Acquisition or Retention? Where Growth Research Says Budgets Should Go

Ask a room of business owners where growth comes from and someone will say it costs five times more to win a customer than to keep one. Someone else will add that a 5% lift in retention adds 25% to 95% to profit. Both lines sound like research. Followed back to their sources, one has no traceable study behind it and the other comes from a much narrower claim. The research on how brands actually grow points the other way: most growth comes from winning more buyers, not from squeezing more out of the ones you have.

That does not make retention worthless. Retention and acquisition do different jobs, and confusing them leads companies to underfund the one that grows the business. This report traces each claim, sets out what the growth evidence found and its limits, and shows how to work out which problem your company has.

The two numbers everyone quotes, traced to their source

“Five times cheaper to keep a customer”

No one has found an original study for this number. The authors of Loyalty Myths (Keiningham, Vavra, Aksoy and Wallard), who tried to trace the wider claim that new customers cost far more than existing ones, report that the earliest references attribute it to work by the Technical Assistance Research Project (TARP) in Washington in the late 1980s. They say it is “difficult to determine the exact origins” of the claim, and that it gained credibility through Hart, Heskett and Sasser’s 1990 Harvard Business Review article “The Profitable Art of Service Recovery” and through Tom Peters’s Thriving on Chaos. The excerpt itself gives no ratio from the original research. Even the size of the gap is not fixed: Amy Gallo’s 2014 HBR piece says acquiring a customer is “anywhere from five to 25 times more expensive” and notes that the ratio depends on which study you believe and on the industry.

The Loyalty Myths authors also question the reasoning behind the claim. It assumes existing customers spend more over time, buy at full margin rather than at a discount and are cheaper to serve; the authors say research has not shown these things to be true. They also argue the ratio changes over a product’s life. Acquisition costs dominate early, retention costs dominate in decline, and in mature markets it can go either way.

“5% more retention, 25% to 95% more profit”

This one has a named source: Frederick Reichheld of Bain & Company. It goes back to his 1990 HBR article with W. Earl Sasser Jr., “Zero Defections: Quality Comes to Services”. Reichheld’s own Bain write-up is more limited than the version people quote: “In financial services, for example, a 5% increase in customer retention produces more than a 25% increase in profit.” The “25% to 95%” range is how Gallo summarised Reichheld’s research in HBR in 2014. An unsigned 2006 article on Bain’s site repeats only the top of the range, saying profits can be boosted by “as much as 95 per cent”.

Three points get lost when the figure is repeated:

• It is about profit, not growth. It models how much more a company earns from customers who stay longer. It does not show that retention programmes grow sales or market share.
• It comes from specific service businesses. Reichheld’s own example is financial services, where customers hold accounts for years. It is not a cross-industry average.
• “5%” is usually a bigger change than it sounds. If retention goes from 90% to 95%, defection falls from 10% to 5%, which halves it. Critics of loyalty marketing make this point, including the B2B Institute’s Jon Lombardo, who draws on Byron Sharp’s critique of Reichheld. Halving churn is rarely cheap or fully in a company’s control.

What the growth research actually measured

Two bodies of evidence are often confused. Here is what each studied.

Ehrenberg-Bass: penetration drives share

Byron Sharp’s How Brands Grow (Oxford University Press, 2010) collected decades of Ehrenberg-Bass Institute research on buying behaviour. Its central finding is that brands with larger market share have many more buyers, and those buyers are only slightly more loyal. Sharp’s summary in his reply to critics is: “Larger brands have higher penetration, and all their loyalty metrics are a bit higher too.”

The most direct test is a 2014 study by Riebe, Wright, Stern and Sharp, “How to grow a brand: retain or acquire customers?” in the Journal of Business Research (vol. 67, pp. 990–997). It used panel data on doctors’ prescribing and a cross-sectional tracking survey for residential finance, and compared gains and losses with what normal churn would predict. In the authors’ words, “for both growth and decline, unusual acquisition plays a stronger role than unusual defection.” A summary published on the Ehrenberg-Bass site says acquisition is “roughly twice as important” as reduced defection. That ratio comes from those categories and should not be read as a universal constant.

Binet and Field: B2B campaign outcomes

The B2B Institute (funded by LinkedIn) commissioned Les Binet and Peter Field to analyse B2B cases in the IPA Databank from 1998 to 2018. Their report, The 5 Principles of Growth in B2B Marketing, found that “customer acquisition strategies tend to be much more effective than loyalty strategies”. It also found that reach strategies, which talk to customers and non-customers together, “tend to be the most effective of all”. On their measure of “very large business effects” (a combined measure of six business indicators) for 2008–18 B2B cases, loyalty campaigns averaged 0.0, acquisition campaigns 1.0 and reach campaigns 1.6.

The caveats come from the authors themselves:

• “Our findings are tentative.” The B2B sample was fewer than 50 cases.
• The Databank is biased towards effective campaigns, because it is built from effectiveness award entries, although the authors describe the bias as modest.
• The cases are skewed towards the UK and tend to have relatively big budgets, as the report notes and independent summaries such as Alex Murrell’s repeat.
• The funder sells advertising reach, so its interest in this conclusion is worth keeping in mind. The finding matches independent Ehrenberg-Bass work, but it is still directional evidence, not a rule.

