Little Known Facts about the Weather Business

Arnold Doray

In Part I we worked out a cost-based benchmark for forecast services. This helped us see that forecast accuracy is the overriding factor to consider if your offshore operations are high value, or cost a lot to repair or (de)mobilize.

In the second part of this series, we will reveal some hidden facts about the weather business and how these directly impact your weather forecast service's accuracy.

Secrets of the Weather Business

Every industry from the manufacture of computer hardware to Mao memorabilia making have secrets that are unknown to the industry outsider.

In each case, the secret involves allowing industry outsiders to believe that "all products are the same" when in fact, important tradeoffs have been made.

For example, did you know that there are 3 classes of hard-drives and computer parts? OEM, OEM Replacement and Retail. The OEM (Original Equipment Manufacturer) parts are the highest quality, followed by OEM-Replacement, and trailing towards the end are those destined for the retail market. You usually can't buy OEM / OEM-Replacement parts on the open market. This is why that PC cobbled together for $42 less than Dell is probably not going to last as long.

Or, if you're thinking of settling in Singapore, and are planning to buy an apartment, did you know that many developers here factor into their prices a 10% to 15% discount for those who ask?

In each case, there is no active deception involved, but by being ignorant of the secret, the industry outsider suffers a higher cost.

The weather business is no different.

Secret #1: Global atmospheric forecasts are 6 to 12 hours old.

Global atmospheric forecasts are the lifeblood of the modern weather industry. They are used to drive wave models, current models, mesoscale atmospheric models, etc., which in turn are used to produce many other products. All global atmospheric forecast data are provided by National Meteorological Agencies.

There are no private sector weather providers that run their own global atmospheric model from scratch. The reason is simple: It takes enormous computing power, a large team of weather experts working around the clock and a great deal of research and background data to get it right.

More pragmatically, much of this data is available for free, or at a reasonable cost. So, weather companies focus on delivering products that meet their client's needs rather than reinventing the wheel.

The problem is that this forecast data is anywhere from 6 to 12 hours old by the time it reaches your weather provider. The problem with old data is that it needs to be checked against reality. An erroneous global atmospheric forecast input will result in incorrect forecasts from other models run by the provider. Garbage In, Garbage Out, as the saying goes.

Responsible providers solve the problem by utilizing multiple global models and selecting the one that best fits the current weather for the client's operating area.

Not all weather providers work this way. Some providers place all their bets on a single model. The forecaster might adjust the few initial forecasts to make them line up with reality. Unfortunately, the quality of the adjustments are greatly dependent on the experience of the forecaster, so your mileage might vary when using this type of service.

The forecasts with the poorest accuracy are probably single source automated services that blindly accept the input global forecast data. There are many such services offering free marine forecasts over the Internet. These are targeted at the recreational sports industry, where occasionally inaccurate forecasts suffice. These services add value in other ways, like easy-to-use graphical displays of forecasts.

So, just like the computer hardware industry, there are also tiers within the weather industry:

TierAccuracy
Multiple models with forecaster selectionHigh
Multiple models, automatedGood - High
Single model, with forecaster adjustmentsGood - High
Single model, automatedPoor - Good
Insight #3: Forecast services are different in the way they handle old data and ambivalent weather.

Savvy offshore operators recognize this difference and what it means for their operations.

Less savvy offshore operators believe weather services from these tiers are "all the same" and select the cheapest service. As we have seen, they might pay a lower price, but suffer higher costs in the long run.

Secret #2: The overloaded forecaster.

The sad reality is that most offshore operators are unaware of this tier, or confuse cost with price and so, select their forecast service based on price alone.

This adversely impacts the bottom line of higher-end weather companies, and they cope in a few ways:

  1. Lower their prices but internally downgrade their accuracy. This usually happens indirectly by asking a forecaster to handle more forecasts to make up for lost revenue. The net result is always lower forecast accuracy. Often, the forecaster may simply "recycle" an old forecast to meet deadlines. If you receive weather forecasts, check for this. It's one sign that forecasters are overworked.

  2. They may "sell pizzas" , that is, make the forecast service a loss leader in order to sell other, more profitable services (like measurement services).

  3. Offer two service levels: one service with forecaster intervention for cost-conscious clients and an automated service for price-conscious ones.

  4. Gradually transform into an automated service provider.

  5. Add value to the forecasts by integrating them with other high barrier-to-entry services (like remote data reception, graphical displays, etc.).

In this author's experience (1) happens most frequently since it is upper management's path of least resistance.

Insight #4: Forecaster overload always leads to a drop in forecast accuracy.

Forecaster overload always results in a drop in forecast accuracy, an increase in operational errors and client neglect (eg. the forecaster forgets the client's operating thresholds and does not provide sufficient warning to the client).

How We Do It

Terra Weather is not immune to these market pressures, but our goal is to be a high-quality forecast provider and to stay that way. We do this by:

  • Utilizing multiple models for all our areas of operations.
  • Setting strict limits on the number of forecasts assigned to each forecaster. This is a decision we made early on, and we find that it translates to very satisfied clients.
  • Investing in technology to give our forecasters better decision making tools (ie, a multiplying effect).
  • Conducting weekly reviews of our operations to improve the forecast process itself. In this way, we have developed software and operational protocols to reduce human error.
  • Investing in R&D to add value to our suite of weather services.
  • Holding training sessions so that experience and knowledge within our team is shared and codified for future use.

In Part III of this series, we'll see how successful offshore companies use these facts to their advantage.

Q & A

We'll answer some questions you might have:

Q: Isn't the price vs. cost argument a cover-up for inefficiency?

The implication in this question is the belief that competition is always good. So if your company can't compete, it must be inefficient.

Competition is good only if "all things are equal". Certainly, IF all forecast service providers are equally accurate, then competition between them on price weeds out the inefficient companies.

That's a big IF (pun intended). As we've discussed at length in this article, forecast services are not all equally accurate. And as we've seen in Part I this means there is a huge discrepancy between cost and price, especially for forecast services.

Due to a lack of information, many offshore operators select their weather services based on price alone. The primary effect is to cause good companies to lower their standards to make up for lost revenue. This in turn lowers the overall profits of offshore operators. It's one of those lose-lose situations.

By focusing on lowering cost, offshore operators can maximize their profits, and competition between weather providers is then beneficial, weeding out inefficient companies. Cost is King.

Related Links


Cost is King

11 Secrets of Successful Offshore Operators

Subscribe to KnowledgeVault

If you'd like to get the latest updates of articles, enter your email below and click "subscribe". We'll send the latest articles to you!