Saturday, February 25, 2012

The NBD model – ways to grow your business

Let’s get mathematical for a change and talk about negative binomial distributions (NBD) – more specifically the NBD-Dirichlet Model as developed by Andrew Ehrenberg (Wikipedia Link Here).

The maths of this need not put you off, however, if you understand the principles that Ehrenberg uncovered you will improve your business decision making rapidly.

For a great introduction to this model and also to read one of the best books on business I have ever read please track down Byron Sharp’s ‘How brands grow, what marketers don’t know.’  The following is a precis of some of the points from Byron’s book.

The above is an idealised version of a negative binomial distribution – I say this to mean that the percentages and frequencies are guaranteed not to match your industry, but the principles of this illustration are what counts. And apologies to Byron Sharp – the following is my poor attempt to paraphrase the key principles.

First off, you can see that light buyers represent the largest part of your buyers. In other words it is infrequent buyers who make or break your business.

I personally think you can also apply this to a service business too if you think of the number of purchases as blocks of billable hours.

Secondly, you can see that improving the loyalty of your frequent buyers won’t make that much of a difference to your revenues.

Thirdly, targeting a particular segment of your buyers is unlikely to make a material difference to your business. You are better off targeting the light, medium and heavy buyers through mass marketing.

Finally, as your market penetration grows the binomial distribution will stay roughly the same shape, it’s just that you earn more sales revenues. Again, this means you are better off targeting all buyers rather than a select few.

Ever since reading this I have been applying (and likely misapplying) this principle all over the place.

Even this blog and the user statistics bear out the NBD-model in a crude way.

My main marketing so to speak is LinkedIn. I post the blog on a Monday and my regular readers are on to it the same day. However, after reading about the NBD model I realised many people may log on to their LinkedIn account weekly if not monthly.

So given that most of us don’t go past the first page of news on LinkedIn by Friday my blog post advertisement would be buried by new posts somewhere on page 2 or 3 of your LinkedIn home page.

As an experiment I started reposting (sometimes) on a Friday, and lo and behold, readership went up as I captured those who log on on Friday and Saturday too.

To take it to the next stage, I realised that many LinkedIn users only log on every month or so – which I see when someone logs on and catches up on 10 or 20 postings – which happens once every week or so. So, the next step would be to repost old blog postings regularly (daily) as the light readers would be unlikely to have read that headline yet. On the other hand, regular readers will become jaded, and many people would start blocking my LinkedIn posts. However, targeting lighter readers appears effective.

One of the key challenges I would like you to take away from this is how you can best service light buyers.  

For example, the following are a challenge to those in the world I live in, service firms:
  • Calling all law firms – while it makes sense to chase large clients which undertake a lot of complicated transactions, don’t neglect the many thousands of small and medium size businesses that only undertake a transaction every 5 to 10 years.
  • Calling engineering consultants – winning major deals that employ hundreds of people for a couple of years does make sense. However, have you thought about how to win and deliver the smaller jobs for smaller clients.
  • Calling bankers – winning the big deal is nice, but I bet if you added up a lot of smaller deals it would be better for your bottom line.

I have been talking to some of you out there about this, and the most common line I am hearing against targeting lighter buyers in the service industry is that there are too many overheads and that it is hard to be efficient.

My counter argument to that would be to say that if you have a large volume of customers in the light buyer category then you are likely to make enough margin to cover the overheads.

As to the efficiency argument, my response would be to commodify those transactions as much as possible. A lot of service firms build themselves up around providing specific tailored solutions to exactly match each client’s exact problems. I’ll bet a lot of customers will be happy with ‘a solution’ as opposed to ‘the solution.’ I am sure you will find that much of what you are doing is the same, except for the names of the companies involved.

So, next time you are talking to someone about their business, feel free to ask how they are affected by the NBD-model. It’s a great party conversation.

P.S. After posting the above I was watching the cable TV show Storage Wars in which the most successful businessman said, "It was when I started selling low end items instead of just high end furniture that I turned my business around and started making a profit." It took this man 20 years to figure that out - don't wait that long.


  1. Efficiently written information. It will be profitable to anybody who utilizes it, counting me. Keep up the good work. For certain I will review out more posts day in and day out. New York SEO

  2. Thank you so much for this nice information. Hope so many people will get aware of this and useful as well. And please keep update like this.

    Big Data Services

    Data Lake Services

    Advanced Analytics Solutions

    Full Stack Development Services