Back to insights

    Marketing strategy · 11 min

    SEO or Google Ads on a limited budget: what should come first?

    Start with one offer, a decision deadline and a realistic acquisition allowance. Use the matrix below to decide which work should be funded first.

    A metal path forks towards an emerald glass ramp and dark stone steps with a green sphere.

    Fund one useful answer before funding two programmes

    With a limited budget, choose one offer, one buyer group and one decision to resolve. Google Ads deserves the first test when buyers already search for the service, the landing page is ready and sales can handle enquiries. SEO deserves early attention when relevant organic visibility already exists but the pages fail to answer buying questions. If the form is broken or the offer is unclear, repair that shared foundation first.

    This is a sequencing decision. An industrial maintenance company may need bookings for its next available service window. A specialist equipment manufacturer may be helping a buyer assemble a supplier shortlist for a project six months away. Giving the two businesses the same channel split ignores the job the investment must do. Begin with the offer and the decision deadline, then choose the work that can produce useful evidence.

    The approach suits a business with a credible product or service and insufficient capacity to run several programmes well. The output is a short agreement: what happens first, what stays outside the initial scope, what evidence the work should produce and when the team will decide again. It is narrower than a company-wide marketing allocation exercise. You do not need to solve every acquisition channel before improving one buying journey.

    Separate the decision deadline from the sales deadline

    Write down when you need an answer and, separately, when you need cash from new business. A paid-search test may reveal whether a group of searches attracts relevant visitors before any sale closes. With a long sales cycle, the first month cannot establish a reliable cost per acquired customer. Track accepted enquiries and meaningful sales milestones while preserving the connection to later outcomes.

    SEO work must be implemented, revisited by search engines and encountered by buyers. Do not build a cash-flow plan around a promised ranking date. A new content programme on its own is a poor answer to an urgent capacity gap. Paid search is not an automatic solution either: relevant demand, a competitive offer and a workable sales process still have to exist. Faster exposure is different from faster profitable sales.

    An established page presents a different opportunity. If it already appears for a specific service and sales confirms that buyers need that service, improving its explanation and enquiry path may be the most sensible first step. Inspect the page and its queries. A domain-wide average position cannot tell you which part of the offer has a useful starting point or why a particular visitor failed to enquire.

    Check five prerequisites before choosing a channel

    First, specify the offer. “Industrial services” is broad. “Preventive maintenance for packaging machinery at the customer's premises” gives you something concrete to describe and qualify. Define the territory, equipment types and suitable customer. Otherwise, a report may mix job searches, do-it-yourself repair questions and enquiries from businesses you can actually serve. That mixture is difficult to interpret after the money has been spent.

    Second, make the destination useful. A buyer should understand who the service is for, what information is needed and how to take the next step. Third, verify receipt of the enquiry. Test the form on a phone, confirm the record reaches the business system and name the person responsible for responding. An analytics event triggered by a button click is not proof that the business received a request.

    Fourth, understand the economics of the order: contribution after direct delivery costs, sales effort and capacity constraints. Fifth, establish a feedback process. Someone must review incoming contacts and identify suitable opportunities. Without that distinction, the channel comparison can reward supplier pitches, irrelevant requests or existing-customer support tickets. A small, consistently reviewed sales record is more useful than a complex dashboard nobody trusts.

    Use a first-investment decision matrix

    Apply the matrix to one offer at a time. Add evidence beside each judgement: a landing-page URL, a reviewed enquiry, a report from your account or confirmation from sales. Where the evidence is missing, write “unknown”. That is an input to the scope of the first test, not a reason to invent an answer. An agency should be able to explain which unknown its proposed work will resolve.

    Treat essential conditions as gates before scoring preferences. A promising click forecast does not help if the business cannot deliver in the selected territory. A substantial SEO content package is difficult to execute if no subject expert can approve the material. Under a tight budget, the availability of the person who understands the buyer may constrain the plan as much as media spend does.

