Planning pay per click advertising in Belfast: budget, tracking and decisions

1. Write the question your paid search test must answer

A useful paid search test starts with a question. Can a specific offer attract suitable enquiries within a defined service area at an acceptable cost? That is more actionable than asking whether advertising works. It establishes what you are testing and which evidence should influence the next decision.

Pay per click describes how advertising is charged. It does not explain what a campaign is trying to achieve. When considering paid search advertising in Belfast, distinguish a campaign aimed at people seeking a service from activity intended to introduce a new product to an unfamiliar audience. The message, landing page and measures may need to be different.

Choose one initial offer and explain who it is for. Write down the practical boundaries: areas served, opening hours, availability, minimum project requirements and exclusions. If your team cannot take a particular type of work, advertising for it creates handling costs as well as wasted media spend. State these constraints before building a keyword list.

Define what a successful test would let you do next. You might continue the offer, change a landing page, improve enquiry handling or stop because the economics are unsuitable. A decision to stop can be a useful result when the test was controlled and the evidence is trustworthy. More activity is not automatically the right next step.

Make an assumptions list. Record which customer need is proven, which conversion rates come from your own records and which numbers are simply planning scenarios. That prevents a spreadsheet from acquiring false authority as it moves between the owner, agency and finance team.

2. Work backwards from an affordable customer acquisition cost

Start with the economics of the work you want, rather than a competitor's apparent advertising budget. Consider the contribution available after the direct costs of delivering a typical job. Decide how much of that contribution the business can reasonably allocate to acquiring a customer, leaving room for overheads and profit. Use your own records where possible.

For an entirely hypothetical example, suppose a service contributes £600 before acquisition costs and the business chooses a £150 acquisition allowance. If one in four suitable enquiries becomes a customer, the corresponding allowance is £37.50 per suitable enquiry. These numbers explain the arithmetic only. They are not local market rates, recommended budgets or predicted campaign results.

Multiply the amount you can afford to spend acquiring a customer by the proportion of suitable enquiries that become customers. This gives an initial allowance per suitable enquiry. If the close rate is uncertain, calculate several scenarios. Converting one in eight enquiries produces a different allowance from converting one in four. Check which assumption your plan depends on most heavily.

Include the cost of managing advertising and handling enquiries when assessing overall viability. A platform's cost per conversion commonly describes media spend against a configured action, not the complete cost of acquiring a paying customer. A form submission can be duplicated, unsuitable or never answered. Sales records are needed to understand what happened afterwards.

Do not turn a calculated allowance into a bidding instruction without checking the measurement and available data. It is a business constraint to guide judgement. Ask whoever manages the account how they will work within it and how they will flag evidence that the original economics need revisiting.

3. Set a spending envelope and a separate loss limit

Create a total test budget with separate lines for media, management, measurement work, landing page changes and internal sales time. Label which amounts are fixed and which depend on activity. This helps avoid a test that appears affordable because the setup and work responding to enquiries have been left out of the calculation.

Set the maximum amount the business is prepared to spend to learn, including the possibility of no sales. That is your loss limit. Choose it according to your resources and uncertainty, not a generic rule that every business needs a particular monthly amount. If the available sum cannot support a meaningful test, reduce the scope or improve the offer before launch.

An average daily budget is not necessarily a strict spending cap for each day. Google Ads documents daily overdelivery and monthly charging limits. Ask the account manager to explain the rules that apply to your campaign and how budget changes affect the plan. Keep the current official guidance beside the agreed financial controls rather than relying on a remembered platform rule.

Give someone authority to change spend and define when additional approval is required. Record a pause route for unexpected demand, staff absence or a broken website. A small business can have a perfectly good campaign and still need to stop taking enquiries because delivery capacity has changed.

Your worksheet should contain the approved envelope, planned dates, responsible person, review checkpoints and explicit stop conditions. Do not divide a national keyword estimate by an assumed click price and present the answer as a local traffic forecast. Real auctions, eligibility, competition and relevance affect delivery, and a planning number does not remove those uncertainties.

4. Keep the first campaign focused on relevant demand

Begin with the language a suitable customer uses when they are looking for the service you actually provide. Group searches by intent: booking a job, researching options, seeking employment, finding training or looking for a free resource. Some research searches may eventually be useful, but a tightly scoped first test should make its priorities explicit.

A hypothetical maintenance business might begin with one service and the locations its team can reliably cover. Searches about jobs, qualifications and instructions for doing the work yourself would need separate consideration. This is not a keyword list to use without checking: the right boundaries depend on the business. Ask the manager to explain how irrelevant intent will be identified and excluded.

