Advertising
How to plan an advertising campaign: goals, budget and enquiries.
A campaign plan should explain what you are testing, what you can spend and how the next decision will be made. Choose the channel after defining these foundations.
In this guide
- Name the action and audience
- Connect the ad with what follows
- Write down the hypothesis and limits
- Assess what you learned
- Describe the buying situation before designing an ad
- Choose a channel around how the decision happens
- Build the budget and review threshold from business data
- Write a creative brief that supports actual production
- Build the page around the question raised by the ad
- Connect recorded events with real enquiry assessment
- Separate launch checks from evaluating the experiment
- End the report with a concrete decision
- Questions and answers
Name the action and audience
Awareness and service enquiries are different tasks. Choose the main action the campaign should support. Describe the person’s situation: their problem, what they do not know and what they need to understand before contact.
Then narrow the offer. Advertising everything a company does makes one clear message harder. Start with something you can explain and deliver.
Connect the ad with what follows
The creative promise needs a matching explanation on the page or in conversation. If the ad names a service, customers should not have to search a general catalogue for it. Check mobile layout, content and contact.
Define enquiry handling too: who replies, where the outcome is recorded and how the campaign manager receives feedback. Otherwise you may optimise form submissions rather than useful conversations.
Write down the hypothesis and limits
Define one question for the test. You might compare an argument about the process with one about the finished work. This is a possible comparison, not a proven outcome.
Everyone involved should understand the plan.
- One offer and a defined audience.
- Materials and the differences between variants.
- Media, management and production costs separated.
- A review point and early stop conditions.
- An owner for enquiry handling and quality feedback.
Assess what you learned
After the agreed period, separate facts from interpretation. Describe spending, available events and assessed contacts. Include limitations: little data, technical trouble, an offer change or missing feedback.
The next decision may be to develop the idea, improve the page or stop the test. Discovering an offer mismatch is information too. It should not be presented as confirmed sales growth.
Describe the buying situation before designing an ad
Collect information from the person handling sales: what people ask, when they drop out, which conditions they must meet and what work the company can deliver. Choose one situation where the offer is useful. A homeowner planning a kitchen replacement gives more direction for copy than everyone aged 25 to 65. This is a hypothetical audience description, not a ready-made targeting recommendation. The need, obstacle and decision context matter.
List facts that can appear in the ad: actual scope, available material, documented projects, price with conditions and a contact method. Separate these from assumptions to investigate. If the company has no available appointments, explain availability honestly or choose another service. The campaign must reflect operating capacity. Record business exclusions too: locations, types of work or applications the company cannot support. These details can make the message more precise before money is spent attracting unsuitable requests. Ask the sales team to check that the chosen situation matches conversations they can genuinely handle.
Choose a channel around how the decision happens
In search, you can respond to a stated question about a product or provider. In social media, you may first need to show the problem and explain the offer. Video can demonstrate operation, a work process or differences that a single image cannot easily explain. These are possible uses, not guaranteed format advantages. Check what material exists and whether the company can support the journey after a click.
A hypothetical emergency repair and a planned furnishing purchase need different messages and response speeds. The first may depend on actual local availability; the second on comparing options and preparing for a consultation. Do not split the first budget across many channels just to appear everywhere. Choose a channel whose role you can explain. If several placements are included, define each job and the basis for comparison. The same film used on two platforms still needs its crop, captions and contact journey checked in both contexts. Reuse is useful when the material remains understandable.
Build the budget and review threshold from business data
List platform spend, production, management and website changes separately. Distinguish setup from recurring work. DigiDraft’s price list puts either Google Ads or Meta Ads at PLN 1,500 net setup and PLN 1,500 net monthly management for each individual option, with media additional. A combined package has its own price. Check scope and minimum engagement in the current proposal instead of assuming every campaign fits one line item.
To assess an acquisition cost, you need margin, fulfilment costs and the share of relevant enquiries that become orders. If these are unknown, use clearly labelled scenarios. Do not give every message the value of a completed contract. Set a maximum test spend and name who can change it. A review threshold need not promise profitability. It can specify when you will reconsider the offer, sales process or measurement quality before committing more money. Keep the cost of learning visible, especially when the test may finish before a long sales cycle has produced a confirmed result.