    Recommended first task based on evidenced conditions
    Starting pointFirst taskEvidence for the next decision
    Unclear offer or broken formRepair the destination and enquiry handlingThe offer is understandable and a test enquiry is received correctly.
    Relevant searches, ready page, available capacityRun a bounded Google Ads testSales reviews contact quality and the complete acquisition cost.
    Relevant organic visibility, weak buying informationImprove the priority page and its SEOChanges are implemented; relevant search and sales outcomes are monitored.
    Unfamiliar category and unvalidated problemInterview buyers and validate the offerBuyers can describe the problem and a meaningful next step.
    An existing channel produces suitable businessMaintain the working scope; test one improvementThe additional task has a hypothesis and an owner.

    Translate click cost into an enquiry threshold

    The following arithmetic is an illustration, not NUMEDIA performance data or a market benchmark. Assume an order contributes €600 after direct delivery costs and the business chooses to allow €180 for total acquisition. If sales closes one in four qualified opportunities, the planning ceiling is €45 per qualified opportunity. This allowance must also cover an appropriate share of preparation, management and sales handling, rather than being treated entirely as a media allowance.

    Now assume half of received enquiries meet the qualification criteria. The ceiling per received enquiry is €22.50. If three in every hundred paid visitors enquire, the theoretical media-only ceiling is €0.675 per click before deducting the other costs. If actual clicks cost more, these assumptions do not support the plan. Simply reducing bids does not entitle the business to expect the same volume and quality at a lower price.

    Run a downside case as well: only one quarter of contacts qualify and one in six qualified opportunities closes. The allowable cost falls substantially. A plan that works only under the most optimistic assumptions needs a smaller initial commitment and clearer stopping conditions. Where repeat orders matter, use observed history for comparable customers. An imagined lifetime value is not cash already available to fund acquisition.

    Give the Google Ads test a narrow question

    A useful first question is whether one group of service searches brings enquiries with the right territory, project scope and timing. Separate searches for your brand from searches for the service. A low-cost contact from someone who already knows the company answers a different question from a new opportunity arising from category demand. Define which search intentions are outside the test before reviewing its aggregate cost per enquiry.

    Keyword Planner can help estimate possible clicks and spending, but forecasts depend on settings and other factors. They do not guarantee a number of sales enquiries. Likewise, your Search Console impressions describe observed visibility for your own website, not the entire market. They omit searches where your site did not appear. Google's documentation for both tools is included below so the estimates and observations remain distinguishable.

    Before launch, agree the maximum total commitment, the permitted duration and the choices available at review. An average daily Google Ads budget is not a hard cap for every individual day; daily spend can vary. Check the applicable settings and budget report for the campaign. Record changes to budgets, targeting and the offer so that a sequence of different conditions is not later presented as one stable test.

    Give SEO a deliverable list

    For the first SEO scope, select a priority service or category page and a small set of supporting answers that help a buyer decide. Verify that the intended page is accessible to search engines and people. Then address unclear positioning, missing selection criteria and dead ends in the enquiry journey. A quota of new articles should not determine the initial work when a commercially important page already exists.

    Ask for a change list with owners and acceptance evidence. For a maintenance service, the page might need to explain supported machinery, the information required for assessment and how to submit a fault description. Acceptance can verify published content, internal links and the functioning enquiry route. Future rankings are not an implementation deliverable under the supplier's control. Keep completion of the work separate from subsequent performance review.

    Google explicitly distinguishes advertising from organic presence: paying for ads does not buy improved organic rankings. The practical connection between SEO and paid search is the buyer knowledge and destination experience they can share. It is not a ranking bonus. Specialist questions about generative search can be taken to geooptimizacija.si; keep this initial investment decision focused on the specific sales problem you need to resolve.

