Pondering this question, the owner went about the usual business tasks for the day. Scanning the new newsletter subscriptions, the owner notes the different sources producing the majority of new subscribers, then moves on to process orders for the day. Folks, recall that response to the newsletter has been falling, and the owner was pleased to see the latest newsletter generated decent order activity.
As usual, some orders stood out from the rest; the owner recognized repeat buyers and people who had just placed an order Recently. "What makes them do that, I wonder?" thinks the owner. They buy something then they buy something else only a week later. Why don't they buy both at the same time? They could save money on shipping, the owner thinks...
As the owner processed orders, thoughts returned to the 30-60-90 bucket analysis. I have all these orders, day after day, the owner thinks, yet most customers have not bought from me in quite some time. How is this possible? It doesn't make any sense.
Then the owner has a brainstorm. What would the 30-60-90 Last Purchase Date information look like just on people who responded and purchased from the recent newsletter? I could match people who bought from the newsletter with their Last Purchase Date before I sent the newsletter, and then could find out how effective the newsletter is at getting my "lost customers" - those who have not purchased in months - to buy again.
In other words, what percentage of people in each 30 day bucket purchased through the last newsletter I sent out? Perhaps this would provide the insight needed to demonstrate what this Recency data means, and provide some insight into the kind of action that needs to be taken to keep people buying for a longer time. The owner sorted responders to the newsletter according to the Last Purchase Date before the newsletter was mailed out, with the following results:
Last Purchase Date Before Newsletter Drop
The owner was slack-jawed. How could this be? Is it possible that (top row) almost 1/3 of the responses came from 3% of customers? That (top two rows together) nearly 50% of the responses came from 9% of customers? The owner's head was swimming! What was the implication here? Is it possible - and just this simple - that the response rate of a customer to the newsletter could be predicted based on how many days ago they last made a purchase? The implications were stunning. One simple calculation. Incredible ability to predict purchase behavior.
Of course, my fellow Drillers, the question really is what can the owner of IMIssAsia do with this new information to make the business more profitable? We'll get to that issue next month, when the owner discovers an even more stunning connection between the newsletter and the purchase behavior of subscribers - and figures out how to increase profits by taking advantage of it.
If you would like to read the next installment of
Q: Hi Jim,
Our industry is facility management services where a headquarters with chain locations contracts with us to manage their facilities in all their markets. The President is interested in a "CRM Solution" but is concerned about the ROI he might expect from implementation. Do you know of any number that I can pass along to him that would placate his insistence on knowing in advance what the ROI will be?
A: Bad news: No, not really.
Good news: You can figure it out, which is something nobody did in the past and is why so many "failed" at CRM. You might not even need any new software to "do CRM", though it depends on what you have now and what the objective of the CRM program is (you do have one, correct?). But the software required is certainly not millions of dollars and if you only have 1000's of clients you can probably do it with some combination of ACT! or GoldMine, MS Access, and MS Excel.
The key question to ask: do you really know how your customers behave? In this kind of contract business, I imagine the central issue is this: Can you predict which customers are likely to re-up a contract, and which ones are not? And then can you use this information to focus on the ones less likely to re-up, and take steps to make them more likely to re-up?
Sometimes it is just a matter of better customer service. In this case, what you need is better service practices, not "CRM". From a distance, it is very difficult to know what the issues might be in your company.
Here's a test you can do to find out where you might be on the road to answering the CRM question. If you cannot accomplish one or more parts of the following, you are not ready to even talk about "CRM", and need to do some more internal research. These steps, by the way, are the ones everybody skipped on the first round of CRM and will pave the way for a successful implementation if you decide to go with a CRM approach:
1. Define a "best customer". It's not just sales, you have to take into account margins, service costs, etc. Don't worry about finding exact financial numbers. Think about best in relative terms - these customers are better than those customers, and you are pretty sure it is true. If you can't get to this point, you probably need better data collection before you think about a CRM project.
2. Once defined, how many of these customers left you in the past year or 2 years or whatever the right time frame is for your business? If you typically sign 3 year contracts, then it might be "in the past 4 years". Also identify best customers who stayed with you and renewed. If you can't get to this point, you probably need better data organization before you think about a CRM project.
3. Group best customers who left and best customers who stayed and compare the two groups. Look for similarities and differences. Is it the kind of business they are in, geography, number of "trouble calls", billing disputes? You will almost always see patterns that will lead you to conclusions on what circumstances create a best customer who stays and one who leaves. If you can't get to this point, you probably need better data analysis before you think about CRM.
4. Now that you know what causes customer defection and retention, figure how much more money you could make if you could keep a certain percentage of these best customers that would otherwise leave. If you can't do this, you probably need better customer reporting before you think about a CRM project.
5. Figure out what it would cost to keep this certain percentage of best customers that would otherwise leave. Is it better targeting in sales upfront, better customer service, better billing practices? If you can't do this, you probably need better financial reporting before you think about CRM.
6. Calculate your ROI and either decide to do it or not. If you can't do this, don't invest in CRM, because it is going to cost you more than you will make by implementing it!
Sorry I don't have "a number" for you, it simply does not work like that. But you can find that number, a number that is right for your business, with a little detective work. If you get hung up with any of the above steps, perhaps I can help. Many businesses can get great results following the above plan. If you are a larger organization with a lot of complex issues, you might need the industrial strength version of the above, described here.
What would you like to
the book with Free customer scoring software at: Learn Customer
Marketing Models and Metrics (site article
Get the book with Free customer scoring software at:
Learn Customer Marketing Models and Metrics (site article list)