RFM analysis: how to spot your best Whatnot buyers slipping away

Your best buyers rarely leave all at once. They show up a bit less, spend a bit less, and one day they are buying from someone else. Online stores have tracked this for years with a simple method called RFM analysis, and it works just as well for a Whatnot shop.

What RFM means

RFM looks at every buyer through three numbers.

#	Letter	What it measures	What it tells you when it gets worse
1	Recency	How long since their last purchase	They may be drifting to another shop
2	Frequency	How often they buy	You may not be stocking what they want
3	Monetary	How much they spend	The most serious warning sign

Each one points to a different problem, and each one has a different fix.

Recency: it has been too long

When a regular hasn't bought in a while, the clock is running. A coupon or voucher can be the right tool to win them back before a competitor does. A comeback offer that only works in one show gives them a reason to turn up to that show.

Frequency: they are buying less often

If someone who used to buy at most of your shows now only buys once in a while, the usual reason is that you are not selling what they want anymore. Common causes:

• You stopped restocking the thing they usually bought.
• They are buying it from someone else.
• They already collected the items you keep showing.
• A customer support or shipping problem left them unhappy.

Figure out which one it is, and if it is stock, get it back in as soon as you can.

Monetary: they are spending less

This is the one that hurts most. When a top buyer's spend drops, message them and ask what changed. Often it is a bad order, a slow shipment or a problem that never got fixed.

For your top buyers, it can be worth making exceptions on refunds or shipping to put things right. Keeping one big buyer happy is worth far more than the cost of a partial refund.

How to work it out yourself

You can do a basic RFM analysis with your Whatnot exports and ChatGPT:

1. In Whatnot, go to Shipments and download the shows you want to look at. The more shows, the better.
2. Upload the files to ChatGPT.
3. Ask it to combine the files, find the buyer, date and amount columns, and calculate recency, frequency and monetary value for every buyer.
4. Ask it to score each buyer from 1 to 5 on each letter, and to flag your top buyers whose orders or spend dropped in the recent half of the data compared with the earlier half.

Check which columns it used before trusting the results. Export formats change, so let it find the columns rather than assuming their names.

If you would rather not do this by hand, HypeStream's customer breakdown tracks recency, frequency and spend for every buyer automatically.

What to do in your own shop

1. List your top buyers by total spend. Start with the top 20%.
2. Check when each one last bought. Anyone who has gone quiet gets a personal message or a comeback coupon.
3. Look for buyers who now buy less often. Ask yourself what you stopped stocking, and bring it back if you can.
4. Message top buyers whose spend dropped. Ask what changed, and fix any order or shipping problem first.
5. Repeat every few weeks. RFM only helps if you catch the drop early.