Why loyalty follows size in B2B

The pattern behind this research is called Double Jeopardy. Smaller brands lose out twice: they have fewer buyers, and those buyers are slightly less loyal. In a May 2021 Ehrenberg-Bass report, Jenni Romaniuk, John Dawes and Sahar Faghidno showed the same pattern in UK business banking and US business insurance, and cited earlier published studies in categories such as concrete supply and aircraft buying. Their conclusion: “You have more loyalty because you are big, rather than you are big because you have more loyalty”, and “it is not possible to sustainably grow by focusing on loyalty alone.”

This matters for planning because loyalty has a natural ceiling set by your size. A mid-sized supplier that sets a target to match the market leader’s repeat rate without first winning more buyers has picked a target the evidence says it will not reach. The same report makes a useful diagnostic point. Loyalty that is unusually high for your size can signal a constraint rather than a strength: the authors link it to the firm’s distribution or its range of offerings, with restricted distribution the most likely cause.

The timing of B2B buying adds to this. John Dawes’s 95:5 heuristic holds that only a small share of buyers is in the market in any quarter, so most of your future customers are not buying today and reaching them now is how you get considered when they do. Our brand and performance budget guide explains the heuristic and its limits.

Where retention work does pay

The evidence argues against relying on loyalty for growth. It does not argue for neglecting customers. Even the Ehrenberg-Bass summary that favours acquisition says keeping service and product quality at a reasonable level still matters. Retention effort earns its cost in these situations:

• Churn is caused by things you control. Late deliveries, slow quotations, warranty disputes and unanswered WhatsApp messages are operational failures, and fixing them stops avoidable losses. Lombardo’s B2B Institute piece makes the same split: advertising for acquisition, customer experience for retention.
• Subscription and contract markets. Where customers hold one provider at a time, such as maintenance contracts, managed services or annual supply agreements, every defection loses the whole account. Renewal management has a clear, measurable payoff; our guide to selling AC maintenance contracts shows one renewal system.
• A few accounts carry the business. The Loyalty Myths authors note that the customers who cost most to retain are often the ones who generate the most profit. If three clients make up half your revenue, protecting them is risk management, not a growth strategy, and it is worth doing.
• Existing customers have real unmet need. Cross-selling works when the customer actually buys the other products elsewhere, such as a contractor who buys your pumps from you and your valves from someone else. It does not work when you are pushing products they do not need.
• Referrals in a small market. In closely connected Gulf sectors, a satisfied client’s word reaches consultants and procurement teams you have never met. That is retention work that helps acquisition.

Retention starves growth when it takes the budget and attention that should go to reaching new buyers. Common signs are loyalty discounts for customers who would have stayed anyway, newsletters only existing clients read, and sales teams that only farm accounts while the list of prospects shrinks.

How to tell which problem your company has

Before you change the plan, find where revenue is actually being lost or not won. Take the last two or three years of sales records and answer these questions:

1. Split revenue change into buyers and spend per buyer. Did revenue move because you had more or fewer active customers, or because each customer bought more or less? If the number of active customers is flat, growth has stalled at acquisition, whatever the repeat rate.
2. Find out why customers left. Call ten lost accounts. Separate reasons you control (service, price disputes, response time) from reasons you do not (project ended, company closed, budget moved, a new procurement rule). Only the first group responds to retention spend.
3. Compare your loyalty with your size. If your repeat rate is roughly what you would expect for a company of your market share, more loyalty spend will mostly reward customers who would have stayed anyway. If it is clearly below that, look for a service problem.
4. Check how many category buyers know you. List the companies in your target sectors that buy what you sell. What share have heard of you, met you or received a quotation from you? That share is your realistic limit on future deals.
5. Look at what the plan actually funds. Count how much of this year’s activity reaches people who are not yet customers. Many B2B plans spend most of their effort on existing accounts without anyone deciding to. An annual marketing plan should make that split a deliberate choice.

A hypothetical example (not a real client): a freight forwarder finds revenue up 8% while its active customer count fell. The growth came from a handful of large project shipments, the kind of income Double Jeopardy suggests will not last. The fix is not a loyalty scheme. It is getting known by the importers and procurement teams who have never asked it for a quote. Working through this kind of diagnosis before the plan is written is the core of a marketing strategy and research engagement.

Give acquisition and retention different owners

The practical answer is not “acquisition instead of retention”. Give each its own owner, tools and measures, so one does not quietly absorb the other.

The split between brand reach and short-term activation is a separate decision, covered in setting your own brand and performance split. The question here comes first: is the money reaching people who are not yet buying from you?

Start with the first check above before next year’s plan is written: from last year’s invoices, count active customers and revenue per customer, and put the two numbers next to the year before. That one split usually shows whether the plan should fund reach or service, before anyone argues over the budget.