not disappear, and there is also a seller who cannot pursue. Amazon’s seller hopes that your products get certain exposure on the platform, bringing more traffic. Amazon Advertising is the way to promote sales, and the Amazon advertising can pay these specific advertising locations through a specific keyword, so that your products are higher than those in the consumer Amazon’s pre-order goods.

Many sellers know that advertising is the most direct way to get traffic. What do you need to pay attention to during daily operations? How to optimize product keywords? Guan Tong gives you different characteristics from birth to maturity, analyzed at different stages How to optimize advertising keywords. The following technique is a wonderful content you can’t miss.

The life cycle of the product can be roughly divided into three stages: new duration, growth period, stable period, different periods of advertisements in selecting keywords, quotes, advertising indicators, have different focuses.

1. During the new product:

New products online, the advertisement in the station focuses on opening the traffic entry, and the user searching the word accumulated vocabulary through the automatic advertisement. Amazon’s automatic advertisement will match the user search term according to the relevance of Listing, so the measurement phase during the new product is also a timing for checking whether Listing is qualified and optimized. The points executed here are traffic, need to pay attention to exposure, click on data.

2, growth period:

Enter the growth period, advertising target attention to sales volume and conversion. This stage is focused on optimizing the keywords of optimizing advertising. The previous period runs a lot of words in the automatic advertisement, in which the performance-effective vocabulary can be placed in a manual advertisement, and the vocabulary of poor performance is negative. How to choose a good performance, just sorted by exposure, or only sorted by conversion rate may be too film. Filter with exposure, click, fees, click rate, ACOS, etc.

3. Stable period:

The advertisements of mature products pay attention to profits and costs, which is more meticulous at cost control. The product enters the maturity phase, the advertising keyword setting is basically stable, and there are historical sales available for reference. According to the change of sales, a certain proportion of sales will be assigned to the advertising budget. Refer to the recent sales data dynamic to adjust the daily advertising budget, stabilize advertising, and ensure the health operation while ensuring the interest.

The above analyzes the keyword characteristics of three different phases. We also need to pay attention to the setting of daily keyword format.

1 First, on the length, keywords best use phrases and languages, try not to use long sentences, longer keywords, customers are not easy to use, the lower the product exposure

2 Key words arrangement recommendations from top to bottom from top to bottom from top to bottom from top to bottom

3 Don’t repeat keywords;

4 Use a concise keyword, but not the abbreviation of rare purchases may not know

5 Key words need to be combined with data to optimize

6 Key words content order should be arranged in normal order, and the content spell should be correct.

Not only must I learn to write keywords, and the optimization of keyword settings is also what we need to understand and grasp. We can analyze from the following different Amazon Advertising Settings, further improve the keywords, and achieve the increasing amount of orders.

– Product Knowledge Research and Learning: Amazon seller must first understand its own products, fully learn our product’s comprehensive knowledge, initially judge and collect keywords.

– Advertising Frame: Amazon Advertising Framework affects advertising effects, repetitive or internal competition cost invalid costs, and clear frameworks can improve advertising effects. A reasonable grouping of project products, making the advertising group clear structure and convenient future optimization.

– Key words Screening: The key words of Amazon advertisements are not corresponding to the product, directly affecting the conversion. There is a lot of costs in invisible. Amazon seller can collect keywords related to the product in the first two steps, plus the assistance of the Amazon Advertising Keyword Tool. For example, we use the captain BI, choose the right advertising keyword, combined keyword, match the appropriate advertising group.

– Key words offer optimization adjustment: Quotation high and low impact advertising rankings, should pay attention to the keyword rankings, adjust the quotation at any time according to the situation. For poor effects of poor effects or low-low keywords, the ROI should be adjusted according to the situation. For example, a keyword conversion is very high, but the display and click, the ranking is also declining, and a keyword for improving the price optimization is very much, but if there is no conversion, you can consider whether to use this key directly. word. I use the captain BI’s advertisement management software to monitor the advertisement you put into, automatically filter the invalid keyword of this section, and use one-click processing directly. Set the keyword quote with the captain of the captain, save advertising costs according to different quotations in time period.