Write a creative brief that supports actual production
The brief needs an audience, one offer, an argument, available evidence and the intended next action. Add restrictions: claims that cannot be made, conditions attached to a price and elements that must remain readable in a small frame. A designer also needs the intended placement and source material. A request for “modern graphics that sell” does not explain these choices or tell the designer how the work will be assessed.
Prepare variants that differ in a named element. A hypothetical installation business could compare a clear process explanation with a completed-project demonstration, provided it has permission to use the material. That is an idea for a test whose result is still unknown. Do not simultaneously change price, audience, page and promise if you later want to attribute the difference to a headline alone. Agree on concepts, revisions, the approver and the deadline for factual input. Review claims and usage rights before submitting the ads for publication. Record the approved version so a later revision is not mistaken for the original test.
Build the page around the question raised by the ad
After clicking, a visitor should recognise the offer they just saw. The first screen names the service and explains the next action. The page then needs scope, important limitations, evidence of work and answers to questions before contact. Order depends on the decision. A simple booking needs a different page from a consultation about a complex service. Do not assume every campaign requires a long landing page or the complete removal of navigation.
Test on a phone with the onscreen keyboard open. Are fields labelled, errors actionable and the button reachable? Check a slower connection, the phone link and confirmation of a successful action. Someone who is not ready to enquire may need scope details, a project example or collaboration terms. Provide that information without creating several competing primary buttons. The form should collect what the next real stage needs and explain honestly how the message is passed on. The confirmation must describe what happened: preparing a message is different from delivering it to the company.
Connect recorded events with real enquiry assessment
Make a short measurement table: action, where it happens, how it is checked and who owns it. Describe visits, form submissions, relevant enquiries and sales separately. Campaign parameters can help identify traffic sources when the chosen system reads and records them. Adding UTM text to a URL does not create a report by itself. Use consistent naming and keep personal information out of these parameters.
Before launch, test a valid enquiry, form failure, retry and delivery to the correct person. Resolve data and consent requirements before activating tools. Then create a simple way for the company to assess contacts. An agreed process recording fit and conversation stage may be enough; an elaborate CRM is not a prerequisite for a first test. Account for purchase delays and differences in attribution methods. If quality feedback is missing, do not describe event totals as acquired customers. Record that limitation as a task with an owner, because the next report will otherwise inherit the same uncertainty.
Separate launch checks from evaluating the experiment
Before activation, the responsible person confirms URLs, materials, audience settings, schedule, billing, budget and the contact route. Save the version approved for delivery so everyone knows what was tested. The first check looks for technical faults and departures from the plan. A wrong destination or unavailable service calls for prompt action regardless of the number of events recorded.
A hypothesis needs suitable observations. If the aim is a causal comparison, plan traffic allocation and use an appropriate experiment mechanism where available. Two independent campaigns need not receive comparable audiences. Choose the main measure and stopping conditions before viewing results. Do not announce victory after a single order or describe a small difference in a tiny sample as certain evidence. An inconclusive result should explain why the question remains open and what decision is possible within the approved spending boundary. Repair technical faults as soon as they are found. Allow enough time for enquiries to be assessed before drawing conclusions about commercial results.
End the report with a concrete decision
A useful report shows spending, observed actions, assessed contacts and limitations. It then explains changes and their reasons. Consider a hypothetical situation in which suitable enquiries arrive but the company does not respond. Fixing handling comes first. If contacts repeatedly ask for a service that is not offered, review the message and targeting. Increasing the budget is not an answer to every diagnosis.
Finish the review with a justified direction: continue the current scope, run another test, repair a particular stage or stop. Give each change an owner, required material and a way to verify it. Record what remains unknown too. A business does not need every channel, advanced automation and many creatives before it can learn from genuine enquiries. For a conversation with DigiDraft, bring the offer, resources, existing data and constraints. We can then discuss what needs preparing, who will handle enquiries and what information you will need to decide whether to continue.
Questions and answers
What should I establish before choosing an advertising channel?
Define the audience, their buying situation and the action the campaign should lead to. Check that the offer, landing page and person handling enquiries are ready. Then compare channels against that specific task.
What should I do if an ad gets clicks but no enquiries?
Follow the journey from the ad to contact on a phone and check that the promise matches the page. Test the form, confirmation and actual message delivery. Then check measurement and audience fit before increasing the budget.
What should a campaign test report contain?
Show spending, the agreed outcome and the quality of enquiries, together with data limitations. Record which hypothesis was supported and where evidence is still missing. The report should lead to a decision to continue, make a specific change or end the test.