Marketing Agency

A customer sees your post, searches your business later, asks a friend, then calls to book. Which marketing channel gets credit? Often, you can’t know. The journey may cross devices, happen over weeks, or include a conversation nobody records.

That does not mean you should set vague goals or give up on measurement. It means your goals should reflect what you can reasonably observe—and should not pretend that a tracking report tells the whole story.

When people search for how to set marketing goals without attribution, they are usually facing a practical problem: marketing may be influencing sales, but the business cannot reliably connect each sale to one campaign. The answer is to set goals at more than one level, combine a few useful signals, and make decisions in proportion to the strength of the evidence.

This approach is especially useful for small businesses, local services, and lean teams. It won’t prove exactly which touchpoint caused every sale. It can help you decide what to keep doing, what to test next, and where not to spend more yet.

First, separate a business goal from a marketing signal

A business goal describes a result the organization needs: more qualified enquiries, repeat purchases, profitable bookings, or revenue from a particular service. A marketing signal is something you can observe that may indicate progress toward that result: a completed enquiry form, a rise in branded searches, or a prospect mentioning a recent campaign.

These are related, but they are not interchangeable. More website visits are not automatically more customers. More enquiries may not help if they are a poor fit or take too much staff time to handle. A useful measurement plan keeps the business result visible while choosing signals that are available and relevant.

AMEC’s Integrated Evaluation Framework distinguishes activities and outputs from audience responses, outcomes, and broader impact. That distinction is helpful here: publishing a guide is an output; a qualified prospect using it before contacting you is closer to an outcome. Neither should be casually described as a sale caused by the guide.

Why perfect attribution is often unrealistic

Attribution tries to assign credit for a result to one or more marketing interactions. In practice, a small business may have missing or unreliable data at several points:

• A customer first hears about you in person, then visits your website later.
• A family member or colleague shares a link, but the eventual buyer arrives through a search engine.
• A phone enquiry is not connected to the webpage, ad, or event that prompted it.
• Several people influence a purchase, while analytics records only one device or session.
• Customers remember the business but not the specific message or channel.

Tools can provide useful clues, but a channel report is not a complete record of how people make decisions. A platform may claim credit based on its own tracking rules; your website analytics may assign credit using a different rule. Treat these as models, not a neutral view of every customer’s path.

There is also a cost to collecting more data. Extra tracking can take staff time, create privacy obligations, and still leave gaps. Before adding a complicated measurement setup, ask whether the answer would change a decision you actually need to make.

Build goals at three levels

A practical goal-setting system has three layers: the business outcome, the marketing contribution you hope to make, and the indicators you can monitor. Keeping the layers separate makes it easier to stay ambitious without overstating what your evidence proves.

1. Name the business outcome

Choose a result that matters to the business, not just a number a platform makes easy to export. For a local accountant, that might be more suitable consultations for a specific service. For a small online shop, it could be more returning customers or increased sales of a particular product line.

Make the outcome specific enough to guide decisions. “Grow the business” is too broad. “Increase qualified enquiries for bookkeeping packages over the next quarter” gives you a segment and a time period to work with. If the business has a revenue or capacity target, note it—but do not assume every part of it can be attributed to marketing.

2. Describe marketing’s intended contribution

Next, state how marketing is expected to help. For example: “Make more local business owners aware of our bookkeeping package and give interested prospects a clear way to request a consultation.” This describes a plausible contribution without claiming marketing is the only influence on a purchase.

This step is valuable when sales also depend on seasonality, pricing, referrals, staffing, or the quality of follow-up. Marketing can influence demand and support decisions, while other factors affect whether an enquiry becomes a sale.

3. Choose a small set of observable indicators

Select indicators that connect the activity to the intended contribution. You might track the number of qualified enquiries, attendance at a relevant event, visits to a service page, and how often new enquiries mention a particular campaign or referral source.

Use a few measures that answer different questions, rather than a long dashboard of overlapping numbers. One indicator might show whether people noticed the campaign; another might show whether they took a meaningful next step. A business outcome, such as completed sales, remains important even if you cannot assign its source precisely.

