Amazon’s refurbishment services: A recipe for disaster

 

Sellers should turn off Amazon’s refurbishment services option ASAP to prevent their sending out adulterated inventory

Imagine that you sold a gorgeous designer sweater on Amazon via FBA. Your buyer wore it, spilled wine on it, let their dog take a nap on it, and returned it.

You likely expect an industrious Amazon fulfillment center employee to open the box, be hit with the scents of wine and dog, notice the red splash of color, and throw the sweater into your unfulfillable inventory. Unfortunately, you may be wrong.

Your inventory is being refurbished, unless you opted out

Amazon’s refurbishment services setting is auto-enabled on all accounts. Unless you turn off the setting, your inventory may be “refurbished” when it is returned to the fulfillment center.

What, exactly, does “refurbished” mean? I’ll let Amazon speak for itself, from the Help Center:

“Depending on the item, FBA can re-tape, re-glue, and re-staple boxes; remove excessive tape, non-product labels, and stickers; and re-box branded and unbranded corrugated boxes. Shoes and apparel refurbishment can include steaming and removing stains and odors.”

Let’s go through these one at a time, shall we?

1. Re-tape, re-glue and re-staple boxes. Isn’t this a formula for Used Sold as New complaints? Are customers blind – so blind that they won’t notice a box has been cobbled back together after being torn open? 
2. Remove excessive tape, non-product labels and stickers. At first glance, this seems reasonable. But unlike careful sellers, who use products like Un-Du and Goo Gone, Amazon FC personnel probably scrape stickers off without a lot of concern about damaging product boxes – or products. Again, this seems like a huge risk factor that could result in complaints for Used Sold as New.
3. Re-box branded and unbranded corrugated boxes. In other words, your branded products can be thrown into generic brown boxes. Then, your customers can think you’re selling them counterfeit products. Awesome.
4. Steaming and removing stains and odors. This cannot be serious. If an item is stained or smells bad, it is no longer new. Period. End of story. No amount of steaming or stain removal will change that. Most confusing about this step? It certainly doesn’t lead to a great customer experience and will almost certainly create complaints against both 3P sellers and Amazon.

You can no longer escape repackaging

In the past, sellers could choose to opt out of Amazon’s repackaging service for FBA returns. No longer.

“FBA repackages your eligible customer returns so that they can be sold as new. The service applies both to retail and FBA return items, and cannot be disabled in your settings.

Each returned unit is evaluated to determine whether it can be repackaged. Repackaging can include replacing the poly bag or bubble wrapping, or re-boxing the item. It can also include repackaging of branded and unbranded boxes, and poly bags.”

This is deeply troubling. It ensures that customers will receive items they believe have been tampered with, are generic, are inauthentic, or are used. And now, thanks to Amazon, there is absolutely no way to turn it off.

Your best strategy? Manage your returns.

Let’s face it – the FBA crew does a lousy job at deciding whether an item should go back until fulfillable. And now, with required repackaging, there’s no way for a seller to know if a returned item has been resold in a condition that would make customers unhappy.

Here are some strategies sellers can use to try and mitigate the damage:

• If you have a small quantity in stock or are nearing the end of your “lot” of inventory, pay attention to repetitive returns for the same ASIN. This could indicate that the same bad-quality item is being resold again and again. At that point, it’s definitely worth recalling your remaining inventory for inspection. Remember, every return is a huge blow to your overall profitability.
• Monitor your feedback, customer comments and product reviews (if applicable) for quality complaints that could be related to repackaging or re-sold returns. Key words might be plain box, generic box, no tags, dirty, worn, used, loose in the box, etc. Again, if this happens repetitively, you may need to remove your inventory for inspection.

How do I find my Amazon Seller ID, Merchant ID and Merchant Token?

 

In my years at Amazon, the recurring frustration I heard from sellers was, “where’s my seller information—and what exactly is my seller information?”

They were frustrated when trying to figure out what exactly it was:

• What is my Seller ID?
• I have a Merchant ID as well?
• What is a Merchant Token?  

Virtually every seller I’ve met agrees that Amazon tends to place critical information in the weirdest places, especially in Seller Central’s maze of pages, tabs, sections, menus, sub-menus, and links.

It is my goal to help make this as easy as possible and reveal the mystery. Let’s talk about Merchant ID, Seller ID, and Merchant Token.

Know It and Find It: The Merchant ID & Seller ID

Your Merchant ID and Seller ID are one and the same. They’re synonymous. They are a combination of letters and numbers created as your primary identifier by Amazon—for Amazon.

