7 AM, you wake up and jump in the shower. You let yourself get in the flow of thinking to startup ideas. Suddenly, it strikes you. An amazing idea. At least, that’s what you think.
You get out of the shower, super excited and confident that you can build this product. You’re sure people want and need this. Right?
You’re so pumped you start building the product, or look for investment or co-founders.
Stop for a minute; we need to talk.
How can you be sure that your crazy idea will work? The truth is, it’s pretty hard. The building part is not the most complex and challenging part, no matter what your idea is. Seriously.
Ok, it’s pretty hard, now what? It turns out that I can help you with the first step of validation. Why just first step? Because there is no real validation other than growing the business for about 2 years and see if you can make it either scale fast or profitable, depending on your goal.
So, how can we get to some level of validation, fast and quite reliably?
If you’re like me and eat Software as a Service (SaaS) for breakfast, lunch, and dinner, you know there are many ways to build a SaaS company.
You can self-fund (aka, bootstrap) or get funding once you’ve got some initial traction. There is The Lean Startup “movement” with its lean canvas. There are all sorts of startup validation tools, and also, some bullshit like “build it and they will come.”
No matter how you fund your SaaS business, there are a couple of things that I can guarantee:
- it’s not gonna happen overnight
- it’ll be a lot of work
- you’ll have to sell something if you want money, so you’ll need something worth paying for
It’s going to be long and you’ll need customers. That’s for sure.
That’s why the generally suggested path to success is to start selling as soon as you can. That is, before even having a finished product.
A lot of people in the industry will tell you not to write a single line of code before starting to sell. I’m one of them. Avoid building something no one wants. That seems to make a lot of sense, right?
You need to validate that people will pay for what you build. The only real validation is money. There are no other validation techniques that are nearly as effective. That’s why I’m going to go through the steps to try to get money from people before building your SaaS. This technique applies to B2B (business-to-business) and I don’t think it would apply for a consumer product (B2C).
I’ll show you 7 specific and detailed steps to get a pretty decent validation before building anything.
Your idea is just a guess
Before diving in the steps to validation, we need to clarify a few things.
You have this amazing idea to build a B2B SaaS and selling it. To do that, you need a problem you want to fix and a solution to that problem. Surprise! Right?
This is commonly called an idea. I call that a guess. This is nothing else than a guess at this point.
You need to identify a problem and its solution, yes, but a huge part of your guess (aka idea) resides in who you think will want to pay $X per month or year to get their problem solved. This is also called "customers". This is what will make you filthy rich.
For example, the initial guess of Metrics Watch was that digital agencies would be willing to pay for Google Analytics alerts, white label them and send them to their clients. Wrong. So wrong.
The point of this post is to help you avoid that kind of mistake and avoid building something people don’t want to pay for.
We’re not going to stick to the guess subject for too long. I’ll assume you have one.
You (think you) know:
- the problem you want to address
- the solution to that problem
- the target customers you want to sell this solution to
- how much you want them to pay for it
This could make a huge blog post itself, but this post is about getting sales before building. Pre-orders, presales, whatever you want to call it. Getting customers before building your B2B SaaS product.
What needs to be validated in your startup idea?
So, you have a guess and a ton of will to take over the world, a market or just have a fun lifestyle business. We want to validate that it’s legit and has some chances to work. Validating your idea is all about risk mitigation.
What needs to be validated exactly?
First, we need to validate that the problem exists and that people are willing to pay for a solution. We also need to make sure that your proposed solution is a good one that people will be motivated enough to use and pay for.
That’s as "simple" as that.
Now, let’s do it!
Step 1: What’s your competition? How are you different?
That might sound counter-intuitive, but looking at your potential competition before starting is great starting point for multiple good reasons.
"Oh, but this is a totally innovative idea, no one has ever done this, I don’t have competitors."
Guess what? You are going to compete for people’s money. Uber had competition in all transportation ways: having a car, taxis, buses, subways, etc. Amazon back in the days? Yeah, maybe not something serious online, but all the bookstores and libraries. Airbnb? Hôtels. Ebay? Newspaper and garage sales.
If you were to create a totally new market, you would still compete for money that’s somewhere else to some extent, plus you would have to create the market, educate people and make them want something they don’t know exists.
You might have a radically new approach, like say Rainforest QA with their on-demand QA-as-a-service, but you still compete with someone.
Constitute a list of your future competitors.
Hint: if you can’t find competitors, there is probably no market and no money. At least, it won’t be easy.
With that list of competitors, find what you do differently. Why would someone buy your solution instead of your competitors that are already on the market and possibly well known already? It needs to be compelling. Usually some key features, or a radically new approach.
