Customer Discovery — How I used LinkedIn to talk to 100+ people who I didn’t know earlier. You can too.


If you are trying to start a new venture, especially in an area that is new to you, talking to customers and people associated in that ecosystem is actually a must. Also known as customer discovery or market research.

Even though the need for it should be intuitive, most folks — especially engineers like I am (or truthfully was) — usually start building a prototype first. Even though this approach gives instant gratification, it usually results in garage projects and wasted cycles.

Steve Blank first taught me that during my MBA (Haas School of Business) with his words “Get out of the building” i.e., get out of your comfort zone and talk to people who would be your potential customers before you start working on your idea. It made perfect sense at least in theory. But then how to operationalize this.

Google “Customer Discovery” and you will find ton of content (youtube videos, blogs, articles etc.). But most of the advice was about how to set up the process and not how can you effectively connect with people you don’t know and start having conversation.

This post is about how I managed to talk to 100+ almost none of whom I knew beforehand. Some people I talked to were even overseas — UK, UAE, South Africa and India.

Some may be tempted to dismiss market research Jobseque-style; Steve Jobs said in a 1985 Playboy interview (yes Playboy; a fascinating must-read)

“We weren’t going to go out and do market research. We just wanted to build the best thing we could build.”

Above works to the extent that you are already an expert in that area (Jobs was clearly that). When I started down the path of surveying CPQ/Subscription Billing/Quote-to-Cash, I didn’t even understand the terminology well, forget being an expert. So, market research was not just essential, it was existential.

In case, you are wondering how I came up with Quote-to-Cash idea, I didn’t. It came up in a casual conversation while doing market research on another idea. I was told that the subscription billing of that company was broken and that there is a massive opportunity for someone who fixes it. That was a big cue to someone who is hunting for $B ideas.

Trouble was that I didn’t have any background in IT and consequently, my LinkedIn network had almost zero IT contacts. So how do I find 100+ people to talk to? If you are thinking what’s so special about 100, you are right… nothing. You just need to talk to a large set of people — definitely more than 50, and 100 is better so that you can convert at least 20% of them into your own customers in future.

Most founders I have spoken with usually usually talk to 5–20 people and consider it a good proxy. Why? Because connecting and talking to people you don’t know is a painful and time consuming task. If you happen to have a shy personality, the problem is even more compounded.

The challenge with talking to a small group of people is that the initial people you talk to are mostly people you know directly or indirectly. As such, the conversation often ends up being biased (people generally tend to agree with you rather than disagree). Also, in my experience, it was only after 5–10 interviews that I got smarter about asking the right questions and maximizing my learnings in a short 30 minute phone call. Moreover, something interesting happens when you talk to a large number of people — you are likely to find advisors, future employees, customers or maybe even a co-founder. One thing to watch out for with increased # of conversations is confirmation bias. If you want to read more about it, this post does an awesome job in explaining what the bias is and how to avoid it with concrete examples.

Coming back to why the magic 100 number, I believe once you get closer to this number, you have a very deep understanding of the industry/product that you are going to spend your news few years on. Such conversations are hard to get post-funding because then frame of reference changes — people expect you to know more than they do. You don’t want that. Being a student is forgiving and there is no limit what you can ask and learn.

Before I go into the details, few things to highlight:

  • It is a time consuming process (in other words, it requires patience). Contact to conversation ratio was similar to sales-close-ratio (typically 20–30%) i.e., to get to 100 conversations you need to contact at least 300–500 folks. That’s a lot of folks to find and connect to especially if they don’t come from your core network.

It took me roughly roughly 150 hrs for entire exercise.

~100 conversations of roughly 30 mins each = 3000 mins or 50 hrs . Getting to each conversation would require roughly 1 hr of additional work => 100 hrs.

  • Never ever spam. Besides being disrespectful, it never works. Why even bother? I never sent an empty LinkedIn connect message (a pet peeve if I receive one). Also, I manually sent all the messages (no automation) and tailored the message based on the audience (more on this later).
  • It costs you (almost) nothing. I used LinkedIn to discover and connect with folks. I sent them LinkedIn connect requests instead of InMails (which require you purchasing LinkedIn Premium). The only challenge was that LinkedIn has a monthly search quota (commercial use limit) and when I hit that, it was a bummer. If you are thinking of creating a new LinkedIn account, it wouldn’t work as the search results are a function of your network. For a brand new account, you won’t be able to see any names etc., I had to pay for one month of LinkedIn Premium to get over that limit (interestingly after the monthly subscription expired, the web version stopped at the limit but unlimited search still worked on mobile). You can also use your spouse’s LinkedIn, which shamelessly I did ;)

Anyways, let’s jump into the details.

Finding who to contact

Keyword-based Search

First, you need to find out the relevant “job titles” that apply to the kind of customer discovery you want to do (Better the search string, better will be quality of search results). This wasn’t an issue with Q2C as the relevant keywords — ‘quote-to-cash’, ‘Q2C’, ‘CPQ’, ‘Subscription Billing’ — gave plenty of search results.