Where is the Merchant ID? The first place to find it is in every URL of every product you list. You can also find your Merchant ID in your Amazon storefront URL. Here’s how:

• In Seller Central, go to your Amazon storefront URL by clicking on your storefront link. This can also be done by going to one of the products you are selling.
• Now, click on the line that says “(Seller Name) Storefront,” where you will see this. Click on the storefront link, which takes you to your storefront. There, you see the Merchant ID.

 

• Now, take a look at your URL from this page (below). See the string of letters and numbers after “me=”? That’s your Merchant ID.

   

 

Increase sales with your Merchant ID/Seller ID?

So what can you do with your Merchant ID/Seller ID? Beyond simply knowing it helps you communicate with Amazon seller support, this ID can be leveraged to increase sales.

With it, you can send direct links to your customers. These links give your customers a quick, easy way to your Amazon product page, and it shows in the Buy Box! (even if you DO NOT HAVE the main-page Buy Box on Amazon).

So, how can you do this? You can add your Amazon Seller ID to the end of the URL that you are selling. Here’s how:

• Find the Amazon Product URL link.
• Once you have that link, add the following:

&m=[your merchant ID]

If we use the ID shown above, it looks like this: &m=A1IPKJZDXF9J3.

This is a great way to drive your customers straight to your product and increase sales. As they grow, winning the main page Buy Box is possible, assuming you are competing with other sellers listing on the page.

What is your Merchant Token?

Your merchant token is a unique seller identifier used in Amazon Merchant Transport Utility (AMTU), third-party applications, and XML feeds. They match up products that you upload with your seller account. CAUTION: Be extremely careful with this information and avoid sharing it with others. Persons with this information can make significant changes to your account.

To find your Merchant Token, you can sign into your Seller Account. From the main page on your seller account, go to settings in the top right corner of your page. A gear symbol will represent this. When you hover over the gear, a pop-out window will appear with options; you want to select the Account info link.

 

This will bring you to your Seller Account Information Page. From here, you will see different links. Find “Merchant Token,” which is located in the Business Information box. Click the “Merchant Token” link, and you are done. Here you will be able to find your Token.

In addition to these identifiers, exploring the Account Information tab in Seller Central is always a good exercise. It gives detailed information that impacts your account status, health, primary contacts, and more. Be sure to keep all that information up to date.

Here is a quick look at each:

• Merchant ID: The primary way Amazon identifies your account and storefront
• Seller ID: This is precisely the same as your Merchant ID and used interchangeably with Merchant ID.
• Merchant Token: Your merchant token is your unique seller identifier. It works “behind the scenes with technologies (Amazon Merchant Transport Utility/AMTU, third-party applications, and XML feeds). The token connects products you upload into your seller account.
  

While seemingly basic to know, many sellers still struggle with locating their information and why it’s important to know and use. And when you need an extra hand managing your account, resolving an Amazon policy violation or even making sure your seller information is up to date and  “Amazon friendly,”

2020 Goal: File for Brand Registry on Amazon and Protect Your ASINs

Brand registry on Amazon matters. Here’s why.

Without brand registry on Amazon, your listings can be hijacked. Counterfeiters and other lousy Amazon sellers can harm your brand image and equity. With brand registry, your company can ensure that you can take back some control from the bad guys.

How to structure your Amazon advertising campaign

 ou have a stunning product on Amazon, but how do you make sure customers discover and buy it?

A good way to get started is to get your products on the first page of search results. How? A strong Amazon advertising campaign.

But there’s a problem. Most sellers are involved with other ways to advertise and promote on Amazon. So, how do you ensure your products get the desired reach and sales through ads, and do it with minimum Advertising Cost of Sale? (ACoS)? This is where the importance of structuring your campaigns comes into play.

It starts with setting up the campaign but managing it well is often complicated. Every one of them must be properly organized and structured. A solid, well-built Amazon advertising strategy helps you get noticed and convert potential customers while optimizing your spend. It also makes it a bit easier to manage.

Structuring a PPC campaign involves many things. You always need to follow the rules and technicalities to make your structure workable. You may be dealing with multiple campaigns at a single time. The strategy you adopt for Amazon advertising campaigns will depend on various factors.
Let us explain the strategies with some examples.

Group similar products in your ad campaign

We recommend grouping similar products in a campaign for better management. You can group the products based on factors including product line, profit margins, and product goals.