They have an unfair advantage vs. you: they are already on the market and have a better understanding of the market than you. Find your unfair advantage.
Why would someone take a Uber instead of a local taxi? Ease of use, relatively exact ETA, simpler payment methods and maybe price for some. That’s compelling.
Why would someone use Metrics Watch for their Google Analytics reporting? It’s simple, directly in your email (no PDF or web-based dashboard) and comes with a concierge service that can help you set everything up, including the choice of data to include in it. Plus, for agencies, it can be white-labeled and have an additional touch of email marketing where you can share news about your agency.
Why would someone use your startup instead of an already established business?
If you can’t find competition, shoot me an email, I’ll find you competition in no time! 😉
Step 2: What’s your pitch?
I’m not talking about a pitch deck and going to pitch to investors here. In a later step, you will have to be able to pitch your product to potential customers so that they buy.
How can you do that?
- find the language your potential customers use and understand
The words you use to describe the problem and the solution might be technically accurate, but it doesn’t mean it resonates with your potential customers.
You can figure out how to pitch to them by looking at the language used by your competitors but also from your potential customers. Where do they hang online? You can look on Quora, Amazon reviews of products in that niche, forums, etc. Go there and look for how people are writing, what language do they communicate with. Ideally, it would be as specific as find people talking about the problem you want to solve but it’s not always easy to be as specific. Do your best.
Next and final step, practice. You should practice a bit before pitching to potential customers. Not too much. You don’t want to spend 10 hours practicing that. Spend maybe an hour or two practicing it alone. It will refine super fast as you speak with potential customers. More on that in later steps.
Step 3: Build a landing page
Next step is generally to build a web presence. Go for something simple.
Don’t spend too much time or money on this. You certainly don’t want to spend thousands of dollars on a designer and someone to build you a site at this point. It could be money down the drain.
Honestly, you could skip this step if you want. I recommend doing it anyway. Why?
Well, you’ll have people that will ask you for a web page to be able to see the details again or some people you will send cold emails to will want to see a page before booking a meeting. More on cold emails and meetings in the next step.
Use the same language you figured out for your pitch. It doesn’t have to be exactly the same pitch, but the vocabulary should be about the same. If you feel like spending $10-$15, you can buy a domain, but you can also use the URL provided by your landing page tool.
This is the point where we take a decision. From here, step 4 could be to get traffic to our landing page & gather email addresses. It could be through blogging, Facebook Ads or Google Adwords, for example. It works pretty well if you know how to do that reliably and fast. Even more so if you already have an audience to sell to.
For most of us, this is just not the case. I prefer to take a different path first that I have more control over.
PS: You should totally gather email addresses and interest from that landing page, but just don’t spend time getting traffic there just yet.
Step 4: Cold email? Yup!
The next step is to talk with people. That’s by far the most effective way to validate your B2B product.
How do you get in touch with those people?
The easiest way is to have a decently sized audience in the space you want to conquer and start talking with them. As very few of us have that, we’ll have to go to the next solution.
The second easiest (and my preferred) way, is to send cold emails to people that you think has this problem and to book meetings with them. I’m not talking about spam here. Don’t get confused between emails sent one by one to individuals you think could be a good fit and could help you validate your guess.
How can we do that in the fastest and most reliable and steady way? Cold emails. Why is reliable and steady? Because you are in control. You are in control of sending X emails per day or per week, to be able to talk with the at least 20 people you don’t know that are potential customers. You decide the pace. Believe me; it’s incredibly effective.
How do you even do that, huh? You need to identify who you think could be interested in your product. Once that’s done, you can cold email them. Easy, right? 😉
There are blog posts and courses dedicated to cold emailing, but I’ll go quickly over the technique that worked for me.
What you want to do is to look for either specific companies you think could be interested in subscribing to your SaaS, or for a role in such businesses.
- Use the search on LinkedIn to find the profiles of the potential buyers you have in mind. If you’re lucky, their work email is on their profile, but this is unlikely.
- Figure out the email address of that person. To do that, the lazy and usually effective way is to install Chrome extensions like Hunter or Prospect.io which will add buttons on the LinkedIn profiles and will give you the email of those people, or at least, possible emails.
- Once you have what you think is their email, if you use Gmail, install Rapportive. Create a draft email and add that email as the recipient. Rapportive should in theory show you a profile in the right bar. If it does, then it’s almost certain that this email is legit. If it doesn’t, you can still use it, but it might not work. If you had a few potential emails, try them all in Rapportive and go with the one that shows a profile.
