Digital marketing FAQ

A full-service digital marketing agency plans, executes and measures every online channel a business uses to acquire and keep customers — paid advertising, SEO, content, video, social, email and messaging automation, ecommerce, web development, conversion optimization and analytics. The meaningful difference between 'full-service' and 'many services' is whether the work is reported against one revenue number or five separate channel dashboards.

Hire a freelancer when you need one channel done well and can manage them yourself. Build in-house when a channel is core to your business, you can afford the salary plus the ramp, and you can retain the person. Hire an agency when you need three or more channels coordinated, want senior capability faster than you can hire it, or need capacity that flexes. Most companies eventually run a hybrid: in-house owns brand and product marketing, an agency owns performance execution.

Ask who specifically will do the work and whether they are on the call. Ask for a forecast with a low case, not just a target. Ask what they would tell you not to do. Ask who owns the ad accounts. Ask for a client in a similar situation to speak to. An agency that answers all five without hesitation is a very different proposition from one that answers with a case-study deck.

Guaranteed rankings or guaranteed ROAS. Accounts built inside the agency's own manager account rather than yours. Reporting that only shows platform-attributed conversions and never reconciles to revenue. A pitch team you never see again after signing. Long lock-in with no break clause. And percentage-of-spend fees combined with a recommendation to increase spend in month one.

You should — always. At Credex Media we work inside your Google Ads, Meta Business Manager, GA4 and CRM under agency access that you grant and can revoke at any time. Creative source files, documentation and playbooks are delivered to a drive you control. An account's conversion history is a genuine asset, and agencies that build inside their own accounts leave you starting from zero.

Most growing businesses invest 7 to 12 percent of gross revenue in marketing. Ecommerce brands in competitive categories often run 15 to 20 percent during an active growth phase. Early-stage companies buying a first market position sometimes exceed that deliberately. The more useful question is the split: a practical early-stage allocation is roughly 55 percent paid acquisition, 20 percent content and SEO, 15 percent creative production and 10 percent lifecycle and conversion work.

For paid search in a competitive category, roughly $1,500 to $2,000 per month in media, because below that you cannot generate enough conversion data for automated bidding to work. For paid social, around 20 to 30 times your target cost per acquisition per week. For SEO, budget matters less than time — a modest retainer sustained for nine months beats a large one abandoned at month four.

We would argue not. A percentage fee rewards the agency for increasing your spend rather than improving your efficiency, and creates a conflict every time the right recommendation is to spend less. Flat retainers in bands, or fixed project fees, keep the incentives aligned with your outcome rather than your budget.

Look at customer acquisition cost against lifetime value and at payback period. A CAC-to-LTV ratio worse than 1:3 usually indicates over-spending or under-monetising. A payback period longer than twelve months constrains how fast you can safely grow. Neither number can be calculated from platform dashboards alone — they require revenue and margin data, which is why we ask for it before quoting.

Technical and on-page fixes on an established site can move rankings in four to eight weeks. Content-driven ranking for competitive commercial keywords typically takes four to seven months. Terms that depend heavily on domain authority take nine to twelve months or more. New domains take longer. Anyone promising competitive page-one rankings in 30 days is either targeting keywords nobody searches or doing something that will eventually cost you the site.

Yes, and arguably more than before — but the definition has to widen. AI systems retrieve and cite from the open web, and the sites they cite are overwhelmingly the ones that also rank well and publish clean structured data. What has changed is that ranking alone no longer guarantees the click, so the work now includes being quotable: direct answers, verifiable statistics, comprehensive schema and consistent entity signals.

SEO optimises for ranked positions on a search results page. GEO — generative engine optimization — optimises for being cited inside AI-generated answers. AEO — answer engine optimization — targets direct-answer surfaces such as featured snippets, People Also Ask and voice assistants. LLMO is a newer synonym for GEO. They overlap by roughly 60 percent, so running them as one programme is cheaper and more effective than buying four products.