• Product line: It’s the most convenient way to group your products in a campaign. In such grouping, you add all the products with a common parent ASIN. However, you don’t have to add all the child ASINs. You can choose the top-selling ones or the new additions that you want to promote.
• Profit margin: If products in your PPC campaigns have almost similar margins, it is easier for you to set a bid. For example, if you plan to set aggressive bids on a product with higher margin, your campaign may contain products with higher margins. You can group the products with smaller margins in another campaign.
• Product goals: A product’s goal would depend on several factors like its lifecycle and inventory availability, etc. For example, your bidding strategy may not be the same for a newly launched product and a product already on the market. Also, a new product or a new campaign may not perform as effectively as an already existing product or campaign. It always takes time for a new product to gather information regarding the effective keywords and strategies. Ensure you strategize such campaigns differently.

Optimize your keyword-based campaigns

 

Set up an optimized structure for your keyword-based campaigns. In this case, you need to create two campaigns for your products. One is automatic, and the other is manual. The automatic campaign will help you get the search terms and the most relevant, high-converting, long-tail keywords.

You can use these keywords when you run the manual campaign to improve the efficiency of the campaign. This process is called keyword harvesting.

You can use all three – broad, phrase, and exact keyword-match types in the manual campaign.

Here’s what different match types mean for your understanding:

• Exact match type keywords: The keyword searched by the customer should exactly match your set keyword, word by word. Let’s say your target keyword is “bottle” with an exact match, your ad can appear only for the search term “bottle” or “bottles.”
• Phrase match type keywords: Your products rank for similar phrases when you use this targeting type. For instance, if your targeted keyword is “bottle” with a phrase match, your ad can appear for search terms like “water bottle” and “red water bottle.”
• Broad match type keywords: The keywords should have all the words and components of your set keyword. They can also have several additional words and synonyms. Let’s say, you targeted the keyword “shoes” with a broad match type, your ads could appear for “shoes for men” or “sneakers” as well.

You can also set negative keywords for which you do not want your campaign to perform.

• Negative keywords: You can find negative targeting under the campaign or ad group in the Amazon advertising console. You can set negative keywords for which your ad should not appear using this option. Besides, add non-performing, irrelevant keywords as negative keywords to save your ad spend.

Create audience-based campaigns

If you target an audience instead of keywords, you may run one automatic campaign and three manual campaigns. The three manual campaigns include generic, brand, and competitor.

• The generic campaign may target anything generic. For example, you sell “Fossil watches” on Amazon. You can target audiences who search for “watches” or “watches for women” in generic campaigns.
• Brand campaigns can target audiences who search for watches from your brand. In this case, you can add keywords like “Fossil watches” or “Fossil watches for men” in your campaigns.
• The competitor campaign aims at audiences who search for competitor products. Let’s say you have a competitor named “Seiko,” your competitor campaigns would include keywords like “Seiko watches.”

Ensure you add appropriate negative keywords in the relevant campaigns based on their campaign structure.

Differentiate your campaigns, use a naming convention

You need to differentiate your campaigns by giving them names. Though there are no strict guidelines on naming a campaign on Amazon, it is often recommended to follow a convention that helps you remember the strategy you followed to create the campaign. For example:

• Campaign type_Targeting_Product Name_Product Attributes
• A generic campaign for watches can be SP_Generic_KW_Broad_Watches_Digital.

Budget allocation and optimization

It is crucial for campaigns to run on sufficient budget allocation. If you do not allocate sufficient funds on your campaigns, they will run out of budget and have minimal results. You can also reallocate the budget from existing campaigns to your converting campaigns to run efficiently. You can also use dayparting strategies if you want to efficiently adjust your campaign budgets.

Automating the Amazon advertising campaign optimization

Optimization of ad campaigns is a time-consuming job that involves lots of monitoring and managing mundane activities. This includes monitoring bids, finding and setting relevant keywords, removing non-performing campaigns, etc. This is where automation comes into effect. With the help of automation and optimization tools, you increase speed and accuracy and save time for other productive activities such as managing your accounts and inventory, etc. Automation will help you seamlessly optimize bids, keywords, and campaigns.

 

As shown in the image, you can create various advertising rules to increase visibility, optimize your ACoS, and effectively manage your budget with SellerApp’s advertising automation tool. Setting rules is pretty easy from the SellerApp dashboard by navigating to “Advertising,” then choosing “Automation,” and then “Automation Blueprints.” These blueprints can be customized. Templates are also available. The tool will help you implement a negative keyword strategy, optimize return on investment (ROI), and increase visibility, impressions, and conversions.

Getting started, now what?