Location settings need attention too. Ask how the campaign defines the people it can reach and how actual enquiry locations will be checked. Do not assume that including Belfast in an ad means every click or enquiry comes from someone you can serve. Review the account configuration and the information collected at enquiry stage together.

Keep searches that name your company and searches that do not understandable in reporting. Someone who already knows your company may behave differently from someone discovering it through a service search. Combining everything into one attractive average can hide whether the test is reaching a new audience or mostly responding to existing recognition.

Avoid changing several major variables at once. If you expand locations, introduce a new offer and rebuild the landing page in the same week, interpreting a change in results becomes harder. Keep a dated change record with the reason for each adjustment. You can then connect observations to decisions without pretending the campaign is a laboratory experiment.

5. Make the landing page ready for a real customer

The page reached after an ad click should continue the same conversation. State the service, relevant area, important limitations and next step clearly. A customer should not have to search through a general website to discover whether the advertised offer applies to them. Review the page as a customer who has never heard of the business.

Use evidence you can support. Show genuine work, accurate service information and permitted customer feedback. Do not add invented project results or vague claims of market leadership to make a campaign sound stronger. If the offer depends on an assessment or quotation, explain that instead of implying an immediate fixed outcome.

Check forms on a real phone. Can someone enter their details comfortably, correct an error and understand what happens next? Ask only for information needed to assess or respond to the enquiry. A very short form may increase submissions while reducing their usefulness; an unnecessarily demanding form can discourage suitable people. Decide which information earns its place.

Telephone enquiries need a process as well. Confirm who answers, what happens outside business hours and how missed calls are reviewed. A count of clicks on a telephone link is not proof that a conversation took place. Where call measurement is used, understand the setup and interpret it alongside the team's actual records.

Before launch, complete the journey with a test enquiry and remove or clearly identify that test in reporting. Verify receipt, notifications, confirmation messages and ownership. If an enquiry enters a customer relationship management system, make sure it reaches the correct person. A technically recorded conversion has little value if nobody follows it up.

When the destination needs work, our website design expertise helps connect the campaign message with the enquiry journey.

• Check the ad promise against the landing page.
• Test a form submission, error state and confirmation.
• Confirm that the intended team receives and owns the enquiry.
• Check telephone handling during advertised hours.
• Identify test submissions so they do not distort reporting.

6. Define conversions and reconcile them with business records

A conversion is an action you choose to measure as valuable. Google Ads supports measurement of actions such as website enquiries, purchases and calls. The important planning decision is which actions deserve to guide optimisation and which are useful observations. A page visit and a suitable enquiry should not silently become interchangeable in a report.

Write a measurement specification in ordinary language. Name the event, what triggers it, how often it should count and how it will be tested. Specify the difference between clicking a submit button and successfully submitting a form. Include what should happen when a visitor encounters an error, refreshes a page or submits twice.

Then add the business classification: received, duplicate, unsuitable, suitable, quoted, won or lost. Use a small set of clear definitions so different team members classify enquiries consistently. For example, a suitable enquiry might require the right service, a location you cover and a genuine contact route. Agree the definition before judging the campaign.

Platform reports, website analytics and sales records may show different totals because they measure different things. Reporting periods and attribution choices matter as well. Ask the manager to explain material differences and maintain a documented source for each decision. Do not force figures to match by changing definitions after the fact.

Measurement also involves personal information and tracking choices. Decide what data is necessary, who receives it and which checks your organisation needs before collection or sharing. The guidance on data protection by design from the Information Commissioner’s Office is a starting point for assigning responsibilities. A marketing worksheet still needs to sit within your organisation’s wider approach to data protection.

7. Review in stages without overreacting to small numbers

Use the first review to check delivery and technical reliability. Are ads eligible, are enquiries reaching the team and are the recorded actions behaving as intended? This is different from judging profitability. A broken form needs prompt action even when there are too few enquiries to assess business performance.

The next review should examine relevance. Look at the intent of the searches where information is available, the locations being served and the reasons enquiries are rejected. An unsuitable lead may point to an unclear ad, a missing exclusion, a confusing page or a genuine mismatch between demand and the offer. Record the reason before choosing the remedy.

Review commercial outcomes only after allowing for the real sales cycle. If quotations normally take several weeks to become orders, a sales total from the same week is an incomplete view. Keep older enquiry groups visible so later outcomes are not lost. Explain which results are still unresolved instead of treating them as either failures or wins.