- Send an email.
Let’s detail this last step a little bit. First, you don’t want to pitch them in the email. Depending on who your targets are, there are different ways to approach them. One way, which I learned from Alex Berman is:
I just found out about YourAwesomeAgency and saw that you did some really good work on the latest Star Wars’ movie website, pretty impressed!
I built a tool that helps agencies like yours close more deals and thought I would reach out as I think it could be a good fit for you.
All the details are here (link to site), but I would love to have the chance to chat with you and get your honest feedback. Are you free for a quick 15-minute call in the next few days?
Look forward to your thoughts!
PS: Fun fact, I also worked for an agency who did a Star Wars’ movie website in the past! 🙂
Let’s go through what this email contains and a couple of rules.
The first sentence is personal. Ideally, it has something flattering. Something you like about their business or the person. It needs to be true; bullshit is going to bite you someday. Ideally, you will want to end the email with a "PS" at the end that is personal too. It can be about something they wrote about on a blog, or a tweet, or something that you have in common. If you don’t find anything there for some people, just skip it.
You then want to give them one or two lines of describing the problem you think they have and your solution. It needs to be short and to the point. The point of this email is to land a quick call.
Another trick for sales email: you always want to finish with a question. Not two, not three, not zero. One question. It forces the conversation to keep going. And that question has to be quite simple, avoid making them think too hard. Make it as frictionless as possible.
Finally, it’s important to know that the hardest reply to get is the first one. After that, people will just reply all the time and rarely stop replying suddenly. To achieve that, you might want to follow up with everyone who didn’t reply a few days or a week later. There are tools to automate that, but you can do it manually for now.
Last time I used this technique was for our reports for agencies and startups, and it worked quite well. I had over 50% reply rate and I think the call booking rate was about 20 or 25%. I was really satisfied with those stats honestly. Don’t expect that necessarily, I remember looking for industry standards and if I recall correctly, I was massively over the industry standards.
Step 5: Booking meetings
Now that you had an email conversation, it’s time to actually book a call.
You want to avoid doing back and forth of "Are you available Wednesday at 10 AM EST?" – "No, but 11:30 AM CST would work" for a few reasons:
- it can take multiple emails, and you have a chance of losing them in the process due to friction
- you need to block that time on your calendar for that person until they reply to avoid double booking, and change it every time you need to try a different time, which means you can’t offer that time slot to other people in the meantime
- this is highly ineffective for both of you
What I found the most effective is to use Calendly or similar tools. It let people choose a time that works for them, in your calendar. Calendly syncs with your calendar, and you can decide when people can book meetings, and it will make sure it doesn’t conflict with other meetings you already have booked. You can even ask people to fill in a few questions that will help you with your call, like maybe the number of employees of the business, or the website of the business, etc. Oh, and if they need to reschedule, they can do it without even talking with you, directly with Calendly. It has a decent free plan.
There are other similar tools to do that, and in fact, Google Calendar has a shitty version of this called "Appointment Slot" but you need to define each time slots manually. You can’t make them recurring, and the UX for your customers isn’t ideal either.
If you sent your calendar link to the person and asked them to pick a time that works for them, make sure they book a meeting. You should follow up with them a few days later if they didn’t do it to see if there is a scheduling problem or whatever and figure out a solution with them.
Step 6: Anatomy of a meeting with a potential customer
So you booked meetings with about 20 people? What’s next? How should you drive those calls?
First, you want to do that over the phone or video conference. I prefer a video conference because you can see the actual physical reaction of people, and that helps a lot in understanding their overall reactions to what you are telling them.
Start by thanking them for their time and give them a quick agenda of the call. Keep the calls short. I aim for less than 10 minutes, sometimes get to 15 and very rarely more than that.
Describe the problem you think they have, ask if that sounds like a problem they have and want to fix. Now is the time to pitch the very high-level version of your product. I tend to keep that to around 1 or 2 minutes.
Answer questions as they come, but don’t ask if people have questions or feedback yet.
Tell them about the price point you planned. Don’t tell them things like "yeah, I was thinking to maybe, I don’t know, $30/month…". No, you need to be convinced your price is right.
Tell the price: "it’s starting at $50/month for 10 reports", then shut up. The only thing you can add here is "What do you think?" or something similar…to invite them to talk. See what comes out from their mouth. Do not ask specific questions; you just want to let them talk.
They’ll react to the pricing and the overall pitch. They’ll ask questions. Note them, and answer those you can right away. Be transparent, be honest. They take some of their time to talk with you, respect that.