– Data Statistical Analysis and Product Optimization: Adjust Budget and Products according to the formulated plan and requirements. The news is mainly provided to provide independent data, promptly understand the advertisements, and suspend / price cut / brings, etc. according to the product situation, and optimize the effects of advertising.

The above methods and techniques provide more operation shortcuts for cross-border operations, but because each seller uses these methods to integrate different optimization programs, the seller also needs to try in subsequent daily operations, dare to practice, and constantly put forward new Question, constantly solve questions (article Source:

.NET CRANK Performance Test Getting Started

Crank is a Microsoft’s new performance test framework, integrating a variety of reference test tools, such as Bombardier, WRK, etc.

Crank can be converted into different benchmark test tool commands through a unified configuration.Refer to Bombardier Job implementation.

Install CRANK

Run the following two commands separately install Crand’s CLI (Controller) and Agent.

DOTNET TOOL Update Microsoft.crank.Controller –Version "0.2.0- *" – GLOBAL

DOTNET TOOL Update Microsoft.crank.Agent – Version "0.2.0- *" – Global

Need .NET SDK 5.0 Environment

Need .NET SDK 5.0 Environment

After the installation is complete, execute the command crank, you will print the parameters and descriptions that can be configured as follows, or GitHub View related parameters.

PS C:\Users\Stack\Desktop> crank

Crank Benchmarks Controller

The Crank Controller Orchestrates Benchmark Jobs Oncrank Agents.

Usage: crank [command] [options]


-? | -h | –help show help information

-c | –Config Configuration File or URL


-J | –Job Name of Job To Define

–Profile Profile Name

– Execute a named available in the configuration files. Can be used multiple

-j | –json saves the results as json in the specified file.

–CSV SAVES The Results AS CSV in The Specified File.

–compare an optional filename to compare the results to. can be used Multiple Times.

–Variable variable

–SQL Connection String of The SQL Server Database to Store Results in

–Table Table Name of the SQL Database to Store Results in

–Session a logical identifier to group relatyd jobs.

– Deion a string describing the job.

-p | –property Some Custom Key / Value That Will Be Added To The Results, .e.g. –Property Arch=ARM –Property


Execute a CRANK-Agent to start the proxy service required for the benchmark test. GitHub View related parameters introduction.

PSC:\Users\Stack\Desktop> crank-agent

Hostingenvironment: Production

Contentroot path: C:\Users\Stack\.dotnet ools\.store\microsoft.crank.agent\ 0. 2. 0-alpha. 21567. 1\microsoft.crank.agent\ 0. 2. 0-alpha. 21567. 1 ools
et 5. 0\any\

Nowlistening On: http: // [::]: 5010

Create a CRANK profile

Configuration file reference official Hello.benchmarks.yml example

Bombardier.yml is introduced in the sample file. Because the large environment is not very friendly to the domain name, you can consider downloading Bombardier.yml to your local, imports introduces local paths or directly add file content to the new configuration file.


-https: //

The use of recommended IMPORTS files in production, and a small amount of configuration can be completed in the CRANK configuration file to complete the reference test, and different can be introduced, based on different benchmark test tools.

Other test tool profiles are under


None: ”

Plaintext: ‘- header "Accept: Text / Plain, Text / HTML; Q=0.9, Application / XHTML + XML; Q=0.9, Application / XML; Q=0.8, **; Q=0.8" – HEADER " CONNECTION: Keep-alive "’

Json: ‘- header "Accept: Application / JSON, TEXT / HTML; Q=0.9, Application / XHTML + XML; Q=0.9, Application / XML; Q=0.8, * / *; Q=0.7" – HEADER "Connection: Keep-alive" ‘

ConnectionClose: ‘- Header "Connection: Close"’

PresetHeaders: None




BranchorCommit: main

Project: src / /

SourceKey: Bombardier

Nobuild: True

ReadyStateText: BombardierClient


Connections: 256 # Set the number of connections

WARMUP: 15 # Set the test preheating

DURATION: 15 # Set the test time

Requests: 0 # Set the number of test request instances

Rate: 0 # Sets the request frequency per second

Transport: fasthttp # | http1 | HTTP2 Settings to send HTTP requests with golang’s fashttp library

