Declining Sales: How to Find the Cause and Fix It

Almost every sales decline traces back to one of five things: fewer people finding you, fewer of them buying, the ones who buy coming back less often, each order getting smaller, or the market moving underneath you.

They all produce the same downward line on the same chart. They need completely different responses. Running a promotion to solve a retention problem, or spending on ads to solve a conversion problem, is how owners burn a month and a budget and end up exactly where they started.

This guide is the diagnostic. Work through it and you’ll know which of the five you’re dealing with, usually in an afternoon with data you already have. Once you know the cause, the broader recovery sequence of cash, focus and offer is covered in our guide to getting out of a business slump.

The Bottom Line

• Don’t act on the revenue number. Split it into customer count, purchase frequency and average order value first, because each one has a different fix.
• Check whether it’s you or the category before you conclude anything. Aggregate small business sales rose 0.8% month over month in June 2026 and 2.4% year over year (Bank of America Institute, 2026).
• Retention problems are the cheapest to fix and the most commonly misdiagnosed as demand problems.
• Compare against the same period last year, not last month. Seasonality accounts for a lot of drops that get treated as emergencies.
• Give any fix a proper test window and measure it against the specific number you’re trying to move, not against total revenue.

The five numbers that explain almost every sales decline

Revenue isn’t one number. It’s the product of several, and only one of them is usually moving.

Written out, revenue equals the number of customers who bought, times how often they bought, times what they spent each time. Layered underneath that: how many people reached you at all, and what share of them converted.

So there are five places a decline can originate:

• Reach. Fewer people are finding you in the first place.
• Conversion. The same number find you, fewer of them buy.
• Frequency. Your existing customers are ordering less often.
• Order value. The same customers order as often but spend less each time.
• The market. Demand in your category has contracted.

Every one of these looks identical on a monthly revenue chart. That’s why the chart is where diagnosis stops for most owners and why so many fixes miss.

How to find out why your sales are down in an afternoon

You need three years’ worth of nothing fancy: last 12 months of orders, and the same 12 months a year earlier.

Step one: compare like for like. Put this year’s month against last year’s same month, not against last month. A florist comparing February to March will always find a decline, and it means nothing. If last year’s equivalent month was also soft, you have seasonality. Plan the cash around it and stop treating it as a crisis.

Step two: split the revenue. For the declining period and the healthy one, count three things: how many distinct customers bought, how many orders they placed between them, and the average order value. Whichever of the three moved most is your cause.

Step three: check whether it’s you. Look at your category. Small business sales at the aggregate level were up 2.4% year over year in June 2026, and NFIB’s optimism index sat at 98.7 in August, above its 52-year average of 98.0 (NFIB, 2026). If the wider picture is stable and yours isn’t, the cause is internal, which is good news: internal causes are the ones on your timetable.

Step four: ask ten customers. Specifically the ones who used to order regularly and have gone quiet. Five conversations will tell you more about the why than the spreadsheet told you about the what. The spreadsheet finds the cause category; the calls find the actual reason.

Cause 1: Fewer people are finding you

The signal is a drop in new customers while your existing ones behave normally. Order counts from repeat accounts hold steady; the new-name column thins out.

Reach problems usually have a specific, findable origin, and it’s rarely mysterious:

• A channel that stopped working without anyone noticing. A referral partner went elsewhere, an ad account got throttled, a search ranking slipped, a directory listing expired.
• Marketing spread too thin. Five channels at half effort produce less than one at full effort, and the decline is gradual enough that nobody notices when it started.
• You stopped doing the thing that worked. Extremely common. The outreach that built the business got squeezed out by the work of running the business.

The fix: find where your last 20 customers actually came from, not where you think they came from. One channel will dominate. Put the whole effort there and pause the rest until volume recovers. If the dominant channel is the one that broke, you have a rebuild rather than a reallocation, and you should expect it to take a quarter.

Cause 2: The same traffic, fewer buyers

Enquiries, footfall or site visits hold up. Orders don’t. That’s a conversion problem, and it’s almost always something specific and fixable rather than something atmospheric.

Where to look first, in order of how often it’s the answer:

• Something broke. A checkout error, a form that stopped sending, a phone line nobody answers at lunchtime, a delivery option that disappeared at postcode entry. Test your own buying process end to end, on a phone, as a stranger would. Owners find something surprisingly often.
• Response time slipped. If quote requests used to get answered in an hour and now take a day, your conversion rate has moved without anything else changing.
• A competitor changed the comparison. Not necessarily on price. On terms, minimums, delivery windows or availability.
• The offer stopped matching the moment. What customers needed 18 months ago isn’t automatically what they need now.

The fix: conversion problems are the highest-return thing on this list, because the demand is already there and already paid for. Fix the mechanical faults first, because they’re free, then work on response time before you touch price.

Cause 3: Customers are ordering less often

Nobody has visibly churned. The wholesale account that ordered fortnightly is now on three weeks. The catering client who booked monthly books quarterly. Revenue slides without a single customer saying goodbye.

This is the most under-diagnosed cause, because there’s no obvious event to point at. It’s also the cheapest to fix. The probability of selling to an existing customer runs around 60% to 70% against 5% to 20% for a cold prospect, and acquiring a new customer costs somewhere between five and 25 times what keeping one does. Bain & Company’s widely cited finding is that a five-point improvement in retention can lift profits by 25% to 95%.

