# How a Small Consultancy Should Measure Marketing What a ten-person consultancy should actually track, why most firms measure the wrong things, and how to establish a baseline that means something. Source: https://artifexa.co.uk/insights/measuring-consultancy-marketing/ --- Home / Insights / Visibility Entry 14 of 16· 8 min read ## How a small consultancy should measure marketing The wrong measures are the easy ones to screenshot. What a baseline actually is, the difference between a measurement and an estimate, and how often to re-check. By Matthew Hillman Published 7 September 2026 For consultancies Request a free review → A measurement needs a scale, a period and someone reading it.Photograph: [Brett Jordan](https://unsplash.com/@brett_jordan?utm_source=artifexa&utm_medium=referral) ### The wrong things are easy to screenshot Most consultancies measure their marketing badly not because they are careless but because the numbers that are easy to read are usually not the ones that tell you anything useful. Sessions go up. Impressions go up. The post got thirty-seven likes. These numbers move, they look like feedback, and a monthly report built from them creates the impression of a measurable programme. The problem is that they can all go up while enquiry volume falls, and they can all fall while enquiry volume rises. They are not wrong numbers. They are just not connected to the question a principal of a ten-person consultancy actually needs answered, which is: did this change anything that matters to the business? The measures that matter are fewer, mostly lagging, and only mean something when they are tracked the same way twice against a stated baseline. That is a harder discipline to maintain than screenshotting a dashboard, which is why most firms default to the dashboard. ### What a baseline actually is A baseline is a number, from a named source, recorded on a specific date, before you changed anything. It does not have to be a good number. It has to be real, dated and findable in six months. Without a baseline, an improvement is just a higher number. A higher number without a baseline tells you nothing about whether you caused it, whether the platform reclassified something, whether the comparison period was different, or whether a single piece of content that has since disappeared accounted for the whole movement. All four happen regularly, and a higher number without a baseline cannot distinguish between them. The practical minimum is this: when you decide to invest in marketing, or when you start working with a supplier, write down the current figures from each source you plan to track, the date you recorded them, and the exact source they came from. Not a summary. The actual numbers, with dates. That document is your baseline, and it is the only thing that will let you evaluate what happened later. Blunsdon Blinds ran 925 unique visitors and 3,101 page views between 1 September 2025 and 31 August 2026. That figure is useful not because it is large or small but because it has a period attached. A visitor count without a period is not a measurement. ### The measures that matter to a small firm A ten-person consultancy does not need a marketing dashboard. It needs four things tracked consistently. | Measure | What it is | Why it matters | How to get it | | --- | --- | --- | --- | | Qualified enquiry count | How many enquiries arrived in a given period, and how many of those were from firms you could actually work with. | This is the outcome measure. Everything else is a leading indicator or a proxy. If this goes up and nothing else explains it, something worked. | Your CRM, your inbox, or a simple tally. No tool required. Record the period. | | Enquiry source | How each enquiry arrived: referral, search, social, event, direct. | Without this, you cannot attribute an increase in enquiries to anything specific, and you cannot decide where to put the next month's effort. | Ask every new enquiry directly. No attribution platform is more reliable than the client's own account of how they found you. | | Target page rankings | Where your most important pages appear in search results for the specific queries your buyers use. | A page that does not appear for the right queries will not send you the right people, regardless of how well it converts. | Search Console, or a free rank-checking tool for the handful of queries that matter to your firm. Not a global rankings report. | | Search-to-contact rate | Of the people who reached your site via search, how many contacted you. | A high click count with a low contact rate suggests the page is answering the wrong question, or that the next step is unclear. | Search Console click counts into your contact or service pages, cross-referenced with your analytics contact-goal completions. | Tracking all four consistently, with dates, produces something you can actually reason from. Tracking ten metrics inconsistently produces a spreadsheet full of numbers and no way to know what any of them mean. ### Measurement versus estimation Not every number a tool reports is a measurement. Some are estimates, and the distinction matters when you are trying to decide whether something changed because of something you did. A measurement has a source you can name and a method that, if repeated under the same conditions, would produce the same result. Search Console click counts are measured: they are Google's own record of clicks from its search results to your site. They are not perfect, but they have a defined provenance. Organic traffic in most web analytics platforms is partly estimated. Direct traffic and dark-social traffic are regularly mis-attributed as organic or direct depending on how the session started, which browser was used and whether the referrer header was sent. If a newsletter sends a hundred people to your site with no UTM tags, most of those sessions will be recorded as direct. The analytics platform has not measured them wrongly. It has done the best it can with incomplete information. The practical implication is this: when a number goes up, ask what kind of number it is before you claim it as evidence. If it is a platform estimate, that is fine, estimates are useful. But they are not the same thing as a measurement, and a report that treats them as equivalent is a report that will eventually mislead you. Artifexa's Visibility Framework makes this distinction explicit. Every finding carries