    Budget for the foundation, the test and the next decision

    Instead of applying a universal percentage split, make three envelopes. The first covers shared prerequisites: the offer, destination, enquiry handling and basic outcome records. The second covers one defined test or implementation scope. The third is reserved for the next decision, an uncovered obstacle or a repeat check. If the foundation consumes all available funding, do not launch an additional campaign that you cannot afford to evaluate sensibly.

    Compare supplier proposals using the same cost categories. For ads, distinguish media, landing-page work, measurement setup, management and sales time. For SEO, distinguish research, expert content preparation, technical implementation and upkeep. Both channels consume paid effort even when there is no charge for an individual organic click. Value internal hours consistently rather than counting employee effort as free in the preferred option.

    If an existing campaign already produces credible business outcomes, do not switch it off simply to start SEO from zero. Identify what should be maintained and restrict the new scope to available capacity. Conversely, do not neglect established organic business because an advertising dashboard is easier to read. The sequence should follow evidence about the offer, including delivery capacity, rather than the convenience of the reporting interface.

    Three businesses, three different sequences

    A maintenance provider has free service slots, a clear offer and someone ready to respond. It does not yet know which searches produce enquiries for suitable equipment. A verified destination followed by a limited search test is a reasonable starting sequence. Questions from those buyers can inform later SEO work, but the first scope need not include a full editorial programme or every available advertising format.

    A specialist supplier already receives organic enquiries, yet buyers repeatedly confuse two product configurations. Its first investment may be a stronger comparison with specifications and an explicit route to a quote. A paid test for a new segment can follow once sales confirms that the main page reduces confusion and the new offer is ready. This sequence improves an existing source of opportunity before funding a new one.

    A business introducing an unfamiliar category cannot identify clear searches and has not validated the customer's problem. Neither a broad campaign nor a large SEO package is the obvious first purchase. Buyer conversations, message testing and evidence of usefulness come earlier. A limited budget does not have to be divided between two channels when neither has a defined job. This is a valid decision, not a failure to choose.

    Review a decision record, not a traffic leaderboard

    Prepare a one-page review. Put the offer, audience, start date and approved commitment at the top. Below, record spending, implemented changes, received enquiries, qualified opportunities, closed sales and cases still open. Add the typical delay to a sales decision and the data you could not observe. Otherwise, a long sales cycle can make a reasonable early test look like a failed first week.

    Choose one of four next steps: continue, change the offer or page, repeat the measurement, or stop. State the reason. Irrelevant contacts call for a review of intention and messaging. Suitable contacts that never receive a proposal call for a review of sales handling. Insufficient evidence does not prove that the other channel is better. It supports a more cautious conclusion and another bounded step, if that step is worth funding.

    To scope this with NUMEDIA, bring one offer URL, your delivery territory, the available investment range and an example of a good enquiry. Identify who approves technical content and who reviews contacts. We can use those inputs to define an achievable first scope and the evidence needed for a further decision. The contact link below opens the existing enquiry form; describe the decision you need help making.

    Sources and methodology

    1. Do you need an SEO? (Google, accessed 2026-11-11)
    2. About Keyword Planner forecasts (Google, accessed 2026-11-11)
    3. About average daily budgets (Google, accessed 2026-11-11)
    4. Get started with Search Console (Google, accessed 2026-11-11)

    Frequently asked questions

    Should a small budget be split equally?

    Not automatically. Fund the necessary foundation and one defined question first. Two incomplete programmes may produce less useful evidence than one properly scoped piece of work.

    Is SEO free after publication?

    An organic click has no equivalent media charge, but content development, expert review, implementation and upkeep consume resources. Include internal team time consistently in the comparison.

    Does buying ads improve organic rankings?

    Paying for Google Ads does not buy better organic presence. The useful overlap is a clear offer, a better destination and the buyer questions uncovered during the work.

    When should we call a test unsuccessful?

    Agree the maximum commitment, review period and stopping conditions before launch. Fix technical faults immediately; evaluate business outcomes with the sales delay and evidence volume in mind. A handful of contacts cannot support a confident channel comparison.