How to Market Research
building marketing
Published: 2022-08-12


Subsets of market

  • TAM(Total Available Market)
    • The total market demand for a product or service.
    • It would be effectively impossible to capture the entire TAM, but this is theoretical.
    • (The population who uses hair care products) x (how often they purchase) x (average price of that purchase) = $1bn
  • SAM(Serviceable Available Market)
    • The segment of the TAM that can be realistically targeted by your products and services which is within your geographical reach.
    • (Global Gen-Z population who uses hair care products and can afford yours) x (purchase frequency) x (avg price) = $500mn
  • SOM(Serviceable Obtainable Market/Share of Market)
    • The portion of SAM that you can capture/own.
    • Your SOM can only really be determined after adoption, although you can try to estimate what share of a new market you could take.
    • If your SAM is $10B and your product makes $1B, then your SOM of $1B represents 10% of the market.

Market Validation

All about validating that a market exists for your intended product/service.

How to conduct market validation research

  • Define your TAM, SAM, SOM to the best of your ability
    • Using general customer personas/target customer profiles that you create
    • Market sizing
    • Customer segmentation
  • Use the internet and offline resources to research those three defined markets
    • Learn about their size, demographics, psychographics, and trends in those markets
    • Use Google Trends, Social Listening
    • Market analysis/research firms have great reports. See if you can get hands on those. Some are proprietary but library networks/universities might have subscriptions to those. There are research librarians who can also help.
    • Revise your customer personas/target customer profiles
  • Find people that fall within those personas/profiles and interview them
    • Qualitative: With small number of prospective customers.
      • DO NOT ask hypothetical questions.
      • Only ask questions around their existing and past behaviors.
      • Ask them about what they are seeking
      • Why they are seeking that
      • How they go about finding and obtaining that - map those workflows and user flows.
      • At the end of every interview, ask if they can recommend anyone with the same problem that you can talk with.
    • Quantitative: With a large enough pool. Survey. If no budget for a survey, rely on survey that other people have done. This can go pretty big depending of what results(market sizing/segmentation work) you want and can be expensive. Outsource it if possible.
  • Research the companies, products, services these people mention to you.
    • Are they sustainable businesses?
    • Are they growing businesses?
    • Are they offering a similar experience and benefit that you intend to offer to these people?
    • How do they connect with their niche?
    • What type of content do they create?
    • What key words are they using?

How to use the research

  • Use all of this information to validate that the market exists. Also use this research to refine your company, product, service, strategies, etc.
  • If you have a successful competitor who gives the same exact offering like you are planning to. The competition has already validated the idea for you.

Product Validation

  • This is all about proving that the market wants YOUR product.
  • The goal of Product Validation is reaching Product/Market Fit.
  • It doesn’t mean much to simply validate the market exists.
    • It in no way validates that the market wants your product.
    • It only validates the market wants what other established players are offering.

What is Product Validation

  • Phase1: People in the market simply committing(buying/using) to your product
  • Phase2: Collect hard data on usage and repeat purchases. (requires time)
  • Phase3: Finding P/M Fit. Market loves your product so much they are highly engaged, frequent buyers, buyers become evangelists
    • You can lose P/M Fit after achieving it.
    • You must continually iterate your product and company to give your customers a great experience, benefits, and value.

How to conduct product validation research

  • You can get creative with it and the process can differ from product to product.
  • You basically need to run tests to validate the market wants your product.
  • You need to prove people are willing to buy your product at the price point you set.
  • Collecting their email address is a bonus.

A basic example:

  • Email collection
    • Collecting email addresses barely qualifies as interest, it does not count towards product validation as the commitment is too low.
    • You should only collect email addresses as a secondary step to a proper Product Validation test.
  • Landing page (Field testing, fake product before build)
    • You can make this process look more “correct” by mentioning early access in the landing-page.
    • Create a landing page that tries to sell your product. Make them believe it exists.
    • Have a call to action to buy your product/book a demo. (based on the product)
    • Take them through the check out process.
    • In the checkout page, before they put payment info, hit them with: “We are so sorry! You missed the early access window! Would you like to join the wait list?"
    • Record the effort to buy your product as a metric. Do not actually process/save their data.

Customer Segmentation and User Personas

There are two ways(customer seg, user personas) we can approach understanding our users(marketing+ux). Sometimes, the user and customer can be different.

  • Case of children’s cereal. Parent is customer, child is user.
  • Case of a free app, where the source of revenue is advertisers

The real kicker is combining the two if you have a chance. Segmentation studies will give you a more rigorous view of what your customer base is. Persona creation will elaborate on why/how behaviors occur. Both will “touch” on the others area, but it’s really just about the level of detail each is capable of going into.

Customer Segmentation

- Guides marketing decisions
- Macro, Generic, Demographic Data, Large dataset

See this post on how PMs can make use of segmentation.

Segments are generally developed through big-data analysis and are defined using a mix of these:

  • Geographic Segmentation: based on geographic boundaries
  • Demographic Segmentation: based on gender, age, household income, education and etc.
  • Psychographic Segmentation: based on lifestyle, values, and beliefs
  • Behavioral Segmentation: based on shopping and purchase behaviors.

When coming up with segments without any data-backing, you should only focus on attributes that are meaningful for what you’re developing. More of an art in a way. Some prompts could be such as, what magazines do they read? What trade shows do they go to? Where do they get their news? etc.

Knowing your segments is valuable when you’ll make different experiences or prioritization calls based on the segments.

User Personas

- Guides design/UX decisions
- Micro, Contexual, Fictional, Qualitative, Small dataset

A customer persona is a fictional but fact-based archetype that represents the key traits of some of those customers, based on qualitative data collected through primary research with real people. It’ll have personality of each archetype.

They do not have to representative of all customers but rather represent the indicative emotional and behavioural traits of Some customers.

Determining market size

It’s basically figuring out TAM, SAM and SOM. Results might not be accurate but as a solo builder, it’s nice if you have some idea about the market you’re building for. These can be time based, geography based etc, eg. TAM on an yearly basis, India TAM, Rest of the world TAM.

When determining it, we can either go top-down or bottom up:

  • Top-down analysis: We got data from some research firms that read that they estimate the ridesharing market to be say $100B.
  • Bottom-up analysis: Based on data from this and this article or data source, we estimate there are 1M cars doing ridesharing every year in the US. We will sell each of them a gadget for $10 so our market is $10M if we capture all of it. This is similar to first principles thinking.
# people on the internet suggested few formulas:
- Price x Customer Base = Market Size (if price is 0, customerbase not matter)
- Price x Customer Base x Frequency of Purchase = Market Size

Two primary steps:

  1. Decide what you want to calculate for. Eg. Do we want to calculate the number of cups sold per year? Or how much money is spent by consumers on takeaway coffee? Are we interested in only India or the whole world etc.
  2. Calculate bottom up. Map out the calculation steps.


  • Do not do top-down. Okay, the industry is $100bn, this industry probably will be willing to spend 2% on software products. That makes our TAM is $2bn. Wrong approach. 🚫
  • Don’t use someone else’s market size. Car(₹1L), accessory for Car(₹20). If TAM for Car is 1Cr(1Lx100 Customers), your TAM is ₹2000(₹20x100). 🚫