I’m Keith Resar and this is the beginning of a journey that is more public (or published at least) than I would typically take when developing something new.
That said, the adage that ideas are nothing and execution is everything really is true, so I’m glad you’re going to be a part of this.
I’m well prepared to succeed with Calenzen. Let’s start with a brief rundown of a few recent projects I’ve run, where you’re sure to find a mix of success and failure.
Dropshipping E-Commerce Retail
My background is purely in technology, which makes the fact that my first successful efforts were in retail a bit ironic. Maybe even sad.
I spent two years selling international edition textbooks to college students across the country. My international co-founder would FedEx books from Malaysia and they would arrive on my customers’ doorsteps two days later. Even with the cost to ship heavy paper and our healthy margins, we kept prices $50 to $75 per copy lower than anything purchased domestically. As a side benefit for customers, since the books were typically soft cover, they were easier to carry between classes and the library!
Software engineering had a place in this effort nonetheless. I built extensive automation to connect marketplaces, payment providers, and ship/email notifications to make the entire process almost touch-less.
Over time the inventory began to dry up and we shut down the enterprise.
Take a deep dive into this entire operation by reading my how I built it blog entry.
Custom E-Commerce Retail
Hot on the tail of a previously successful e-commerce operation I got into the custom manufacturing space. The switch from relatively high-cost textbooks to a few pieces of plastic dramatically reduced the risk exposure stemming from customer satisfaction issues that always seem to pop up. Instead of losing $50 or more with a shipping problem, my actual cost was well under a dollar.
As an avid cyclist, I owned one of the first GPS bike-mount computers. It was my stroke of luck that there was a fatal design flaw in the flimsy plastic used to mount these to the bike – and replacements were in the $15 or higher range.
After noticing this, I contracted with a Wisconsin-based engineer to model a prototype replacement bike-mount using SolidWorks 3D CAD, then solicited bids from a dozen different manufacturers via Alibaba.
Within two months, I had 10,000 rubber and plastic kits sent to my home and was selling hundreds every month through Amazon, my own e-commerce site, and through eBay.
The sales were so successful that we ran through a second 10,000 unit order before the GPS model was old enough to be out of fashion.
Next up was a purely technology focussed SaaS side project called Zmonitors. This set of tooling would monitor servers and applications for performance and generate alerts when they went down.
My day job was in IT Outsourcing (ITO) at the time and I spent a lot of effort defining the contractual language around availability SLAs for the services we hosted (e.g. 99.95% available). I saw a problem in how best to monitor and report on availability as a neutral third party. Luckily it also seemed like a really fun challenge!
The tech architecture was pretty amazing, for the time. High availability and distributed hosting across multiple regions and providers was clearly a requirement to get accurate monitoring data.
To add more value to the offering, I pioneered moving from simple page loads to rendering using the first headless browsers. This enabled me to capture screenshots to better document changes.
The customer visible side of the web site was incredibly powerful… and complicated. I developed templating language so customers could create the exact monitoring scenarios and pixel perfect customized reports.
Regrettably, I was so focussed on creating a powerful reporting tool that I lost sight of the level of knowledge needed for someone to successfully learn to use the tool. The target persona was someone who would be hands-off on technology and this solution was the exact opposite.
Email Address Formats
The idea behind the email-format.com project came after a job interview. I wanted to send the hiring manager a follow-up email but I never got his address. What I needed was a web site that showed the email address format in use at every company.
The initial implementation combined a job search plugin called Job-o-matic with API access to Yahoo.com search data. With every visitor, the app automatically generated a list of company names thanks to Job-o-matic’s advertising engine, then used my limited API credits to search for formats.
After almost two years, traffic increased enough that Adsense advertising was actually cutting a monthly check.
The next phase was to dramatically increase the number of companies represented, which was pretty easy thanks to a few awesome resources:
The OpenCrawl database hosted by AWS (a full list of all crawled web page content across the Internet)
A list of voter registration polls that included millions of first and surnames This didn’t seem like it should be public information but it was the data set I needed (Thanks Ohio!)
EC2 spot instances on AWS (What a broken marketplace – the cost for hundreds of machines was literally pennies.)
I extracted what appeared to be email addresses from the OpenCrawl dataset, then grouped these by company domains. Each grouping went through an ML engine seeded with the list of first and surnames trying to determine what email address combination I was likely seeing (first.last, first initial then last name, etc.).
The long tail on this new set of data was amazing and the search engine traffic skyrocketed, along with Adsense revenue. Additionally, the data was so valuable that my revenue model expanded beyond advertising into monthly access fees and metered API usage.
This disappointing effort failed twice, each time under a different name. The idea was to create a localized marketplace for used children’s toys – think Craigslist for Toys.
Buyers on the site would register their kids gender, birth year, and address. Sellers had a number of easy ways to list their offerings including photos tagged with many items.
Item data was expanded using Amazon hosted product APIs so I could get authoritative pictures, names, descriptions, appropriate ages, and other tagged attributes.
Buyer search results were visible based on proximity and their children’s ages.
Regrettably, starting a marketplace is always a challenge, but it wasn’t my only problem. Even after seeding the marketplace with items and reaching out to daycares and preschools, I still didn’t have any buyers.
Like all my projects, Appointment.one came from a personal need. I was in sales at the time and coordinating meeting times with over a dozen customers every week was an awful experience for everyone.
A typical Monday might see me leaving voicemails and emails that offered every prospect two or three windows to meet. By halfway through the day, my entire week was “soft-booked” with potential meetings.
It didn’t take long to discover services like Calendly which helped expose my calendar to make booking meetings easy. That said, I found a handful of core features lacking and decided to build the perfect solution.
The end product, called Appointment.one, was light years ahead of other solutions. It supports more calendaring systems, interesting ways of combining availability across teams, and a simple visual calendar for booking appointments.
My newest effort and its success is yet to be determined. Is everyone else in the corporate world fed up with dialing long and confusing conference bridge numbers?
Will I fall prey to the same mistakes I’ve made in the past? How much effort will be devoted to tech eye candy at the expense of market validation and pounding the pavement for new customers?
Follow along on the journey as I implement and market Calenzen.