avatar

Michael Tapson

Technology Leader

Sample Sentry

project project project project project

Challenge

The "sample closet" is a term used in the Pharmaceutical and Healthcare industry to refer to anywhere in a healthcare provider’s (HCP) office where product samples are stored. The sample "closet" may not actually be a "closet" at all. In some situations, it may be a desk drawer, or another physical location in an HCP’s office.

Samples that are stocked in the sample closet are primarily distributed to HCPs via sales reps during scheduled visits. Bear in mind, these visits are often viewed as unwelcome by physicians unless they need samples. In areas that lack coverage from the sales force, samples may be requested via Fax or the Internet. Once samples are in the closet, there is no good mechanism to track inventory levels, expiration and intelligently execute rules based on location inventory. Furthermore, it is not uncommon for products in the closet to expire, be misplaced, or even taken home by office staff.

Solving the Problem

Like most inventions, Sample Sentry was attempting to solve a pain point. Once the initial idea was conceived, there were numerous issues that were technical and strategic in nature.

I needed to solve for:

  1. How do you reliably track samples in a wide variety of physical spaces without physically altering the appearance of the sample boxes?
  2. How do you make installation of the solution as easy as possible?
  3. What's the benefit to Pharmaceutical companies that deploy the product?
  4. Can the benefit to the healthcare provider be substantial enough to overcome the objection of placing a device in their office?

Solving the strategic issues first allowed me to justify the effort involved a technical proof of concept (PoC). The benefit to the pharmaceutical industry was clear:

  1. A sample tracking solution would give sales teams insight in to stock levels to justify an in-person visit.
  2. In hard reach offices, samples could be automatically shipped when sample levels are low.
  3. A tracking solution would give a pharma company much better insight into where their samples are going, how quickly they are being consumed, and data that would enable a analysis on the true ROI of samples at the individual HCP level.

The value proposition needed to overcome the objection of placing a device in the HCP's office proved to be a bit tougher. With some research and interviewing some physicians, I uncovered a few requirements that physicians demanded in order to consider putting this kind of solution in their office.

  1. Installation had to be dead simple.
  2. The solution needed to provide physicians with insight into stock levels and expirations in order to reduce risk.
  3. The solution needed to improve the quality of in-office visits by reps.

After developing a clear set of business requirements, I defined prioritized viability criteria for the technical proof of concept (PoC). These criteria established that the PoC must be battery only, must remain functional for no less that 6 months on a charge, must be able to accurately track at least twenty samples, must be no larger than a defined size, and the solution must not require any physical modification of the sample container.

Once the viability criteria were defined, I created an RFID-based PoC. It was quickly evident that unpowered RFID lacked the range necessary to accurately track numerous samples and a powered solution would not be able to meet the size criteria. The next PoC utilized Bluetooth Low Energy (BLE) "tags" and proved to be a viable path. Since the solution needed to require minimal setup, the PoC also utilized a cellular radio to transmit data. In order to overcome the power criteria, the PoC could not continually stream data. Energy consumption was calculated and programming refinements were made to the PoC to only transmit status at a specified interval.

Once the hardware PoC was in a good spot, we developed a lightweight web service API that received the data from the transmitter and also made the data available to the graphical web interface. The PoC web interface was developed in parallel with with the API as a way to demonstrate that the hardware and API were functioning as expected. The UI displayed product stock levels, device power and GPS location of the deployed device.

The complete PoC took about 80 hours to complete. Presenting the solution to new and existing clients bolstered the agency's perception as a leader in innovation.

Skills Utilized