In January 2020, we met with about 45 companies. This enabled us to refine our vision (3 pivots in that month ), understand our go-to-market, and start to build a list of potential clients. Once we felt confident about the product, we entered into full implementation mode. It took 5 weeks to build the first alpha version of the product and 6 weeks to have our first paying customer. This article’s goal is to shed some light on what happened during these 6 weeks, and to take a step back on what we did right and wrong.
Continue to meet with other customers to identify patterns When you start to build your product, you want to have as much uninterrupted maker time as possible. You want to give yourself as much chance to get in the zone as possible, so you can build fast.
The issue with this is that you might lose track of the problem you are solving for your customers and whether or not the solution you’re building actually solves this problem for your potential customers.
There is a very high chance that the early feedback you got was unique to the few companies you met with. You absolutely need to meet with several companies to identify patterns and understand which companies land in the general use case and which ones have specific needs. Then, you can decide to use the former as design partners - companies with whom you’ll go back and forth while building your product and that will give you feedback along the way.
Even when you have design partners, you should still continue to meet with other potential customers to ensure that the patterns you are focusing on are the general use case. You might identify several general use cases by doing that.
Now how do you optimize for both maker time and customer meetings? Assign maker time in your calendar first, and leave some blanks for some potential meetings. Try to fill the slots with customer meetings, and if you can’t, that’s just more maker time :).
Never lose sight of the customer’s context One thing we learned while continuing to meet with customers is the context they are in, and your product will be in. Indeed, if you build a B2B company, your product will sit in a stack and will most likely have to integrate well with some tools.
Depending on the stage of the company you’re targeting, the context will be completely different. If you can meet with several companies in different stages, it will help you understand if the first versions of your product will provide a lot of value, and what the requirements should be.
Get intermediary validation with early mockups The first thing we did in early February was to work on the architecture and the mockups of our product. As soon as we had basic mockups, we tried to get feedback on them from customer meetings - but only if the potential customers were experiencing the problem we were trying to solve. We wanted educated feedback on the mockups.
This feedback was really helpful, as it showed us sometimes that we were not focusing on what provided the most value to our customers.
Sell the product immediately, as soon as you have a demo or even mockups (if you can) The best way to gauge the interest of a potential customer is to see if they are actually willing to pay for your product right now. And that means using order forms, and not letters of intent (LOIs).
LOIs aren’t proof of anything I don’t personally consider them a validation of anything. A portion will surely lead to a purchase at one point. But a LOI is non-binding.
It means that even though your product might need the approbation of the security / privacy team, you can get a LOI without going through them. That means you can get LOIs even though your product might never go through the security compliance process successfully. At my former startup, we could get LOIs easily, but we learned over time that our sales cycle was way longer than we would have imagined.
It also means that you can put any pricing in there; it’s non-binding and it doesn’t show that your potential customer is actually willing to pay this amount.
Use binding order forms with conditions to be triggered instead There are several reasons why this model is way better than trying to get LOIs.
As soon as an order form is signed, it means this prospect has become a paying customer. You can have an order form signed without having the features ready. You just need to add triggering conditions to your order form. This will help you align your customers with the essential features they want to see. You should definitely ask whether or not they’re willing to pay without one of the features. It will give you a better idea which feature has the most value in the eyes of your customers. You will be able to really test your pricing with them. Order forms are binding. So you can get a good idea of what your clients are actually willing to pay. Order forms will give you a better idea of your sales cycle as well. Even if you don’t have the product yet, it will go through a company’s standard procedure. Paying customers have a lot more value in the eyes of an investor. They are an actual validation of your product. Here is a great article on the ladder of proof that investors will consider. You can start selling your product with order forms as soon as you can with LOIs. So don’t just settle for LOIs.
Get requirements for the free POC and conditions to get them to pay Sometimes companies won’t be ready to sign an order form directly. They want to make sure your products work before they are willing to talk about pricing. That’s when you want to offer a free proof of concept (POC) - or better yet, a paid one if you can. For enterprises, POCs are the way to go, because they’re used to it.
On our side, Airbyte originally was a new category of product (we enabled you to access your ad-blocked traffic across your analytics and attribution tools, without reenabling ads). So every time we had a potential customer, we included a free POC. What we would do is:
First set up our tool with their Google Analytics to show them that what we were saying was true - you are missing between 20% and 40% of your analytics and attribution data, and we give it back to you. While we were setting it up, we would start discussing which other tools they wanted us to unblock, and would start building the order form for the triggering conditions. We didn’t talk about pricing until it was all set up. Why? Because they had their “aha!” moment when they saw this inbound flow of new data and leads they had not seen before. This was the best leverage for the best negotiations. Don’t wait to have the self-service ready to sell your product At the time of writing, we still don’t have our self-service ready. We have quite a few paying customers that we onboarded manually. You can definitely do that with your customers. You don’t need to have everything ready on your site (settings, Stripe, unsubscription, authentication, etc.) to start selling. You just need the features you’re selling. The rest is just an excuse you give yourself - wrongly! - not to start selling yet. Ok, that might be a bit harsh...there are definitely some exceptions, but you get the mindset.
Hope that was useful. Let me know if you think we could have done differently and better! Always happy to learn!