This step was needed for another idea that I was pursuing — Corporate Training i.e., helping HR folks find trainers. Keywords for it were sparse — “HR training”, “corporate training”. I was looking to speak with both trainers and corporate HR folks. Using those keywords, I found out plenty of relevant titles: Elearning program manager, Corporate training instructor, Corporate trainer, LMS specialist, Instructional designer, Instructional design, Elearning designer and developer, Executive coach, Sales education training, Independent training professional, Training manager, People operations, People analyst, L & D Consultant, Learning and development professional, Enterprise learning solutions director, Learning and development manager, etc.

Armed with these titles, the search becomes easy.

Small trick: If I found folks who had significant industry experience, I scrolled down to the sections — “People also viewed” and “Recommendations”. These two often contained a curated set of relevant folks. (Recommendations usually came from peers or bosses so you can go up the food chain as well).

In my search, I rarely asked for any referrals from my existing network for following reasons — it’s not instant and often times we connect to folks we don’t know that well. However, if you are trying to reach to folks higher up in food chain, referrals are kinda needed unless you yourself have some kind of star power. I find it better to talk to such folks later in my research than earlier as you gotta be well-prepared before such conversations. I should add that I managed to talk to one Board Director at a company through a cold LinkedIn connect…it rarely happens though.

Value-chain based Search

You want to make sure that you are talking to all the relevant folks pertaining to the industry/product you are evaluating so that you can get a balanced perspective. Such folks are buyers, users (may not be same as buyers), implementors etc.

For Q2C:

Buyers: Sales (buyer for CPQ), VP Finance (buyer for Billing)

Implementors: IT Sales team (for CPQ), IT Finance team (for Billing), System Integrators

Users: Sales, Sales Ops, Finance, Pricing/Marketing

I quickly realized that talking to System Integrators will be very helpful — they work with multiple clients so their perspective will be interesting. So I doubled down in talking to folks from big firms like Accenture, Deloitte etc., boutique firms like ATG, Cloudely, Simplus and even very small firms (they are very hands-on and can readily give in-the-trenches perspective).

For Corporate Training:

Buyers: HR, Managers (who request training/indirect buyers, training budget could come from their specific department as well)

Implementors: Not applicable

Users: Trainers, Managers (even though employees attend trainings, Managers are looking for RoI on time/money spent), Employees

Talking to competitors

Don’t be shy in reaching out to folks from existing vendors / future competitors. First, there is nothing certain about the idea that you are pursuing and you are in early stages of market research. Also, I found that folks were willing to give you their perspective of the industry without sharing internal details of the company they work for (which you are not looking for anyways). I learnt a lot from such conversations.

Crafting the message

This was the message I first started with:

“Hi <>, Hope you are doing well! I am a Bay area based entrepreneur currently exploring creating a scalable subsc. billing platform with top notch UX/APIs. Given your deep expertise in this vertical, would love to get your quick perspective, if possible. Will mean a lot. Thank you!”

You want to keep the wording at a minimum and Linkedin connect request has a word limit as well (that meant abbreviating ‘subscription’). Also, I felt that I needed to clarify that I am requesting only few mins of their time and hence, “quick perspective”. It usually means a 30 minute phone call and also, the recipient knows that I respect their time.

Even though I got replies with above message, I later removed some unnecessary verbiage e.g., “Hope you..” pleasantry, “Bay area based”. Also, I added more information about the product idea.

Updated version —

“Hi <>, I am an entrepreneur exploring building an enterprise-ready CPQ+subscription billing platform (UX, AI, scalable, easily-extensible) to disrupt middle office which seems to be broken. Given your deep expertise, would love to get your perspective for few mins, if possible. Thank you!”

Results were better with above message.

For people to reply to a cold connect, I think you need to dazzle them with something, so that they feel naturally inclined to connect/talk to you. In my case, it was disrupting middle office tech. Passion is infectious.

I also added some small personalization details which may have worked as well. For example, if I was sending the message to someone who has 1–2 years relevant work experience, I omitted “deep expertise” and instead wrote “experience” instead. Be truthful. If they had worked in the same companies as I did in the past, I’d say

“I am ex-Cisco, ex-Palo Alto Networks”.

If the person is living abroad I’d say

“I am California-based”

If the person is higher up in the food chain, I’d say

“disrupt $10B Middle Office”

I am not sure about how effective each of these were but I like receiving (and, by extension, sending) personalized messages.

I did the entire exercise on my mobile phone. I felt it was much easier and faster than web.

When folks accepted the invite, I sent them following:

“Hi <>, appreciate you accepting my connect request! Please let me know if you have few mins to talk over phone this/next week. Thank you for your time!! <my phone>

I included my phone #so that they can message/WhatsApp me in case they don’t use LinkedIn messaging.

There were several cases where people accepted the invite but didn’t respond to above message. I sent them a follow-on message in a couple of days with some twist —telling them how can the conversation be worth their while as well.

“Hi <>, would love to get your perspective if possible. Also, happy to share what I have learned so far from my <>+ conversations so far.

Some folks replied but some still didn’t.

I did send a follow-on 3rd message to unresponsive folks but it was rare that someone replied after that so eventually I gave up on it.