The fix:

• Measure reorder intervals per account, not just total revenue. Build a simple list: last order date, typical gap, days overdue. The accounts at twice their normal gap are the ones to call today.
• Remove the friction that makes reordering an active decision. Standing orders, scheduled recurring deliveries, a one-click reorder. If a customer has to remember you, some percentage of them won’t.
• Fix the operational irritations first. Late deliveries, missing items and unpredictable arrival windows don’t produce complaints in a business-to-business relationship. They produce quiet reduction. Reliable, trackable delivery with proof on arrival is retention infrastructure more than it is logistics, and it’s the sort of consistency a dedicated local delivery platform like Metrobi is built to give food, floral, catering and wholesale operations.
• Call the quiet ones. Not an email blast. Ten phone calls to your best lapsed accounts is the highest-yield hour available to you this week.

Cause 4: Orders are getting smaller

Same customers, same frequency, less money per order. Average order value is falling while everything else holds.

Usually one of four things:

• Customers are trading down within your range, buying the cheaper option they previously skipped.
• Your mix shifted. A high-margin line went out of stock or out of favour, and the volume moved to a low-margin one.
• A bundle or upsell disappeared from your process, often when a staff member left.
• You raised prices and they compensated by cutting volume, which nets out as smaller orders.

The fix: this one responds to structure rather than persuasion. Look at minimum order values, bundles, and what gets suggested at the point of purchase. Then check the arithmetic on your delivery and fulfilment costs. A fixed per-run delivery cost against a shrinking basket turns a profitable order into a break-even one without anyone deciding it should, and that’s a margin problem hiding inside a revenue problem.

Cause 5: The market moved, not you

Sometimes the diagnosis comes back clean. Reach is fine, conversion is fine, retention is fine, and revenue is still down. That means demand in your category contracted.

Signals that this is what you’re looking at: competitors are visibly quiet too, industry publications are reporting the same thing, and your customers are citing their own budgets rather than anything about you. Around 12.1% of small business owners reported declining revenue in a 2026 outlook survey while 87.8% reported stable or growing revenue (Small Business Expo, 2026). That’s a substantial minority, but still a minority, so this diagnosis deserves scepticism before you accept it.

The fix: external causes are the only ones where waiting is a legitimate strategy, and even then it’s waiting plus two things. Extend your cash runway so you can outlast the contraction, and use the quiet period to build the thing you never have time for: the process documentation, the second channel, the adjacent customer segment. Businesses that come out of a category downturn ahead are usually the ones that spent it building rather than cutting.

Matching the fix to the cause

The time column matters as much as the fix column. Conversion problems pay back fastest, which is why they’re worth ruling in or out first even when your instinct says the problem is demand.

A 30-day plan for reversing a sales decline

Days 1-3: diagnose. Split the revenue into customers, frequency and order value. Compare year over year. Name your cause.

Days 4-7: talk to ten customers. Focus on lapsed and reduced accounts. Ask what changed for them, not what they think of you.

Days 8-10: fix everything mechanical. Broken checkout, slow quote response, an unanswered phone, a delivery option that fails at certain postcodes. These cost nothing and they’re the only category of fix that works immediately.

Days 11-20: run one intervention. One, matched to your diagnosis from the table above. Not three. Three at once means you learn nothing about which one worked.

Days 21-30: measure against the right number. If frequency was your cause, watch frequency. Total revenue is too noisy to read in 30 days and will only tell you what you want to hear.

Somewhere in that month, protect the time to actually do it. Owners who are buried in day-to-day operations tend to discover a revenue problem a full quarter after it started, which is why handing off operational work without micromanaging it is a prerequisite for this kind of work rather than a nice-to-have.

What not to do when sales drop

• Don’t discount first. A discount addresses exactly one cause, price resistance, and disguises every other one. It also resets what customers expect to pay, which is very hard to undo.
• Don’t cut marketing to zero. You’ll save a few weeks of cash and extend the revenue problem by months.
• Don’t launch something new. A new product line while the existing one is underperforming splits attention you don’t have and delays the diagnosis.
• Don’t change five things at once. You’ll recover, possibly, and you’ll have no idea why, which leaves you exactly as exposed next time.
• Don’t assume it’s the economy. It’s the most comfortable explanation and it’s usually the wrong one. Rule out the four internal causes before you accept the fifth.

Frequently asked questions

Why are my sales down when nothing has changed?

Something has changed; it’s just not visible from the revenue line. Split the number into customer count, order frequency and average order value, and one of the three will have moved. If none has and revenue is still down, check whether a single large account reduced its volume. One customer can hide a lot in a small business’s totals.

How many months of decline count as a real problem?

Three consecutive periods against the same periods last year. One month is noise, two is a pattern worth watching, three is structural and won’t resolve on its own.

Should I lower prices when sales are declining?

Only if your diagnosis says price is the cause, meaning you’re losing deals specifically on cost and hearing it from customers. A discount driven by cash anxiety is usually deeper than one made by choice, and it’s difficult to reverse.

What’s the fastest fix for declining sales?

Conversion faults, by a wide margin. If demand is arriving and failing to convert, repairing the mechanical break turns existing traffic into revenue within days and costs nothing.

How do I tell a sales decline from a slow season?

Compare year over year rather than month over month. If the same weeks were quiet last year, you have seasonality and the answer is cash planning, not a rescue plan.