Write a goal that is specific without pretending to know more than you do

A useful goal has an audience or offer, a timeframe, an intended result, and a measurement method. It should also make room for uncertainty. For example:

Over the next 12 weeks, increase qualified enquiries for our commercial cleaning service from an average of 8 to 11 per month, while recording enquiry source when customers volunteer it and reviewing sales outcomes monthly.

This goal does not claim that a specific channel will create three extra sales. It gives the team a business-relevant target and a repeatable way to review progress. The numbers should come from your own baseline and capacity, not a generic benchmark.

If you do not have a dependable baseline, start by measuring the current level. You can set a learning goal for the first month—such as consistently recording enquiry type and source—then choose a performance target once you understand the starting point. That is not avoiding a goal; it is reducing the risk of building a target on a guess.

Choose measures that fit the decision

Not every campaign needs the same scorecard. Begin with the decision you expect to make. Are you deciding whether to continue an activity, improve its message, or increase its budget? Then choose the smallest set of measures that can inform that decision.

For awareness and reach

Possible signals include reach within a relevant audience, branded search activity, direct visits, or customer responses to “How did you hear about us?” Each has limits. Reach is not proof that people paid attention, and direct traffic can include people who already know the business. Look for a pattern across signals rather than treating one as a complete answer.

For consideration and enquiries

Track actions that show intent: calls, consultation requests, quote requests, email replies, or attendance at an information session. Agree on what counts as qualified. For example, a qualified enquiry might come from a business in your service area that needs the service within a realistic timeframe. A raw form count can rise while useful opportunities stay flat.

For sales and retention

Review total sales, repeat purchases, average order value, cancellations, and the quality of customers where those measures fit the business. Compare periods carefully. A quiet month may reflect seasonality or limited availability rather than a marketing failure. Record major changes—such as a price increase, new competitor, or reduced opening hours—so the review has context.

For a broader planning process, Nomao’s 90-day marketing plan guide can help you connect activities, owners, and review dates.

Use several imperfect signals instead of one false certainty

When sale-level attribution is incomplete, use a simple evidence stack. Combine what customers say, what they do, and what the business records. No single source has to carry the whole conclusion.

1. Ask consistently. Add a short “How did you hear about us?” question to enquiry forms or ask it during calls. Include “Not sure” and “Other” so customers are not pushed into a false answer. Treat self-reported sources as useful memory, not exact tracking.
2. Record campaign context. Note which offer, event, content piece, or promotion was active when an enquiry arrived. If a customer names a campaign, record that too. Keep the process brief enough that staff will actually use it.
3. Watch relevant behaviour. Review visits to important pages, email replies, event attendance, or repeat visits alongside enquiries and sales. These indicators can help explain a change, but they do not prove what caused it.
4. Compare with a sensible baseline. Use a previous period of similar length where possible. Note seasonality and changes in capacity. A comparison is more useful when you explain what else changed during the period.

Suppose a home renovation firm runs a local workshop and updates its project gallery. Over the next two months, several attendees request consultations, more prospects mention the workshop, and visits to the gallery rise. That is a credible indication that the activity is helping generate interest. It still does not establish that every new project came from the workshop—or that the gallery alone caused the change.

Use language that matches the evidence: “enquiries rose during the campaign,” “some customers reported seeing the event,” or “the activity appears to be contributing.” Reserve “caused” for situations where your method can support a causal conclusion.

Set a review rhythm, not a reporting ritual

Measurement is useful when it changes what you do. Choose a review schedule that matches the pace of the channel and the sales cycle. A weekly check may catch a broken form or a campaign spending too quickly. A monthly or quarterly review may be more appropriate for a service with a long decision process.

At each review, ask three questions:

• What changed? Compare the chosen indicators with the baseline and note unusual events.
• What is the strongest reasonable explanation? Consider marketing activity alongside seasonality, pricing, staffing, referrals, and sales follow-up.
• What should we do next? Continue, adjust, pause, or run a more focused test—and state what evidence would change your mind.

A useful review can lead to a small adjustment rather than a major strategic pivot. If a campaign brought in enquiries but few were suitable, refine the audience or message before declaring the channel a failure. If the activity generates attention but no meaningful next steps after a fair trial, investigate the offer and landing experience—or stop investing.

