Are you chasing the GMV equivalent in IoT space ?
IoT (Internet of Things) is a buzz word today. Several studies have hyped up the potential of IoT. Some recent projections are humongous — 24 billion IoT devices by 2020 (growing annually at 41% CAGR), 6 trillion of investments in IoT solutions over the next 5 years which will generate 13 trillion of ROI by 2025 with governments and business being the primary driving force (consumers being the third in all respects — devices installed, money spent and ROI achieved). These statistics have been borrowed from here.
While time will tell whether large proliferation of IoT devices indeed will see the light of the day or not. What’s true is that real time monitoring and control that can be enabled today is providing people with data never seen before and hence opening up many new market segments. While platform will play a major role as and when ubiquitous connected devices (with open API for any third party to connect to them — and that itself is a big assumption) become reality, IoT of today is about connecting the existing infrastructure (using external gateways and devices) to cloud.
So what should be the metrics that a startup (or hence an investor or maybe even a customer choosing between two different IoT vendors) should go for?Let me first discuss what are the often reported but not so important metrics in IoT domain. These are equivalent of Gross Merchandise Value (GMV) which all e-commerce companies used to chase until this year (when funding became scarce and people started looking into true business metrics).
Several articles (eg1, eg2) during the second quarter of 2016 have dubbed end of GMV as the metric to follow for e-commerce. So, what are these often reported, but with no real meaning, metrics in IoT space and why they do not make sense? These are:
1. Number of data points collected (or GB/TB of collected data) — One can simply change the rate at which data is collected from each sensor from 1 minute to 1 second and just scale it by 60 without any meaning. Similarly, adding an extra sensor may not give enough RoI but will increase this metric.
2. Average size of the customer — While average transaction size from each customer does make sense but what often gets reported is the total volume of the commodity (that IoT application is trying to optimise) associated with the customer. For example, for an IoT company in energy space, they will often report sum total of the electrical load across the customers (in MegaWatts or GegaWatts). What matters is how many IoT devices are installed to monitor the sum total of the commodity (and hence the value extracted by the customer) and not the total capacity of the commodity. You will get further clarity in this regard in my next point.
Simple metrics, that an IoT company who wants to be true to itself and ensure that they are on the right track, are:
1. Systems deployed vs systems operational at any point of time — If an IoT company can not show you at least a daily report (if not live information) on how many systems are live, and of those that are offline, what could be the potential problem then they can not scale. This basic piece of information is critical to understand what is causing field failures and address them in a timely manner.
2. Innovation trajectory — How much has been the innovation from the first customer to the most recent customer? It is hard to build the perfect solution in the first place. Similarly, it is important to innovate as one faces new challenges.
The innovation path that the company has taken to address all the challenges faced so far will give you a clear picture of how technologically sound the company is. After all, the IoT company of today will also be the analytics company of tomorrow, and has to be deep rooted in technology.
3. Ease of deployment — How quickly can you deploy? Get numbers on how many days it took the company to execute the largest deployment they have done. Get into the details of the challenges faced and what they have learned from those challenges to improve upon the execution time. If you can not execute fast enough you can not grow fast enough
4. Customers’ confidence — If you are selling in a SaaS model, then how many customers are giving you repeat orders or how many are continuing with you beyond the first year is a good indication. If you are selling hardware, then repeat orders and timely payments are good indicators of customers confidence and they should show upward trend.
5. Revenue — If a customer is accruing real benefit from the IoT product or service, it should very well translate into the revenue of the IoT company. Let us take a simple example from the energy segment to which I belong. Often IoT companies in energy report CO2 saved, units (KWh) or percentage of electricity saved etc. Shouldn’t this translate directly to the revenue of the IoT company? Let us do some back of the envelope calculations to show how the numbers just do not add up. If you claim to save 30% energy and you have 100 KW under management then (considering price of electricity at Rs 10 per unit, 50% as average load and customer willing to share 50% of savings), this should translate to the revenue of Rs 54000 per month (100 kWatts * 24Hours * 0.5load * 30days/month * 0.3 saving * Rs10/unit * 0.5saving share) or approx USD 10,000 per annum. If the company has served customers worth overall capacity of 100 MW, then the company should be earning USD 10 millions of annual revenue. Do these numbers in terms of revenue match those reported in terms of load under management and claimed savings?