Structuring an Amazon campaign involves many factors, such as keyword optimization, suitable budget allocation, creating compelling ad copies, and tracking campaign performance every now and then. SellerApp can help you optimize and track your campaigns with our dedicated Amazon PPC Managed Services. We help creates strategies to help you understand the customer journey for better advertising insights.

Ultimately, running a campaign without the right structure is like driving a car with no clear destination. Know where you’re going and beat your competitors by building and structuring campaigns proven to work.

Wrestling with intellectual property violations on Amazon

 

Why did you receive this intellectual property violation? What do you do now?

No Amazon seller is safe, even those who sell private-label goods. Intellectual property violations can pop up for just about any listing.

You might be thinking, “Amazon is a platform for reselling; how am I violating intellectual property (IP), and why is my ASIN suppressed?  I thought I was protected under the First Sale Doctrine!” 

Unfortunately, the First Sale Doctrine no longer applies on Amazon. The First Sale Doctrine, which is part of U.S. code, states that an “individual who knowingly purchases a copy of a copyrighted work from the copyright holder receives the right to sell, display or otherwise dispose of that particular copy, notwithstanding the interests of the copyright owner.”

But because Amazon owns the platform where products are listed, it is required by law to take down products if rights owners report their intellectual property has been violated. They give the reporting party the benefit of the doubt and leave it to you, the honest, legitimately operating seller, to sort it out. 

What is an intellectual property violation, anyway?

Let’s start by identifying the different types of IP Infringement Amazon recognizes:

1. Trademark infringement. Trademark is a report against sellers who have listed against another entity’s (possibly another seller’s) brand. There are some common mistakes sellers make that result in complaints for trademark. These include listing a generic product against a branded good, incorrectly list replacement parts using the wrong naming convention, or wrongly using the brand name of another product in your listing’s keywords or description. 
2. Counterfeit. Counterfeit IP violations occur when a seller is offering an inauthentic product that does not match the good on the product detail page.  This is different from a standard inauthentic complaint a seller would receive from Seller Performance, since the brand owner is directly making the complaint. 
3. Copyright infringement. These violations can occur for media, such as books or movies. But they can also occur for non-media if the rights owner claims that the listing’s images or text were unlawfully copied. Don’t upload an image to which you don’t have the rights – even if it matches your product 100 percent.
4. Patent infringement. This is a report on a good that too closely resembles another product for which there is an existing patent. Some sellers may receive patent violations if they sourced a product from China, where many manufacturers have no qualms about using another entity’s design. Even if you purchased the violating inventory in good faith, you can’t sell goods that violate existing patents.  

Are all IP violation reports true?

Look at  Amazon’s page on Infringement Reporting https://www.amazon.com/report/infringement. If you read the third paragraph on this web page, you will see that Amazon does not intend or allow its IP reporting process to be used to enforce distribution agreements. Amazon requires “sellers to only list against detail pages that exactly match their items,” and unless your products are incorrectly matched, you may not have done anything wrong.  

Despite these rules, rights owners frequently report sellers purely to control distribution. Whether you are certain of the problem, or you can’t identify any wrongdoing on your part, many sellers are left to wonder, “what am I to do now?”

How can I get my ASIN reinstated?

When it comes to IP issues, to be considered for reinstatement, Amazon generally wants at least one of the following: 

1. An invoice from the brand owner/manufacturer – this legitimizes your inventory and your right to sell their goods
2. A letter of authorization from the brand owner identifying you as a permitted seller of their products
3. A rights owner retraction

In addition, Amazon typically asks for a Plan of Action explaining why the violation occurred, how you’ve remedied the violation, and how you will prevent similar violations in the future. 

For these kinds of appeals, persistence is key. The team that handles IP complaints oftentimes doesn’t accept even ideal documentation the first time around. Be prepared for a long battle, including multiple responses to the “notice dispute” team, as well as executive escalations.

When you have more questions than answers

Riverbend has widespread experience navigating intellectual property violations.  We often hear our clients say:

• “I bought my inventory from a reputable seller. I know it’s not counterfeit. How do I prove it?”
• “The reporting party’s email seems suspicious; this looks more like a competing seller, and I know they won’t retract their complaint. Why does Amazon keep asking me for the same thing?”
• “The violation is only on my Product Policy Compliance scorecard – I didn’t receive a performance notification or an email notifying me of the person who reported me.  What now?”
• “What if I don’t want to sell this ASIN anymore? How do I get it off my scorecard and will this violation cause my account to be suspended?”