Take a lot of notes. You can even record the calls if they agree (don’t record without asking!!!).
You want to note feature requests and most of all, objections. You want to note their reaction when you talk about the problem you’re trying to fix when you pitch them the product and its benefits. You want to note how they reacted when you talked about the pricing. You want to note everything they tell you about.
There’s no such thing as too many notes.
You then tell them about the next steps to get started. Until this point, you don’t have to tell them you pitch something that doesn’t exist yet.
If you don’t have a product, the next step could be asking for money upfront with a money-back guarantee. Honestly, I found that part hard. It’s hard to get people to pay you for a solution that doesn’t exist and that they can’t try. Especially if they don’t know and trust you.
"But JP, I’m an introvert, I think it’s weird and uncomfortable to do those meetings."
At first, the idea of meeting people might feel weird or uncomfortable. Especially for introverts. Remember, they’re just human beings. You don’t force them to have a meeting; they are meeting you because they think it’s worth their time and that you will help them. If you are stressed or uncomfortable for the first few calls, it’s ok. It’ll get better after 5 or 10 calls.
Remember, they are humans, and you just have a chat with them to help them!
Step 7: Interested or invested? (aka, filtering the noise)
How to know if people you are talking to are seriously interested or not really?
You’ll see if it’s a real pain that they are eager to give you money to get it fixed. They won’t find the price too high, or if they do, they will suggest a more decent price for them. After all, you could be wrong with the price.
The thing here is to find whether they are interested or invested. There is a HUGE difference between those, and they can feel similar at first.
Someone interested is not going to be paying you. Someone interested will act a bit like if you were talking about your product to a friend around a drink that finds it cool that you’re doing that and will bounce random ideas. Your job is to detect those people and ignore most if not all of their feedback and feature requests.
The only feedback and feature requests that are important are the ones from paying customers. As you don’t have that (yet), you need to find ways to figure out who will pay for real vs. those who find it cool or say that they will pay but won’t. It’s much easier saying that you will pay than actually getting your credit card out and pay. Beware.
Someone invested is eager to try the product, understands the problem very well and think your approach to fixing it makes sense. Those are the people you want to work with and focus on while you develop your product. This is the kind of people who will thank you for following up or reaching out to them in the first place.
Ideally, you would have hundreds like them. More realistically, you’ll need a good number, maybe 10 or 20 to start with.
Not everyone you will meet will be invested. So you will need to meet a bunch of people before hitting this number.
Once you start to find patterns and people that accept to pay you, tell you they want to pay for that, and you really believe they will as it sounds like a pain for them or like if they found a jewel, you might be onto something worth pursuing.
Now that you spoke with a bunch of people, you should have a good idea if your guess was right or wrong, or, how wrong it was.
What if I was wrong?
Sometimes the product idea is not too bad, but the potential buyers are not the one you thought. Speak with as many people as you need to figure out what people want.
If you realize the original guess is wrong but you want to work with the kind of people you spoke with? Maybe you’re not solving a real problem for them with your initial guess. Find one that is expensive for them and that they want to pay. How? You are already in touch with them; you can just ask them, directly. What the founders of Rainforest QA did, where I was an early employee, was to get in touch with their YC fellow and ask: "What problem would you be willing to pay $1000/month to solve?". That’s one way to do it: just ask.
On the other hand, if you want to solve the problem you think some people have, but the people you talked to doesn’t have that problem, try to find the people that really have this problem if such a thing exists.
Sometimes, it’s also worth just trashing your idea and starting over from scratch. You might have realized it’s not a real issue, or that you don’t want to work on that product for a few years.
Oh yeah, I was right!
For now, let’s say you find a product that solves a problem for people that want to pay to get that problem fixed. What’s next?
Building the product.
You’re not done. You’re soooo not done. You just did the easy part. What’s gonna come from here is a few months to build the product and a couple of years to try to get it "somewhere".
At this point, you probably have a good idea of what product should be built, the features and overall user experience.
My guess is that you probably have an overkill version of what you should be building in mind. Don’t worry; it’s quite common. You need to cut all the nice to have. When you think you’re done removing features, remove more and more features until you just can’t remove anything.
Your goal for the next phase is to build a minimal version of your product to bring value to some people, onboard them, get the bug fixed and start getting money for it.
If you end up building just a subset of the features, which you should totally do, do not lower your price or give access for free (except maybe for a short beta and maybe a trial period). You should give discounts instead until all required features are shipped. Make it clear what the price will be once you’ve shipped X, Y and Z features.
Building the MVP of the product, onboarding your first customers and getting data for future feature interest is a whole different topic.