Many B2B publishers are sitting on a gold mine of rich data disguised as dusty old directories, sold to libraries and a dwindling population of industry customers who order more out of habit than need.  Even when those directories are turned into an online application, they rarely deliver on their potential to be significant revenue drivers.  Directories often get moved online in ways that make them hardly more useful than the printed product, with clunky interfaces that discourage users from digging into the information.  That’s a shame, because even a moderately successful data product generates 50% margins.

I can’t tell you what your publication’s gold mine will look like, but I can help you find it.  Here are five basic rules about turning a sleepy directory business into a serious data products business.

Convenience trumps pretty much everything

Raw data is useless to most people.  If somebody needs data structured or presented in a certain way, they will pay for it, even if the raw data is publicly available for free.  It’s even possible to sell data that your publication offers for free in other formats.

For example, if you publish information about monthly factory output for your industry, you may find that companies are willing to pay for that data if you organize it into downloadable spreadsheets that they can plug into their financial systems immediately.  Don’t underestimate the power of convenience.

Mashups can extend the reach of your data

After data has been structured, you can often increase its value by combining it with other data. As long as you make it easy for users to query and visualize the results, you may be able to deliver a profitable data product if you mix it with one or two other layers of data.

For example, the U.S. government offers a growing store of free data that goes far beyond just Census data.  The potential for profitable permutations of your proprietary data layered over some large public data set is pretty huge.

Presentation is important – but don’t overdo it

Data visualizations are all the rage because it’s easier for most people to understand pictures than columns of numbers.  If the visualizations are interactive, that’s even cooler.

However, most data products don’t need a lot of eye candy.  They simply need to allow subscribers to get to what they need easily and quickly.  Spend some effort on professionally designing the user interface of the data product so that people want to use it.

Get the right people

Building a paid data product doesn’t require a genius, but it does require some pretty specific skills.  Here’s what you’ll need:

• An editor or publisher with enough market insight and vision to brainstorm ideas and validate them with readers
• Access to a strong developer with prior experience building data products.
• Access to a designer with strong user interface/user experience chops to ensure your data product can be used by someone other than the developer who built it
• A product/project manager who understands the business goals and the technical solutions and will ensure the final product works as intended, will scale, and will prosper and grow without the developer babysitting it constantly.

Promote it, sell it

Once the data product is live, produce a “screencast” demo that shows prospects just how wonderful and easy it is to use.  You’ll close a lot more sales with this kind of demo, and it’s easy and inexpensive to do.  Here’s a quick screencast how-to demo to give you an idea.  You can spend a bit more time or money to dress it up with some fancy effects or video, but it’s probably not necessary.

You’ll also want to promote the product heavily on your site, in the press, in email, in print and through social media channels – wherever your target audience can be reached. If you publish an article about a business challenge that your data product addresses, don’t forget to link to the demo. Give influential bloggers free access to get their feedback and access to their network.

Finally, make it easy to buy.  There are plenty of easy-to-implement self-service payment solutions, starting with PayPal. If your product is expensive, and you want to build a broader audience, you may want to try a “freemium” pricing model, where some data is available for free after registration, but the really juicy stuff is only available in the paid version.

Price your products in line with competitive offerings.  If there aren’t comparable products, model the likely revenue (or savings) a customer will generate by using your data product.  If the data is difficult to find from anyone else, you can add a premium.  If the data product can’t be easily mapped to revenue or cost savings, figure out your development costs and a target margin and price to recover initial costs and reach your target margin within 6-18 months.

Building a successful data product isn’t generally an easy or fast process, but a good data offering is an excellent hedge against the ups and downs of advertising.