Logo retention is 70%, net dollar retention is 115%. Which one matters for your stage.

A founder shows two retention numbers in the same deck. Logo retention is 70 percent. Net dollar retention is 115 percent. Neither is a typo. One says the company lost nearly a third of its customers over the year. The other says the dollars from existing customers grew. Both are measured off the same book of business, and an investor will read them together, not separately.

What the two numbers actually count

Logo retention counts customers. Take a cohort that started twelve months ago and ask how many are still transacting today. If 100 signed up and 70 are still active, logo retention is 70 percent. It does not care whether the survivors spend more or less than they used to. A customer is either present or absent.

Net dollar retention counts dollars from the customers you already had. Start with last period's revenue from a set of accounts. Add expansion, subtract contraction and churn, and compare to where you started. At 115 percent, the accounts you kept, plus the ones that grew, more than replaced everything that shrank or left. New logos are excluded on purpose. This is the base you already owned, measured again.

The two can diverge hard, and the gap is where the story is.

How you lose 30 percent of logos and still grow the base

The arithmetic is not a paradox once you see it. Say the 30 customers who left were small, each contributing a little. The 70 who stayed were larger, and a handful of them doubled their spend. Expansion inside the survivors outran the revenue that walked out the door. Dollars up, logos down.

This is the signature of a land-and-expand model working as designed. It is also the signature of dangerous concentration. A 115 percent NDR carried by three accounts that tripled is a very different business from a 115 percent carried by broad, even expansion across the base. The headline number is identical. The risk underneath it is not.

So the second question an investor asks after "what is your NDR" is "who is expanding." One answer is a flywheel. The other is a countdown.

Which one your stage lives on

For an early company selling into a large market, logo retention is the leading indicator. It tells you whether the product holds people once they arrive, before you have enough expansion history for NDR to mean much. A 70 percent logo retention at seed is the market telling you something about fit, and no amount of expansion from the survivors fully offsets that.

For a later company with a proven expansion motion, NDR is the engine of the model. Efficient growth stops leaning on new-logo acquisition and starts compounding off the installed base. Here a high NDR is the number that justifies the multiple, and churn among small accounts can be tolerated if the accounts that matter grow.

The mistake is leading with the number that flatters you and hoping the other one does not come up. It always comes up.

What the pair tells an underwriter

Read together, 70 and 115 sketch a shape. Small customers leak, large customers expand, and the business is quietly sorting itself into a smaller number of bigger relationships. That may be exactly right for an enterprise trajectory. It may also mean the top of the funnel is filling with customers who were never going to stay, and the sales and marketing spent to acquire them is a recurring tax on a leaky bucket.

An underwriter will not take either number at face value. They will look at logo retention cohort by cohort to see whether the 70 is stable or sliding, and they will look at how revenue splits into growth and loss to see whether the 115 is broad or balanced on a few accounts. The single figures are the headline. The cohorts and the composition are the diligence.

Two numbers, one honest picture

Levian computes both.