How Good Is Your CRM? Your Customer Retention Rate Can Tell You

Ask a dealer principal how many customers came back last year and you usually get a feeling, not a number. That gap is the best test of a CRM there is. If your system cannot tell you who you kept, who you lost and who you never heard from again, it is a contact list, not a customer relationship tool.

When this post first ran in 2020, it said your CRM reports would show how many customers return. This version, for dealer principals, managers and BDC leads, shows how: the formula, three numbers to keep apart, how to pull them honestly, what moves them, and the data your CRM must hold first.

The customer retention rate formula

The standard formula is:

Customer retention rate = ((E − N) ÷ S) × 100

• S is the number of customers you had at the start of the period.
• E is the number of customers you had at the end of the period.
• N is the number of new customers you gained during the period.

Subtracting new customers is the whole point. Without it, a store that lost half its owners and replaced them with first-time buyers would look as if nothing had changed.

A worked example

The numbers below are hypothetical, for a single-rooftop store. "Active" means the customer had at least one unit sale, service repair order or parts and accessories sale tied to their named record during the year.

Dividing end by start says 95% and hides that 900 of last year's customers did not come back. The textbook formula counts reactivated customers as "retained" even though they were not in last year's group. The cleanest method is to match customer IDs between the two years and count how many of the 2,000 appear again: 1,100, or 55%. That is the number to report.

Three retention numbers a powersports store should track

One blended rate mixes customers who buy a unit every few years with customers who buy oil every month, so it moves for reasons you cannot see. Split it into three numbers, each with its own question and time window.

Sales repurchase rate

Take a cohort old enough to have had time to buy again. In our hypothetical store, 400 named customers bought a retail unit in 2020 and 2021, after merging duplicate records and removing wholesale, employee and internal deals. Seventy-two of them bought another unit from the store within 48 months of their purchase. Sales repurchase rate: 72 ÷ 400 = 18%.

The 48-month window is an assumption, not a rule. Check it against your own data: for customers who have bought twice, how many months passed between the first and second unit? Use that median to set your window, then keep it fixed so this year's number compares with last year's.

Service retention rate

In the same hypothetical store, 280 customers bought retail units between October 2023 and September 2024. Five have since been flagged as moved out of the area, leaving 275. Of those, 154 had at least one repair order at the store within 12 months of their purchase. Service retention: 154 ÷ 275 = 56%.

Decide up front whether warranty, customer-pay and pre-delivery work count. Pre-delivery inspection should not; it happens on every unit and says nothing about whether the owner chose to come back.

Owner contact rate

This is the early-warning number. Of 1,500 owners who bought or serviced a unit in the last 24 months, 1,140 have a working phone number or email and a valid permission for at least one channel. That 76% is your reachable rate. Only 690 had a logged two-way conversation with the store in the last 12 months: a reply to a text or email, a connected call, or a visit. Owner contact rate: 690 ÷ 1,500 = 46%.

Sales and service retention tell you what already happened. Contact rate warns you about what comes next.

How to pull the numbers honestly

Most bad retention numbers come from loose definitions, not bad arithmetic. Settle these once and run the report the same way every quarter.

1. Define "customer." A named person or business with a transaction tied to them. Unnamed cash counter sales cannot be counted; say so.
2. Merge duplicates first. One rider entered by sales, service and parts looks like two lost customers and one retained one. Match on phone, email and address before counting.
3. Match on customer ID, not name. The CRM and your dealer management system (DMS) should share an ID, or you need a reliable match key.
4. Fix the windows. Calendar year, trailing 12 months or a cohort window: pick one per number and do not change it to flatter a quarter.
5. List the exclusions. Wholesale deals, employee purchases, internal accounts, and customers flagged as deceased or moved. Exclude only what is flagged in the data, not what someone remembers.
6. Separate new, retained and reactivated. Reactivated customers are good news, but they belong in their own line.
7. Report the counts with the rate. "56% (154 of 275)" lets anyone check the math and see when a small cohort makes the rate jumpy.
8. Keep a definitions sheet. One page that states every rule above, so the number means the same thing when someone else runs it.

