How I Quickly Test and Validate Startup Ideas
Friday, Oct 28th 2011 at 6:04 am — comments
Since I’ve technically been unemployed now for a week, I’ve spent most of that time brainstorming some new ideas and putting them into a Google Spreadsheet.
This week I began going through that list and pulling out the ones I think have a chance and I narrowed a pretty large list down to just a few.
Let’s take one of the examples from my list. It doesn’t have a name, so let’s just call it Patient Connect.
The idea behind Patient Connect is that it integrates with any type of scheduling and patient management software any private medical practice or clinic uses to setup appointments for patients.
The vast majority of private practices and clinics have their receptionist or anyone with some downtime call patients with upcoming appointments to give them a friendly reminder. If everyone is busy, those calls are often not made. I was told a statistic a few years ago by my Dentist, he said in the medical industry, over 50% of missed appointments are simply because the patient forgot about it.
If a patient doesn’t show for an appointment without advanced warning, that is lost time that could have been billed. A few of those dead spots in the schedule could make or break payroll for a small or new clinic.
So the idea behind Patient Connect is that it integrates with whatever software the office is using now to schedule patients and it will automatically send out an SMS, email and robo call reminder to the patient. My idea goes a lot further than that, but it’s not relevant for this article.
So how do I test this idea in the market to see how much traction it could get and to see if people really need this? Using Pay-Per-Click (PPC) campaigns (AdWords, AdCenter, etc). It’s easy, cheap and will only take you about two hours worth of work and between $100-$500.
First step is to create a good landing page for your campaign. If the landing page is terrible, nobody will respond to it and you will be in the same position you’re in now. You could create this by hand if you’re creatively inclined enough to do so, or you could use a tool like Unbounce. I use Unbounce simply because it’s cheap ($25/month) and it allows me to work and iterate faster on multiple projects than if I created them all by hand.
I can’t take full credit for this next step, but I can’t remember who taught me this or where I read it. On your landing page be as brief and concise as possible. Make your message powerful and do not get into the details, just tell them you can solve their problem.
With our example we say “Want to decrease missed appointments by 50%?” and then we blatantly tell them that “We know how!”. It doesn’t matter how, as long as we solve their problem they will be interested and they will convert on this page, unless this isn’t a problem worth solving.
Now that we have a good landing page, head over to Microsoft’s adCenter, not Google’s AdWords, and create an account with them. You read that correctly, stay away from AdWords. You’re bidding against much more competition on AdWords so it will cost you at least double per click for most keywords. There are definitely more eyeballs on Google, but for this project, Yahoo! and Bing will work just fine.
Even though we’re not going to be spending any money with Google today, we’re still going to use their AdWords Keyword Tool. Head over to it and use it to generate a lot of different keywords, hundreds if you can. The longer the tail, the cheaper it will be.
Take that list and import it into adCenter. Now you have a landing page and a bunch of keywords to bid on. My bidding strategy is simple. I set it to $1 max CPC, I turn ads off on the Content Network and I set a maximum daily limit of $20 for the first day. I set the daily limit low on the first day so I can feel it out and see what’s happening without burning through any real cash. If everything looks good after day 1, I will up the balance to $50 and let it run for 2-3 more days depending on how many clicks and conversions I’ve had.
Now you need to build the text ad. The same factors apply here as they did when we built the landing page, but character limits mean you have to be a little more creative. Point out that you have a solution to their problem.
Remember, don’t point out features and don’t explain how you’re going to do what you say you’re going to do. The point is, if you can’t get them to give you their email address for the silver bullet fix to their problem, they simply won’t give you their money for your product.
Setup your analytics tool of choice and start sending traffic to your landing page. Watch it closely, and after a few days, if a high percentage of people converted on your landing page there is probably a market for the problem you’re solving. Your next step should be to repeat this process to test more granular ideas or features of your product to find out exactly those people want.
You also should now have a good amount of email addresses of people who will be your future customers. Contact them and find out what they’re looking for and what they need. Fill their needs and you have a business.
If you weren’t lucky enough to have a high conversion rate you can scrap the idea or rinse and repeat this process and tweak or pivot on the original idea.
This is a process you should not leave out. This process might cost you a weeks worth or time and a few hundred bucks, but in the end, it’s a much better than building a product for 9 months just to find out nobody wants it.
Right now I’m using this method on 3 ideas. One of the ideas I’ve since scrapped and it was the one I was most eager to jump into — now I’m glad I didn’t. I’ll update this post or make a new one with how the others turn out in the end.