Serverscheme: http

ServerAddress: Localhost

Serverport: 5000


Bodyfile: # Path or Url for a file to use as the body content

VERB: # get when nothing isoti

CustomHeaders: # list of headers with the format: ‘<name1>: <value1>’, E.g. [‘Content-Type: Application / Json’]

Arguments: "- c {{connection}} – w {{duration}} – n {{requests}} – {{requests}} – INSECURE -L {% if rate!=0%} –rate {{{{{ {% ENDIF%} {% ENDIF%} {{{transport}} {{{%}} {{headers [presetHeaders]}} {% for h in customeraders%} {% assign s=h | split : ‘:’%} – header "{{s [0]}}: {{s [1] | Strip}}" {% endfor} {% if serveruri==blank or serveruri==EMPTY%} { {serverscheme}}: // {{serveraddress}}: {{serverport}} {{path}} {{{{}} {{{}}} {{{}}} {{path}}} {{path}}} { % IF BodyFile!=Blank and bodyfile!=EMPTY%} -f {{% ENDIF%}} {% IF%} {% {%}=blank and verb!=EMPTY%} -m {{verb}} {% ENDIF% "


# – job.timeout=Number (job.variables.duration) + Number (Job.Variables.warmup) + 10;


Source: # Specify the item you need to test, this article uses the local path directly

LocalFolder :.

Project: crank_demo.csproj

ReadyStateText: ApplicationStarted.

# Source: Specify the remote warehouse address of the test item and specify branch via BranchorCommit

# repository:

# branchorcommit: main

# Project: Samples / Hello / Hello.csproj

# ieStateText: Application Started.

Scenarios: # Configure the reference test scene

CRANK_DEMO: # Defines the scheme name and specifies the name when executing the crank command.


Job: Bombardier # Specify Test Tool Bombardier



-http: // localhost: 5010

Start CRANK-Agent

After starting the agent, execute CRANK, there will be an action of installing SDK as follows

[09: 29: 05.261] Runtime: 6.0.0 (CURRENT)

[09: 29: 05.262] SDK: 6.0.100 (CURRENT)

[09: 29: 05.263] ASP.NET: 6.0.0 (Current)


[09: 29: 05.266] Desktop: 6.0.0 (CURRENT)

[09: 29: 05.266] Installingsdk’6.0.100 ‘…

So get a DotNetHome parameter plus the Agent to avoid repeating the SDK.

crank-agent –dotnethome ‘C:\Program Files\dotnet’

启动Crank crank — config.\demo. benchmarks.yml — scenario crank_demo –profile local

–SCENARIO CRANK_DEMO: Specifies the defined test scenario

–Profile Local: Set the output result to the local, namely the console output

You can output to a local JSON file (-output results.json) or database by parameter designation results (–SQL [Connection-String] –Table [Table-Name])

You can output to a local JSON file (-output results.json) or database by parameter designation results (–SQL [Connection-String] –Table [Table-Name])

Here, the log output during the compaction is omitted, and the results are listed directly.

| Application ||

| | | | |

| CPU usage (%) | 56 |

| CORES USAGE (%) | 447 |

| Working Set (MB) | 140 |

| Private Memory (MB) | 157 |

| Build Time (MS) | 7, 232 |

| START TIME (MS) | 501 |

Published Size (KB) | 91, 292 |

| .NET CORE SDK VERSION | 6.0.100 |


| .NET RUNTIME VERSION | 6.0.0 + 4822E3C |

| LOAD ||

| | | | |

| CPU usage (%) | 46 |

| Cores Usage (%) | 370 |

| Working Set (MB) | 29 |

Private Memory (MB) | 30 |

| Build Time (MS) | 11, 891 |

| Start Time (MS) | 226 |

Published Size (KB) | 68, 228 |

| .NET CORE SDK VERSION | 3.1.415 |

| ASP.NET CORE VERSION | 3.1. 21+ 458D974 |

| .NET Runtime Version | 3.1.21 + DF8ABC0 |

| First Request (MS) | 185 |

| Requests | 912,005 |

| Bad Response | 0 |

| Mean Latency (US) | 4,207 |

| Max Latency (US) | 138, 999 |


Crank readme:

Benchmarks Crank Refactoring the Benchmarks

Sample code: