Web3 Job for $160K a year! Do your Research!

Become a researcher and get $$$

Web3 Job for $160K a year! Do your Research!

I get asked a fair bit about research and most of what I do is actually research. This may not surprise you but the majority of web3 opportunities and careers require a good research ethic. I have been researching topics and writing now on various topics from architecture, politics, marketing and now crpyto for over 2 decades.

Researching requires patience to read through information like whitepapers and resources. I get asked where I conduct my research and the truth is from practically everywhere. If I don’t know a subject I read and read and read till I start to understand it. There are also different type of research also. I am not going to cover all types of research but the ones I feel are most important in getting a head start.

Its also important to recognise that the First rule of research is you have to have an interest or you don’t deserve to know. You will struggle to get anywhere if you don’t have at least some interest in a subject.

Online search engine research is probably the most common, Using search engines to find information from articles and reputable sites is actually a good start but its also a huge timewaster sifting through the useful stuff. I would also air caution to identifying information researched between primary and secondary sources as they can sometimes have minor discrepancies. Journals are always going to be the best source of information however many of them are subscription locked.

Data Analysis takes longer, and if its a poll, you will probably find more use of information if you ask someone with influence if they would be willing to run a poll for you. Its much easier to get information form an account with 70k followers than from 5k. Unfortunately that is just how things are unless you combine it with a giveaway.

I also monitor social media platforms more so for sentiment and trends but opinions is one area of research that can sometimes require a bit more work to get but is sometimes the most valueable.

Opinion Matters

Specifically, I mostly research opinions which is harder to get because of a couple of reasons. One is actually finding opinions on a particular niche subject is just hard. Second is the value of those opinions because its difficult to actually find good opinions online.

In terms of finding opinions you can find good opinions from individuals by using one of my favourite tests: a controversial politically driven question. Why? Because then you get absolute crack-heads responding, completely off topic with a cult-like ‘our leader can do no wrong sort of way’ and you can instantly strike them off for their opinions for the rest of your existence. The rest is usually valuable.

An example of a test is, Who would be a good Democratic presidential candidate against Trump? I mean just the mention of Trump will send the cult looking for non-existent angles to the question. If the answer goes off topic about how so and so is a lizard from Venus, or that the opposition is part of some lude cabal then it is a good indicator that you are never going to have a constructive conversation with that or those individual ever. Ignore, block if you must and move on and remember my post on people wasting your time.

So Twitter, Reddit and Facebook groups are great places for sources of opinion. I have started using Gigabrain a fair bit as it is great for scanning reddit for you in seconds and also gives you an AI TL;DR, Consideration and best posts on the subjects whilst scanning the whole of reddit. As an example this is the TL;DR and consideration:

Gigabrain

You can then easily scroll through the responses and pick ones you want to refer to and filter through also.

Data Analysis

Data Analysis refers to the use of digital tools and techniques to gather, analyse, and interpret data related to behaviours, market trends, and marketing campaigns. It involves the collection of information from various digital sources such as websites, social media, online surveys, and analytics platforms. This data is then analysed to gain insights into customer preferences, target audience demographics, competitors' strategies, and overall market dynamics.

The purpose is to make data-driven decisions, identify new opportunities, and optimize efforts for better results. If its for a client this helps them to understand decisions and direction much better. There is no point in explaining to a determined client that a mint of 20K NFTs at 8Eth is not going to work if you don’t show research to support this.

It also allows us to understand needs, tailor messages, and reach target audience more effectively. With the constant evolution of digital technologies, data analysis actually plays a crucial role in staying competitive and relevant in today's fast-paced landscape.

In Research also its important to have a structure to ensure your gathered information is more effective and efficient. I follow a simple 5 stage approach which is relatively the norm for researchers:

  1. Defining Objectives: 

    Clearly define the objectives of the research strategy. It is important to also determine what specific insights you want to gain from the research, such as understanding preferences, identifying popular trends, and evaluating competitors' marketing strategies.

  2. Data Collection Plan:

    Create a plan for data collection, specifying the types of data to be gathered and the frequency of data collection. For example, you could set up social media monitoring tools to track mentions and sentiment about a brand and competitors in real-time.

  3. Competitor Analysis: 

    Analyse competitors' strategies, including their social media presence, online advertising campaigns, and website content. Identify strengths and weaknesses in their approaches to learn from their successes and avoid their mistakes. It’s sometimes easier to get an idea of what has gone wrong and add this as a lessons learnt.

  4. Implementation:

    Implement the insights gained from the research to optimise campaigns and offerings. For example, if the research reveals that customers are interested in a particular colour of a product or taste of a product , the company could launch a drink in that colour and taste.

  5. Monitoring and Iteration:

    Research is an ongoing process, so set up a system for continuous monitoring and analysis. Regularly update the strategy based on new insights and changing market dynamics.

Job prospects

Researching gives opportunities into writing. Actually writing newsletters and articles for web3 companies is surprisingly forgiving when it comes to grammar so there are quite a few opportunities. I actually get asked about jobs regularly so will look into how to get this information collated and to subscribers but for now here are some opportunities offered for research specific writers:

  1. Content Writer, Specialist

  2. Content Manager

  3. Copywriter & Ghost-writer

  4. Social Media manager

  5. Educational content specialist

  6. Technical Writer

  7. Researcher content creator

  8. Editor

  9. Journalist

  10. Columnist

As you can see there are lot of opportunities and salaries can average around $75K for a research content creator. I have seen salaries of $40k to $250k, just keep in mind many contracts are short term and not for the year. Also previous experience play a big role in salary expectations. Note also that salaries have come down and from some of the research I have done, a salary in Jan 2023 of $155k is now around $95K just to give an indication of the market. However August is looking promising with an increase in jobs posted.

In terms of tools I also get this asked a fair bit and I think people assume I am using the latest state-of-the-art AI tools…I am but most are not even commercially available yet but nothing compares to a trusted notebook.

I have an A5 and and A4 journal for doodling in but also use Google Keep. Google keep is actually awesome and i use it daily to capture Twitter spaces discussions, meetings, To Do lists and notes on subjects. I have notes dating back years and its great for searching as its all automatically indexed.

Keep exploring and learning,

Chewie (Chewie On This) ✏️

100% written by human

AI Tools to try:

For AI, as mentioned before I am a big fan of Gigabrain, but Redderai is suppose to be good also. I’ve actually used a few AI research platforms but I feel that this area is still evolving and that ChatGPT although very generic is still a good tool for some research. Live data though is still best to use traditional methods of research like news and online publications.

Not AI but useful: Also worth noting i am using Twitter Hunter useful more for politics rather than web3 from what I have been using for but good for social trends. It’s a useful tool that works in parallel to twitter and can be used to search large pools of data including your followers.

Hope that has given some insight into some research techniques and also into some tools and if you have any subjects you would like me to cover feel free to hit me up.

Airdrop Campaign: referral code for the Supra Oracles:

Supra is a team of blockchain experts, cryptography researchers and web3 enthusiast who want to make blockchain better, faster, more secure, cheaper and more accessible.

They have been building over 5 years and they have a number of missions you can complete to get $Supra which will drop once the project goes public. So worth doing the missions to get those $Supra and keep an eye on what it grows into. Soon they are giving away a Toyota Supra so absolutely worth entering that competition also.

NFA / DYOR as always.

Chew on it 😋😋or Spit it out? 🤮🤮

Ask me about something and i will tell you if i would chew on it or spit it out. 😁Simple. (FYI if you send me a DM likelihood is I will add it here 😂 Anonymous ofc unless you ask me).

I am seriously thinking of leaving web3 to get a job outside of web3 and not sure if I will miss the bull run, what do i do?

Anonymous

Chew on it 😋😋

Honestly, get a job! We have no idea what is going to happen. Don’t miss an opportunity to get a job for just in case something happens in web3. You can still be part of communities and be involved in web3 after work and depending on what you do at work you might also be able to attend twitter spaces. Don’t let people persuade to stay in web3 with FOMO, unless you have a guaranteed job opportunity in web3.