Huge ABH Pharma recall snafu leads Amazon to take down wrong supplement ASINs

 

PL sellers who once used ABH Pharma to manufacture their items see whole catalog deactivated

A massive pharma recall in the supplements category has essentially put many Amazon third-party sellers out of business – even though their products should never have been subject to the recall.

ABH Pharma is a contract manufacturer that has worked with hundreds of private-label supplements brands in the past several years. Unfortunately, the company violated good manufacturing practices regulations, resulting in the recall of dietary supplements they created over the last six years.

The consent decree information released by the U.S. Food and Drug Administration (FDA) was sparse. Instead of the typical level of detail such as lot numbers, dates and specific products affected by the recall, only a list of brands was provided.

That’s where the trouble started. Last week, Amazon began suspending some or all ASINs belonging to the brands listed by the FDA. But in some cases, these products were completely unrelated to the recall. In fact, for our clients, the vast majority – or entirety – of ASINs suspended for the recall were never manufactured by ABH Pharma.

The situation is much more dire for these sellers than simply having suspended ASINs. Amazon has sent out emails to customers who ever purchased any of these sellers’ brands, telling them the products were recalled (even if they were not) and offering refunds. Some sellers have already seen automatic refunds topping a half-million dollars.

The problems go even deeper.

Inventory has been thrown into stranded status. And Amazon is sometimes insisting inventory be recalled for inspection – though it seems impossible to actually place the removal order in some cases.

The path to solving this issue is crooked and different for each seller. Multiple Amazon departments are involved, from Recalls and Legal to Seller Support and Credit Ops. There is no simple appeal letter to Seller Performance. There is no common sense being applied internally at Amazon.

CoronaVirus Exploratory Data Analysis

 

In this post we are going to see how to apply EDA in live problems. For that we taken the coronavirus data set from kaggle and going to apply EDA . This post will help the beginners to learn how to apply EDA in data science field.

What is Coronavirus

2019 Novel Coronavirus (2019-nCoV) is a virus (more specifically, a coronavirus) identified as the cause of an outbreak of respiratory illness first detected in Wuhan, China. Early on, many of the patients in the outbreak in Wuhan, China reportedly had some link to a large seafood and animal market, suggesting animal-to-person spread. However, a growing number of patients reportedly have not had exposure to animal markets, indicating person-to-person spread is occurring. At this time, it’s unclear how easily or sustainably this virus is spreading between people - CDC
This dataset has daily level information on the number of affected cases, deaths and recovery from 2019 novel coronavirus.

The data is available from 22 Jan 2020.

Define the Problem

Coronaviruses are a large family of viruses that are common in many different species of animals, including camels, cattle, cats, and bats. Rarely, animal coronaviruses can infect people and then spread between people such as with MERS, SARS, and now with 2019-nCoV.
Outbreaks of novel virus infections among people are always of public health concern. The risk from these outbreaks depends on characteristics of the virus, including whether and how well it spreads between people, the severity of resulting illness, and the medical or other measures available to control the impact of the virus (for example, vaccine or treatment medications).

This is a very serious public health threat. The fact that this virus has caused severe illness and sustained person-to-person spread in China is concerning, but it’s unclear how the situation in the United States will unfold at this time.

The risk to individuals is dependent on exposure. At this time, some people will have an increased risk of infection, for example healthcare workers caring for 2019-nCoV patients and other close contacts. For the general American public, who are unlikely to be exposed to this virus, the immediate health risk from 2019-nCoV is considered low. The goal of the ongoing U.S. public health response is to prevent sustained spread of 2019-nCov in this country.

Precautions

Health authorities and scientists say the same precautions against other viral illnesses can be used: wash your hands frequently, cover up your coughs, try not to touch your face. And anyone who does come down with the virus should be placed in isolation. "Considering that substantial numbers of patients with SARS and MERS were infected in health-care settings", precautions need to be taken to prevent that happening again, the Chinese team warned in The Lancet.