Use tests when the decision is important

When you need stronger evidence, design a modest test around a specific uncertainty. A local service business could promote one service in one area for a defined period, while keeping its usual activity elsewhere. A retailer could compare two email messages sent to similar groups. The aim is not to create a perfect experiment; it is to reduce the number of explanations for a result.

Be realistic about what the test can show. Small audiences, changing demand, and overlapping campaigns make results noisy. If you change the offer, audience, creative, and channel at once, you may learn that the package changed performance but not why. For a limited-budget team, testing one meaningful difference at a time is often more useful than trying to imitate a large company’s experiment program.

Before launching a new channel or test, check whether you have enough time to run it consistently and respond to the demand it creates. The new marketing channel checklist helps teams consider readiness and practical constraints.

Common mistakes to avoid

Setting activity targets as if they were outcomes

“Publish eight posts” is an activity target. It may help the team plan its work, but it does not say whether the posts are reaching likely customers or helping the business. Keep activity targets as operational commitments, then pair them with an audience or business signal.

Using every available metric

A crowded report can make it harder to spot what matters. If a measure does not affect a decision, remove it from the regular review. You can still keep operational data available for troubleshooting.

Changing the target after seeing the result

If you redefine success once the numbers arrive, it becomes difficult to learn. Set the goal and review date in advance. If circumstances change, record the reason for adjusting the target rather than quietly replacing it.

Giving all credit to the easiest-to-track channel

A click or last recorded visit may be visible, while a recommendation or earlier conversation is not. Don’t confuse visibility in a report with influence on the decision. Consider how the customer learned about you and what else happened during the buying process.

Expecting every campaign to pay back immediately

Some activity supports future demand, trust, or repeat business rather than a same-week purchase. That does not mean every brand-building activity deserves unlimited time. Define a reasonable review period and use evidence suited to the kind of effect you expect.

A simple worksheet for your next goal

Write down the following before you start:

• Business need: What result matters, and why now?
• Audience and offer: Who are you trying to reach, and what are you asking them to consider?
• Marketing contribution: How should marketing help, without claiming it controls the whole outcome?
• Baseline: What is happening now, and how reliable is that information?
• Indicators: Which two to four signals will you review?
• Timeframe and owner: When will you review the result, and who will record the data?
• Decision rule: What would lead you to continue, change, or stop?

For example, a two-person design studio may want more projects from a particular industry. It could set a quarterly goal for qualified consultation requests, add an optional source question to its enquiry form, and review project wins and losses at the end of each month. The studio might also track visits to a relevant case study, but should treat those visits as supporting evidence—not proof that the case study generated a client.

Keep the process proportionate. If a spreadsheet and a short monthly discussion answer the question, there is no need to build a complex dashboard. If the budget or risk is substantial, more careful measurement may be worth the time. A sensible marketing budget should account for both the activity and the effort needed to assess it; see Nomao’s guide to setting a small-business marketing budget.

FAQ

Can I set marketing goals without knowing which channel caused each sale?

Yes. Set a business outcome, describe marketing’s intended contribution, and choose observable indicators such as qualified enquiries, repeat purchases, or customer-reported sources. Be clear that these measures show progress or association, not necessarily cause.

What should I measure if customers do not remember where they heard about me?

Offer a “not sure” option and do not press customers to guess. Combine their answers with enquiry counts, sales patterns, and relevant campaign activity. The aim is to build a useful picture, not to force every sale into a channel category.

How many marketing KPIs should a small business track?

There is no fixed right number. Start with a small set—often two to four indicators tied to a specific decision. Include at least one outcome or meaningful customer action, and remove measures that do not affect what you do next.

Should I use revenue as my only marketing goal?

Revenue matters, but it may be too far removed from the activity to diagnose what needs attention—especially when sales cycles are long or attribution is incomplete. Pair revenue with nearer-term measures such as qualified enquiries, conversion steps, or repeat purchase behaviour.

How long should I wait before judging a campaign?

Match the review period to the customer decision cycle and the amount of activity you can reasonably generate. Check early for operational problems, but avoid judging a small campaign on too little evidence. Set the review date before launch and note major factors that could affect results.