Small samples move sharply. One additional sale can transform an early calculation of the cost of acquiring a customer. Show the underlying numbers alongside percentages, and avoid announcing a trend from a handful of events. Set a review date, but allow the conclusion to be that there is insufficient evidence for a confident expansion.

Write a short decision note after each review: what was observed, the likely explanation, what remains unknown, the agreed change and the next check. That note is more useful than a series of unexplained budget adjustments. It also helps a new team member understand why the account looks the way it does.

Use a simple review worksheet with five columns: observation, evidence, possible explanation, action and owner. For example, several enquiries outside the service area are an observation; the locations entered in the enquiry record are evidence. Possible explanations include settings, unclear copy or people making enquiries on behalf of someone elsewhere. Inspect the relevant evidence before changing the campaign. The action might be a clearer statement of the area served, with the manager responsible for checking the next group of enquiries. This keeps reasonable hypotheses separate from established causes and leaves a useful record of the team's judgement.

8. Use a hypothetical test plan to connect the pieces

Consider a fictional repair business with one clearly defined service area. It can handle ten additional suitable enquiries during the test period. The owner wants to understand whether search advertising can attract that type of work, but has no reliable advertising history. This is a planning example and contains no forecast or delivered client result.

The initial brief names one service, the customer problem and exclusions. The page explains what the team can repair, where it travels and how assessment works. The enquiry form requests contact details, location and a short description. A staff member checks each enquiry and records suitability, next action and eventual outcome.

The budget worksheet separates media spend, setup and management. The owner approves a total learning envelope that the business can afford to lose. The campaign manager documents the applicable platform budget rules and who can authorise changes. The team also agrees to pause if the form stops working or delivery capacity is exhausted.

At the first review, the team checks whether the measured enquiries are real and received. At the next, it reviews suitability and recurring questions. If people misunderstand the repair scope, the response may be clearer copy rather than additional spend. If suitable enquiries arrive but nobody answers promptly, the process for responding to enquiries becomes the priority.

The final review considers resolved and unresolved enquiries separately. The owner can choose to continue, revise or stop, with reasons recorded. The test has value because it leaves a clearer understanding of demand and the business process. It does not need a dramatic headline percentage to justify an honest decision.

Give the fictional test a specific handover exercise. Another authorised person should be able to open the account, find the approved budget, identify the conversion definitions and locate the record of enquiry quality. They should know how to pause activity and whom to contact if the website fails. Check this before the regular manager is absent. The point is not to encourage unplanned account changes, but to ensure spending that is important to the business does not become inaccessible when one person is unavailable. Record the agreed emergency route beside the normal review schedule and keep access current.

9. Agree ownership and the next decision before launch

Keep business access to the advertising account, measurement setup, billing records and landing pages. Establish which party creates each account and who can grant or revoke access. Do not wait until a supplier changes to discover that nobody in the business can inspect historical performance or update payment arrangements.

Ask for a handover checklist in the initial scope. It should cover campaign structure, conversion definitions, change history, exclusions, audience assumptions, creative assets and unresolved issues. Include documentation of website changes and any tools from other suppliers. A successor needs to understand the reasoning behind the work as well as see the current settings.

For ongoing reporting, require a concise decision summary alongside the figures. State spend, valid enquiries, qualification status, resolved sales outcomes and key uncertainties. Keep definitions consistent over time. If a measurement change alters comparability, mark the date clearly so a future reader does not mistake the new definition for improved performance.

Before increasing spend, check operational capacity and whether the evidence supports the proposed expansion. A campaign can be relevant yet commercially unsuitable, or promising yet too uncertain to scale. Ask which assumption the next stage will test and what additional money would make possible. More budget should have a reason beyond reaching the end of the first month.

Podium's growth service page gives the service context for a paid search discussion. Bring a specific offer, your service boundaries, available capacity and any existing lead records. Those inputs make it possible to discuss a controlled test and sensible measurement without inventing click prices or treating broad market estimates as a promise of local customers.

When a test is stopped, retain a concise learning record. Include the dates, offer, service area, major changes, measurement limitations and reasons for the decision. Remove access that is no longer needed while preserving the business's reporting history. A future campaign should be able to learn from this work without treating every old assumption as current. If the offer, website or sales process changes substantially, explain why the previous evidence may have limited relevance. That makes both continuing and stopping a responsible business decision.

Use the test plan to discuss paid advertising and Google Ads management, including the media budget, management scope and reporting responsibilities.

• A written test question and a definition of a suitable enquiry.
• A complete spending envelope and named spending authority.
• Verified enquiry receipt and measurement checks.
• Agreed review dates, pause conditions and sales feedback.
• Business account access and a usable handover record.