Coronavirus Exploratory Data Analysis

We can explore the analysis of the corona virus affected stats

%matplotlib inline
import matplotlib.pyplot as plt
import pandas as pd
import numpy as np

The data source is from https://www.kaggle.com/sudalairajkumar/novel-corona-virus-2019-dataset

1 2	nCov_df = pd.read_csv('2019_nCoV_data.csv') nCov_df.columns

The output is

Index(['Sno', 'Date', 'Province/State', 'Country', 'Last Update', 'Confirmed',
       'Deaths', 'Recovered'],
      dtype='object')

Column Description

1. Sno - Serial number
2. Date - Date and time of the observation in MM/DD/YYYY HH:MM:SS
3. Province / State - Province or state of the observation (Could be empty when missing)
4. Country - Country of observation
5. Last Update - Time in UTC at which the row is updated for the given province or country. (Not standardised currently. So please clean them before using it)
6. Confirmed - Number of confirmed cases
7. Deaths - Number of deaths
8. Recovered - Number of recovered cases

The sample data are given below,

nCov_df.head()

the output is given below,

 Sno Date Province/State Country Last Update Confirmed Deaths Recovered
0 1 01/22/2020 12:00:00 Anhui China 01/22/2020 12:00:00 1.0 0.0 0.0
1 2 01/22/2020 12:00:00 Beijing China 01/22/2020 12:00:00 14.0 0.0 0.0
2 3 01/22/2020 12:00:00 Chongqing China 01/22/2020 12:00:00 6.0 0.0 0.0
3 4 01/22/2020 12:00:00 Fujian China 01/22/2020 12:00:00 1.0 0.0 0.0
4 5 01/22/2020 12:00:00 Gansu China 01/22/2020 12:00:00 0.0 0.0 0.0

nCov_df.info()

the output is,
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 1199 entries, 0 to 1198
Data columns (total 8 columns):
Sno               1199 non-null int64
Date              1199 non-null object
Province/State    888 non-null object
Country           1199 non-null object
Last Update       1199 non-null object
Confirmed         1199 non-null float64
Deaths            1199 non-null float64
Recovered         1199 non-null float64
dtypes: float64(3), int64(1), object(4)
memory usage: 75.0+ KB

Based on the above information ,The Province/State having some missing values

nCov_df.describe()

the output is,

 Sno Confirmed Deaths Recovered
count 1199.000000 1199.000000 1199.000000 1199.000000
mean 600.000000 276.213511 5.961635 14.617181
std 346.265794 1966.264622 58.082724 103.959136
min 1.000000 0.000000 0.000000 0.000000
25% 300.500000 2.000000 0.000000 0.000000
50% 600.000000 10.000000 0.000000 0.000000
75% 899.500000 82.000000 0.000000 2.000000
max 1199.000000 31728.000000 974.000000 2222.000000

nCov_df[['Confirmed', 'Deaths', 'Recovered']].sum().plot(kind='bar')

The out put of the above line is ,

Observations

1. The data set is contains many countries like China, Japan, US, India and so on.
2. The comparison of confirmed with Recovered, It clearly states that the recovery action from virus is dead slow.
3. The data clearly indicating the spread of virus

Data Clean up

Removing the unwanted columns from the data

nCov_df.drop(['Sno', 'Last Update'], axis=1, inplace=True)

nCov_df.columns

The result is 

Index(['Date', 'Province/State', 'Country', 'Confirmed', 'Deaths',
       'Recovered'],
      dtype='object')

Converted the date data type object into datetime

nCov_df['Date'] = nCov_df['Date'].apply(pd.to_datetime)

nCov_df['Date'].head()

The result is

0   2020-01-22 12:00:00
1   2020-01-22 12:00:00
2   2020-01-22 12:00:00
3   2020-01-22 12:00:00
4   2020-01-22 12:00:00
Name: Date, dtype: datetime64[ns]

Replacing the wrongly mapped country value towards states

nCov_df[nCov_df['Province/State'] == 'Taiwan']['Country'] = 'Taiwan'
nCov_df[nCov_df['Province/State'] == 'Hong Kong']['Country'] = 'Hong Kong'

nCov_df.replace({'Country': 'Mainland China'}, 'China', inplace=True)

Listing all the countries which is affected with corona virus

nCov_df['Country'].unique()

The result is

array(['China', 'US', 'Japan', 'Thailand', 'South Korea',
       'Mainland China', 'Hong Kong', 'Macau', 'Taiwan', 'Singapore',
       'Philippines', 'Malaysia', 'Vietnam', 'Australia', 'Mexico',
       'Brazil', 'France', 'Nepal', 'Canada', 'Cambodia', 'Sri Lanka',
       'Ivory Coast', 'Germany', 'Finland', 'United Arab Emirates',
       'India', 'Italy', 'Sweden', 'Russia', 'Spain', 'UK', 'Belgium',
       'Others'], dtype=object)

Country based virus affected people information

nCov_df.groupby(['Country']).Confirmed.count().reset_index().sort_values(['Country'], ascending = True)

