Digital Marketing Strategy Consultancy For Search-Led Growth

Attribution is the right tool for choosing campaigns, keywords and creative inside a channel, and the wrong one for setting the split between channels. GA4 now offers only data-driven attribution and two last click models, after the first click, linear, time decay and position-based models were retired in November 2023. Direct visits get no credit unless the whole path is direct, and exposure that never produced a tracked visit is invisible to it. We use attribution for what it does well, and no further.

When a large budget line has never been proved, we test it. Google Ads Conversion Lift can split test groups by users or by geography, and the geography version supports offline conversion data. Geo tests, switching spend off or up in chosen regions, work for any channel. We prioritise by spend multiplied by doubt and agree the smallest lift worth detecting before launch. Brand search is often first, as our article on testing whether brand bidding adds sales explains.

Marketing mix modelling estimates each channel's contribution to total sales, alongside price, promotions and seasonality, without user-level tracking. It needs history and variation. Google's Meridian documentation works through a model with 26 parameters and two years of weekly data, four data points per parameter, and calls that too low to estimate reliably. We tell you honestly whether a model is worth building yet. Our guide to marketing mix modelling, attribution and incrementality covers the trade-offs.

AI answers influence buyers who may never click, so last click reporting understates them. We track AI share of voice across a defined prompt set, separate referral visits and conversions from AI assistants in GA4, and watch branded search demand for signs that AI mentions are reaching people. That puts AI search into the same budget conversation as paid and organic search, judged on the same commercial targets, as set out in our guide to measuring LLM traffic in GA4.

We set the role of each search channel against your commercial targets: where SEO captures demand over the long term, where PPC buys speed and control, where Digital PR earns the authority and mentions that feed both rankings and AI answers, and where GEO work is needed to be named by assistants. The output is one plan with priorities ranked by expected revenue, rather than four channel plans competing for one budget. Our piece on SEO and PPC as one plan shows how shared data changes the decisions.

Annual splits rarely survive a trading year. We set channel budgets from the best evidence available, then move them as results come in, judged on blended cost of acquisition and revenue rather than each platform's own conversion claims. At campaign level, Google Ads Performance Planner simulates relevant auctions over the last seven to ten days and may propose a zero budget for campaigns that add nothing to efficient spend. We treat tools like this as one input, never as the strategy.

We design measurement around the budget decisions you need to make: attribution for daily optimisation, incrementality tests for the largest unproven spend and, when the history supports it, a marketing mix model to set the channel split. Open-source tools such as Google's Meridian, made available to everyone in January 2025, and Meta's Robyn cut licence costs, but they still need clean weekly data and an analyst who understands the assumptions. We are clear about which method you need now and which can wait.

Every measurement method depends on conversions and revenue being recorded consistently. We check that GA4 key events, ad platform conversions and campaign tagging agree with each other, that offline and CRM outcomes are imported where sales close away from the website, and that platform revenue can be reconciled with the finance figure. Until those foundations hold, any strategy is built on numbers nobody fully trusts, which is why this is usually where we start.

Most of the brands we work with run a hybrid. We help decide who owns which decisions: strategy, product knowledge, content approval, development priority and data accounts usually stay in-house, while specialist depth, Digital PR capacity and fast-moving disciplines such as GEO often sit with an agency. Then we set one backlog and one set of numbers so nothing is done twice. Our view on the in-house and agency hybrid that works explains the split.

Boards ask what marketing returned, what it cost and where the next pound should go. We report on one page against the targets agreed at the start, such as revenue, new customers, blended cost of acquisition and share of search and AI visibility, with what was tested, what was learned and what budget moved as a result. Brand traffic is separated from non-brand so it cannot flatter the picture. Our guide to SEO reporting for the board covers the KPIs a marketing director should own.

Where several agencies run different channels, someone has to look across all of them. We set shared targets and definitions, review each supplier's reporting against your own data and flag where two channels are claiming the same customer. Because our own PPC management is a fixed fee rather than a percentage of ad spend, we have no reason to push more money into paid search than the evidence supports. The principle is to measure your agency on revenue, not rankings.