It is a real trade-off. If your revenue depends on ad-supported pageviews, blocking has a defensible logic. If your revenue depends on being discovered and shortlisted by buyers, blocking removes you from a surface where discovery increasingly happens. Most marketing-led businesses should permit AI crawlers on commercial and educational content and consider restrictions only on genuinely proprietary material.

Consistency beats volume. A cluster of eight to twelve genuinely useful pages on one subject, published over a quarter and properly interlinked, outperforms forty scattered articles. For most B2B businesses, four to eight substantial pieces a month is sufficient. What matters far more than count is whether the pages demonstrate real first-hand expertise.

If people already search for what you sell, start with Google Search — it is the cheapest way to buy existing intent. If demand does not yet exist, start with Meta or short-form video, because you have to create demand before you can capture it. If you sell to businesses with contract values above roughly $10,000, LinkedIn is worth its premium. If you sell physical products on marketplaces, marketplace advertising usually beats everything else on efficiency.

There is no universal answer because it depends entirely on gross margin. Break-even ROAS is one divided by your gross margin: a 40 percent margin business breaks even at 2.5×, a 70 percent margin business at about 1.4×. Any benchmark quoted without reference to your margins is decoration.

Different attribution models, different lookback windows and different conversion definitions. Meta counts view-through conversions Google does not. Each platform sees only its own touchpoints, so their claims overlap and sum to more than your actual sales. Divergence of 15 to 20 percent is normal; numbers that do not reconcile to your revenue at all indicate an implementation problem.

Yes — for many consumer and ecommerce businesses Meta remains the largest single source of new customers. What stopped working is the old playbook of narrow interest targeting and one static image. Accounts that perform now produce genuinely new creative every month and send clean server-side conversion data through the Conversions API.

Three to four weeks for a restructured account to exit learning phases and produce stable data, and eight to twelve weeks before the programme can be judged fairly. Accounts with high conversion volume stabilise faster. Judging a paid account in its first fortnight measures luck.

Through incrementality testing. The cleanest method is a geo holdout: switch a channel off in matched regions and measure the difference in total sales. Platform conversion-lift studies and spend-down tests are lighter alternatives. Without one of these, all attribution is an educated estimate presented with unearned precision.

For most businesses, contribution margin after media — gross profit minus marketing spend. It is immune to attribution disputes, it cannot be inflated by a platform, and it is the number that actually funds the business. Marketing efficiency ratio, total revenue divided by total ad spend, is a useful simpler proxy.

If you spend meaningfully on paid media, yes. Browser-based tracking loses a substantial share of conversions to privacy features and ad blockers, and that loss degrades both your reporting and the quality of automated bidding. Server-side tagging typically recovers a large part of the gap and often pays for itself through better bidding alone.

A short written update weekly, a substantive review monthly, and a business review with a re-forecast quarterly. Daily dashboards should exist and be self-serve. Anything more frequent than weekly as a formal report is theatre; anything less frequent than monthly means problems compound before anyone notices.

Three months for single-channel retainers, because meaningful paid or SEO data does not exist before then. Multi-channel growth programmes run on six or twelve month terms with quarterly break clauses. One-off projects — a website build, a listing sprint, a video package — are quoted as fixed-scope work with no retainer at all.

Forty-five minutes, no deck. We ask about margins, current channels, what you have tried and what good looks like in twelve months. You leave with at least two specific things to fix whether or not you hire us — usually a measurement gap and a budget allocation issue. If we do not think we are the right fit, we say so on the call.

Most clients do. Common entry points are Google Ads, Meta Ads or SEO, and the common second addition is email and WhatsApp automation, because it is the fastest way to raise the return on media you are already buying. There is no penalty for adding or removing services at a month boundary after the initial term.

Yes — India, the United States, the United Kingdom, the UAE, Australia, Canada and Singapore. Reporting, calls and creative are delivered in English, and account teams maintain at least four working hours of overlap with your time zone.

We tell you before you notice and we say why. If a channel is not working we would rather recommend cutting it than defend it. If the whole programme is not working, we will say that too. After the initial term you can leave with 30 days' notice and we hand over documentation, playbooks and the roadmap.