The output is,

 Country Confirmed
0 Australia 56
1 Belgium 7
2 Brazil 1
3 Cambodia 15
4 Canada 38
5 China 618
6 Finland 13
7 France 18
8 Germany 15
9 Hong Kong 19
10 India 12
11 Italy 12
12 Ivory Coast 1
13 Japan 20
14 Macau 19
15 Malaysia 18
16 Mexico 1
17 Nepal 17
18 Others 4
19 Philippines 13
20 Russia 11
21 Singapore 19
22 South Korea 20
23 Spain 11
24 Sri Lanka 15
25 Sweden 11
26 Taiwan 19
27 Thailand 20
28 UK 11
29 US 113
30 United Arab Emirates 13
31 Vietnam 19

Top most Severely affected countries

nCov_df.groupby(['Country']).Confirmed.count().reset_index().sort_values(['Confirmed'], ascending=False).head(10)


The output is,

 Country Confirmed
5 China 618
29 US 113
0 Australia 56
4 Canada 38
13 Japan 20
27 Thailand 20
22 South Korea 20
26 Taiwan 19
21 Singapore 19
14 Macau 19

List all the Provinces/States that were affected with Virus

nCov_df['Province/State'].unique()

The output is

array(['Anhui', 'Beijing', 'Chongqing', 'Fujian', 'Gansu', 'Guangdong',
       'Guangxi', 'Guizhou', 'Hainan', 'Hebei', 'Heilongjiang', 'Henan',
       'Hong Kong', 'Hubei', 'Hunan', 'Inner Mongolia', 'Jiangsu',
       'Jiangxi', 'Jilin', 'Liaoning', 'Macau', 'Ningxia', 'Qinghai',
       'Shaanxi', 'Shandong', 'Shanghai', 'Shanxi', 'Sichuan', 'Taiwan',
       'Tianjin', 'Tibet', 'Washington', 'Xinjiang', 'Yunnan', 'Zhejiang',
       nan, 'Chicago', 'Illinois', 'California', 'Arizona', 'Ontario',
       'New South Wales', 'Victoria', 'Bavaria', 'British Columbia',
       'Queensland', 'Chicago, IL', 'South Australia', 'Boston, MA',
       'Los Angeles, CA', 'Orange, CA', 'Santa Clara, CA', 'Seattle, WA',
       'Tempe, AZ', 'Toronto, ON', 'San Benito, CA', 'London, ON',
       'Madison, WI', 'Cruise Ship', 'Diamond Princess cruise ship'],
      dtype=object)

Impact in india

nCov_df[nCov_df.Country == 'India']

The output is,

 Date Province/State Country Confirmed Deaths Recovered
432 2020-01-30 21:30:00 NaN India 1.0 0.0 0.0
491 2020-01-31 19:00:00 NaN India 1.0 0.0 0.0
552 2020-02-01 23:00:00 NaN India 1.0 0.0 0.0
611 2020-02-02 21:00:00 NaN India 2.0 0.0 0.0
675 2020-02-03 21:40:00 NaN India 3.0 0.0 0.0
745 2020-02-04 22:00:00 NaN India 3.0 0.0 0.0
815 2020-02-05 12:20:00 NaN India 3.0 0.0 0.0
885 2020-02-06 20:05:00 NaN India 3.0 0.0 0.0
958 2020-02-07 20:24:00 NaN India 3.0 0.0 0.0
1030 2020-02-08 23:04:00 NaN India 3.0 0.0 0.0
1102 2020-02-09 23:20:00 NaN India 3.0 0.0 0.0
1175 2020-02-10 19:30:00 NaN India 3.0 0.0 0.0

Country most affected

nCov_df.groupby(['Country']).Confirmed.max().reset_index().sort_values(['Confirmed'], ascending=False).head(20).plot(x='Country',
                                                                                                                      kind='bar', figsize=(12,6))

Country most recovered

nCov_df.groupby(['Country']).Recovered.max().reset_index().sort_values(['Recovered'], ascending=False).head(20).plot(x='Country',
                                                                                                                      kind='bar', figsize=(12,6))

Country faced more deaths over the world

nCov_df.groupby(['Country']).Deaths.max().reset_index().sort_values(['Deaths'], ascending=False).head(20).plot(x='Country',
                                                                                                                      kind='bar', figsize=(12,6))

Recovery vs Deaths in world wide

nCov_df[['Country', 'Deaths', 'Recovered']].groupby('Country').max().plot(kind='bar', figsize=(12, 7))

Recovery vs Deaths in world wide other than China

nCov_df[nCov_df['Country'] != 'China'][['Country', 'Deaths', 'Recovered']].groupby('Country').max().plot(kind='bar', figsize=(12, 7))

