At many businesses, the honest answer is the second one. Contact happens when there is time to spare, when the month’s sales are weak or when someone remembers an old customer. The team polishes the wording of the message and almost never thinks about the day it will arrive.
- The timing of the contact changes how many customers buy again.
- The repeat purchase interval is how often your customers usually come back, and it comes from the sales history you already have.
- Reach out close to that interval, start with those who are already past it, and measure the result over a 15-day window after the contact.
What timing does to repeat purchases
In a real case at a retail chain, MeContaí compared customers contacted at different points after their last purchase.
- Contacted within 30 days of the purchase
- 36 of 100 bought again
- Contacted only after 120 days
- 6 of 100 bought again
A real case, not a promise of results. The numbers vary from business to business.
Both groups were customers of the same chain, and the difference between them was when the message arrived. The analysis also tracked a control group that received no message at all, to separate those who came back because of the contact from those who would have come back anyway.
Right after a purchase, customers still remember the store and how they were treated. As the months go by, they start meeting the same need somewhere else.
Every business has its own interval
That doesn’t mean 30 days is the right timing for everyone. A business that sells pet food, cooking gas or party snacks has one cycle. One that sells clothing, paint or beauty services has another. Even within the same business, different customers come back at different paces.
So the useful question is how often your customers usually come back. That interval is the repeat purchase interval, and you can find it by looking at the sales history you already have.
How to calculate the repeat purchase interval
You need a list of sales with the customer and the date of each purchase. Most systems can export this to a spreadsheet.
- Export your sales from the last 12 monthsWith the customer and the date of each purchase.
- Keep only those who bought two or more timesCustomers who bought only once don’t have an interval to measure yet.
- Count the days between one purchase and the nextFor each customer. If someone bought on Mar 5, Apr 4 and May 10, the intervals are 30 and 36 days.
- Average the intervalsIf your customers are very different from one another, use the median (the middle value), which isn’t skewed by a few extreme cases.
- Write that number down: it’s your repeat purchase intervalIf it comes to 35 days, that is how often the typical customer of your business comes back.
To build the list, look at each customer. Their next expected purchase is the date of their last purchase plus their own interval. Customers with only one purchase use the business’s repeat purchase interval.
- Purchases on Mar 5, Apr 4 and May 10: intervals of 30 and 36 days, an average of 33.
- Adding 33 days to the last purchase, the next one is expected on Jun 12.
- If they haven’t bought by then, they’re past their interval. They’re the ones you contact first.
If your business sells products with very different uses, it’s worth calculating by category. The interval for someone buying a replenishment item is not the same as for someone buying a durable one.
What to do with this number
Reach out close to the repeat purchase interval
That’s when customers usually need you again. In the timeline example, the message goes out the week of Jun 12.
Start with those who are already past it
List the customers whose last purchase was longer ago than the repeat purchase interval and who haven’t come back. Contact those who passed it only a few days ago first, before they turn into a 120-day case. The further they get from the interval, the lower the chance they come back, so don’t let this list grow.
Give the response time to turn into a purchase
Not every customer buys on the same day they get the message. In a real delivery case, only 6.6% of the purchases generated happened on the day of the contact, and 47% took more than five days. That’s why measuring only the day of the send leaves out most of the result. Set a window (for example, 15 days after the contact) and count everything that comes in during it.
Recalculate from time to time
Customers’ pace changes with the seasons. Redoing the math every quarter keeps the number useful.
To find out how many customers haven’t even made a second purchase yet, see how to count how many customers bought only once.
Be careful with contact
- Only message people who have bought from your business and agree to receive messages.
- Use their name and mention what they bought last time.
- Don’t contact the same person several times in a row. If they buy, stop the sequence.
- Anyone who asks to stop receiving messages should be taken off the list.
Frequently asked questions
What is the repeat purchase interval?
It is the average time, in days, between one purchase and the next among customers who buy again. It shows how often your customers usually need you.
How do you calculate the repeat purchase interval?
Take your sales from the last 12 months, keep only the customers who bought two or more times, count the days between each customer's purchases and take the average or the median.
When is the best time to contact customers again?
Close to your business's repeat purchase interval. In a real case, 36 out of every 100 customers contacted within 30 days of their purchase bought again, compared with 6 out of every 100 contacted only after 120 days.
What about customers who bought only once?
They don't have an interval of their own yet. Use the average repeat purchase interval of your other customers as a reference for the first invitation to a second purchase.
Do I need software for this?
To do the math once, a spreadsheet is enough.
In a base of a few thousand customers, someone passes their interval every day, each on a different date.