I noticed another interesting trend — as I continued the activity, the close ratio improved. I attributed this to my better messaging (?) but something interesting was also happening — as I talked to people, I was also building my network in that domain. So the next person I was reaching out to would likely find more mutual connections and intuitively more likely to accept the connect request.

Operationally speaking, big challenge was to manage these conversations on LinkedIn Messaging — it has unfortunately a very basic interface (even web version) with no concept of folders/tags so everything was getting mixed up. Slight respite was “archival” feature using which messages can be moved to a different list. I moved my unresponsive contacts in that list.

LinkedIn Connect vs InMail

I used a few InMails to send messages initially but the response wasn’t great. Furthermore, I realized that I am somehow accustomed to ignoring most InMails I receive as they have been largely taken over by recruiters… I guess it was happening with others as well.

Also, with InMail, the recipient is not connecting with you on LinkedIn. That has two big disadvantages:

  • First, your searches will be impacted i.e., you will likely have a more restricted set of search results. Remember that LinkedIn search results are a factor of your network (and rightfully so) and if people are connecting to you explicitly, they are also implicitly providing you access to their 1st and 2nd degree connections (who likely are working in the same area) who will now start showing up in your search results. This is immensely valuable.
  • Second, when you are reaching out to new folks, they are likely to talk to you if number of mutual connections is high.

So I decided to pivot on LinkedIn Connect. Its free and it worked.

Scheduling Conversation

I wish LinkedIn Messaging had a quick calendar scheduling thing. Otherwise, creating and sending invites was another pain. iOS Calendar doesn’t remember text from prior invites and for some reason neither does Google calendar so it meant either typing everything out or copying it over from somewhere else every time you want to send an invite. Ugggh.

Mostly, folks gave their cell number for me to call. In other cases, I set up a conference call, which I later realized was not needed. Instead, I did this —

“<>, please call Sandeep at <my number>”.

That way they had my phone number and they could text me for any last minute changes instead of hunting my email etc.

How to have the conversation

When someone (who I haven’t met before) reaches out to be for advice, following are the questions in my mind:

Who are you? What’s the context of this call? How can I help you?

I followed the same template now that I was on the other end of the conversation.

First though, I always started with thanking them for their time (I am really appreciative of folks whom you don’t know but are still willing to take out time to speak with you).

“Thank you so much for your time today. I really appreciate you doing this!”

Then, a quick 15 sec self-intro.

“ My background is in enterprise software primarily and I have worked for companies like Juniper, Cisco, Palo Alto Networks in various roles — Engineering, Product Management, Business Development).

Then, broader context.

“I am interested in creating an enterprise software startup. I don’t have any background in Q2C but stumbled across it when someone mentioned that their subscription billing was broken. I talked to more companies and found that it is a consistent theme.

Once you have had more conversations on this topic, you can be more authentic as well in terms of summarizing what you’ve heard. e.g.,

“I have heard that there are issues with product catalog, integration with subscription billing solutions take time due to different data object models, amendments are tough etc.

Then give the floor to them.

Would love to get your perspective on what doesn’t work or works.

Almost 95% of times, you will find the other person agreeing to what you have said and they will mention what their quick perspective is.

If they don’t agree, then politely ask them why they think that way. In my case, one person mentioned that Q2C is convoluted but companies make do by hiring more people / SIs. And then he was silent. It seemed like a quick way to end the call so I asked him if he thought if there was a related big problem in his mind that needed to be solved instead…he stated “its better to disrupt Salesforce by creating a next-gen platform”. I asked him Why? going by 5 Why’s philosophy (must-read if you are not familiar with it) and we ended up talking about it for 20 more mins or so. Point being avoid confirmation bias and try to maximize learning from each and every call. Later, some other folks gave ideas such as CPQ for Professional Services, MDM for Product etc.

For Q2C, here are some generic questions I prepared initially:

  1. What are some of the pain points in your Q2C?
  2. What are you currently using today (CRM/CPQ/Subscription Billing/ERP)?
  3. Did you make the decision to buy them? (for buyers)
  4. If you had to make the decision again, would you do something different?
  5. What are some things that you really like about your existing system?
  6. What do you think vendors are missing providing?

After more calls, I prepared a list of more specific questions:

  1. Do you think the problem is with the product catalog?
  2. How is the renewal/amendment workflow?
  3. How often you are asked to create a new SKU and how much time does it take to create one?
  4. How much time / money is spent on customizing tools?

Make sure to ask for referrals at the end of the call.

“ Would you recommend any other folks I can speak with”.

Also tell them that you are not asking them to refer you personally (which takes time on their part and they just met you), but that you want to know who to reach out to. This helped tremendously in finding the “right” people to speak with and in some cases, when a name was mentioned more than once, then I knew I will be talking to an expert.


This story is published in Noteworthy, where 10,000+ readers come every day to learn about the people & ideas shaping the products we love.

Follow our publication to see more product & design stories featured by the Journal team.

Engg, Product, Marketing, BD | Creator of Leela Kids — World’s First Kids Podcast App used by 100K+ kids and teachers | Host of Customer Support Podcast

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store