Philippines clearly show that the no recovered happen

nCov_df[nCov_df['Country'] == 'Philippines'][['Country', 'Confirmed', 'Deaths', 'Recovered']].groupby('Country').max().plot(kind='bar')

When did Virus Confirmed initially?

nCov_df['Date'].min()

The result is,

Timestamp('2020-01-22 12:00:00')

When was the Virus Confirmed recently?

nCov_df['Date'].max()

The output is,

Timestamp('2020-02-10 19:30:00')

How many total no.of persons were identified with Virus on each day

nCov_df.groupby('Date')[['Confirmed', 'Deaths', 'Recovered']].max().reset_index()

The output is,

Date Confirmed Deaths Recovered
0 2020-01-22 12:00:00 444.0 0.0 0.0
1 2020-01-23 12:00:00 444.0 17.0 28.0
2 2020-01-24 12:00:00 549.0 24.0 31.0
3 2020-01-25 22:00:00 1052.0 52.0 42.0
4 2020-01-26 23:00:00 1423.0 76.0 44.0
5 2020-01-27 20:30:00 2714.0 100.0 47.0
6 2020-01-28 23:00:00 3554.0 125.0 80.0
7 2020-01-29 21:00:00 4586.0 162.0 90.0
8 2020-01-30 21:30:00 5806.0 204.0 116.0
9 2020-01-31 19:00:00 7153.0 249.0 169.0
10 2020-02-01 23:00:00 9074.0 294.0 215.0
11 2020-02-02 21:00:00 11177.0 350.0 295.0
12 2020-02-03 21:40:00 13522.0 414.0 396.0
13 2020-02-04 22:00:00 16678.0 479.0 522.0
14 2020-02-05 12:20:00 16678.0 479.0 538.0
15 2020-02-06 20:05:00 22112.0 618.0 817.0
16 2020-02-07 20:24:00 22112.0 618.0 867.0
17 2020-02-08 23:04:00 27100.0 780.0 1440.0
18 2020-02-09 23:20:00 29631.0 871.0 1795.0
19 2020-02-10 19:30:00 31728.0 974.0 2222.0

Case confirmed for each countries

nCov_df.groupby(['Country']).Confirmed.max().reset_index().plot(x='Country', kind='bar', figsize=(10,6))

Case confirmed other than China

nCov_df[nCov_df['Country'] != 'China'].groupby(['Country']).Confirmed.max().reset_index().plot(x='Country', kind='bar', figsize=(10,6))

The virus spreadness over the confirmed, Deaths and Recovered in globally

nCov_df.groupby('Date')[['Confirmed', 'Deaths', 'Recovered']].max().reset_index().plot(x='Date',
                                                                                      y=['Confirmed', 'Deaths', 'Recovered'],
                                                                                      figsize=(12, 7))

Spreadness of virus , Deaths and recovery data other than China

List the States in China which were affected

nCov_df[nCov_df['Country'] == 'China'].groupby('Province/State')[['Confirmed']].count().reset_index().plot(x='Province/State',
                                                                                      y=['Confirmed'],kind='bar',
                                                                                      figsize=(12, 7))

nCov_df[nCov_df.Country == 'China'][['Province/State', 'Deaths', 'Recovered']].groupby('Province/State').max().plot(kind='bar',
                                                                                                                   figsize=(12, 7))

Countries those have worst recovery services

nCov_df[nCov_df['Recovered'] < nCov_df['Deaths']][['Country', 'Confirmed', 'Deaths', 'Recovered']].groupby('Country').max().plot(kind='bar')

Countries death rate high and 0 recovery rate

nCov_df[(nCov_df['Recovered'] < nCov_df['Deaths'])&(nCov_df['Country'] != 'China')][['Country', 'Confirmed', 'Deaths', 'Recovered']].groupby('Country').max().plot(kind='bar',
figsize=(12,7))

In those countries are having the more infacted people along with hign deaths. There is no recovery happened.

Very slow recovery in china

nCov_df[(nCov_df['Country'] == 'China') & (nCov_df['Recovered'] == 0 )&( nCov_df['Deaths'] != 0)][['Province/State', 'Confirmed', 'Deaths', 'Recovered']].groupby('Province/State').max().plot(kind='bar',
figsize=(12, 7))

From all the above information the recovery rate is very slow and the death rate is increasing day by day.  Many of states are strugling to recover their own citizen. The goodness is India protecting their people well compared with china's other neighbours.