What makes retention numbers move

Retention is the result of many small handoffs. The table links each lever to the number it moves first.

The visits right after the first purchase matter most. An analysis published in Powersports Business in 2008 followed parts-counter customers at one store from July 2006 to June 2008. Only 10 to 13 of every 100 new customers came back in their second month, and about 4 of every 100 were still returning a year later, while 67% of established customers were still buying after 12 months. That is one store, one department, nearly two decades ago: a pattern to check in your own data, not a benchmark.

For the reminder and trigger side, see our guide to personalized marketing for powersports dealers. For a structured owner program, see our article on building a powersports owner retention program, and for measuring how customers feel about each visit, our piece on measuring the impact of customer experience.

Why published retention benchmarks rarely fit your store

You will find retention benchmarks online. Before you compare yourself to one, check what it measures.

• Car dealerships, not powersports. An Auto Remarketing report on VehicleLyfe's analysis of more than 2.7 million dealership records from DMS integrations defined loyalty as customers returning "for repeat purchases or service". It reported that about 20% of dealership sales came from repeat customers and that VehicleLyfe considers one-third healthy. The report does not publish the exact formula or how repeat customers were identified, and the data covers franchised car dealers.
• One store, parts only. The 2008 Powersports Business figure of 67% covers parts-counter customers at a single store.

Those two numbers measure different things in different businesses, so neither is a target for a powersports store. We know of no current, openly published powersports retention benchmark with a stated method. Your best benchmark is your own number last year, measured the same way.

What your CRM must record to measure retention

This is the real answer to "how good is your CRM?" Without these fields, no dashboard can produce honest retention numbers. Use this as a checklist.

• One record per customer, with duplicates merged and a stable customer ID shared with the DMS
• Customer-since date, the date of the first transaction with your store
• Units owned: VIN, year, make, model and sale date, with a flag for units bought elsewhere
• Every transaction linked to the record: unit deals, service repair orders with pay type, and named parts and accessories sales
• Status flags with dates: moved, deceased, sold the unit, do not contact
• Contact details with a last-verified date: phone, email and mailing address
• Consent by channel, with the date and how it was given, for calls, texts and email
• A communication log that records calls, texts and emails, inbound and outbound, and whether the customer replied or the call connected
• Notes and dispositions on every lead and owner conversation, including why a deal did not happen
• Assigned salesperson, service advisor and rooftop, so each number can be split by person and store

Consent matters for the reachable rate. Under the FCC's rules, advertising texts sent with an autodialer need the customer's prior express written consent, and customers can revoke it in any reasonable way. This is general information, not legal advice.

If you can tick every box, run the three numbers this week. If not, the missing fields are your first project. For choosing a system with retention in mind, see our guide on how to choose a powersports CRM.

How PSX Ascend fits

The AI-powered powersports CRM from PSX Ascend brings customer, market, communication and historical data into one connected view, with less manual entry, and keeps that view across every rooftop in a dealership group. PSX's process requires sales associates to leave quality notes and accurate dispositions when a deal does not happen, which is the data that explains a lost customer.

ALMA, the Automated Lead Management Assistant, does the repetitive work of re-engaging stalled prospects and customers who chose not to buy, and brings them back to a live conversation with a person. It makes no AI voice calls. PSX's AMP marketing automation platform matches content to each customer's lifecycle and interest and coordinates email, text, phone and sales, so a reply from an owner reaches the right person with the history attached.

Whichever system you use, the order is the same: clean data, written definitions, then a retention number you can trust.

You can also see the same AI-powered powersports CRM on our sister site, PSXDigital's AI-powered powersports CRM page.

Sources

• Auto Remarketing, "VehicleLyfe data shows 'loyalty crisis' as dealership customer retention drops"
• Powersports Business, "Sept. 1, 2008: The 'deep pool' of returning powersports customers" (Hal Ethington)
• 47 CFR 64.1200, TCPA rules on prior express written consent and revocation (Cornell LII)