an evidence label, from a defined set: measured, observed, partly checked, not checked, not knowable, client stated or analyst note. The label is provenance, not severity. A number labelled "measured" and a number labelled "analyst note" are different kinds of evidence, and you should weigh them differently. The full method is at [the Visibility Framework page](https://artifexa.co.uk/visibility-framework/). ### How often to recheck Checking too frequently is a real failure mode. Marketing metrics that change slowly look like noise when you check them weekly, and reading noise as signal is how firms make confident decisions based on nothing. A reasonable rhythm for a small consultancy is this: a monthly note of enquiry count and source, a quarterly look at rankings and search-to-contact rate against the baseline, and a formal recheck every six months that goes back to the original baseline document and asks whether anything has moved enough to warrant a change of approach. The formal recheck is the discipline most firms skip, because it requires the baseline to still exist. If you did not write down the numbers before you started, the six-month recheck has nothing to compare against, and you are back to reading higher numbers and inferring that something worked. Next Layer Consulting had an audit score of 80 in June 2026 and 94 less than three weeks later. That figure is worth something because it has a specific start point, a specific end point, a named period and a named method. Take any one of those away and it is just a high score. ### Starting with what is already free A firm that has not measured its marketing before does not need to buy anything to start. Three sources will cover most of what matters at the beginning, and all three cost nothing. Search Console is the most useful of the three. It reports actual click and impression counts from Google, broken down by query and by page, with date ranges you control. It shows you which queries are sending people to which pages, how those pages have performed over time, and where coverage is thin. It is not a perfect window into search performance, but it is the closest thing to a direct record that a firm can access without paying for it. If you have not connected your site to Search Console, doing that before anything else is the single highest-return action available to you. Standard web analytics, whether Google Analytics 4 or an alternative, covers on-site behaviour. Set up goal tracking for your contact form submissions and any other enquiry paths so that you are measuring completions rather than visits. Visits are context. Completions are the point. The third source is not a tool. Ask every new client directly how they found you, and record the answer alongside the date and the enquiry source you logged at the time. No attribution platform is more reliable than a client's own account of how they arrived. Platforms mis-attribute. Clients usually know whether they searched for you, were referred by someone specific, or came across an article. That answer is a measurement, and it is the one most small consultancies never collect systematically. Once those three are in place and you have a dated baseline from each, you have enough to run a quarterly recheck that tells you something real. That is a more solid foundation than most ten-person consultancies have, and it costs nothing but the discipline to write the numbers down. If you want a read of where your site currently stands before you establish that baseline, Artifexa's free [Visibility Review](https://artifexa.co.uk/visibility-review/) produces a written, prioritised list of findings. Every finding carries one of the evidence labels described in section 04, so you know which numbers are measured and which are judgements. The report is yours to keep whether or not you go further. ← All insights In this entry - ¶01The wrong things are easy to screenshot - ¶02What a baseline actually is - ¶03The measures that matter to a small firm - ¶04Measurement versus estimation - ¶05How often to recheck - ¶06Starting with what is already free Next entry · 15 Outsourced marketing operations in the UK → Asked and answered · served as FAQ structured data ### The questions this entry exists to answer. What should a small consultancy actually measure in its marketing?[+][−] Four things are worth tracking consistently: how many qualified enquiries arrived and from where, how your target pages rank for the specific searches your buyers use, how many people who found you via search went on to contact you, and what clients say when you ask how they found you. Everything else is context at best. Sessions and impressions will move without anything happening in the business, so tracking them without the four things above tells you nothing you can act on. How is a measurement different from an estimate?[+][−] A measurement has a source you can name and a method you can repeat. An estimate is a tool's inference from data it does not fully control. Search Console click counts are measured. Organic traffic in most analytics platforms is partly estimated, because direct and dark-social traffic is regularly mis-attributed. The difference matters when you are deciding whether something changed because of something you did, or changed because the platform reclassified it. What is a baseline and why does it matter?[+][−] A baseline is the number you had before you changed anything, recorded at a specific date, using a specific source. Without it, an improvement is just a higher number, and a higher number without a baseline is not evidence of anything. The baseline does not have to be impressive. It has to be real, dated, and documented in a way you will still be able to find in six months. How often should a small consultancy check its marketing metrics?[+][−] A weekly check is too frequent for anything that changes slowly, which is most of what matters. Monthly is a reasonable rhythm for ranking positions and enquiry counts. The formal recheck, where you compare against the baseline and ask whether anything has moved enough to act on, belongs every quarter. More often than that and you are reading noise rather than signal. What free tools can a small consultancy use to measure marketing?[+][−] Google Search Console is the most useful single free tool for a firm that depends on search: it reports actual click and impression counts from Google, shows which queries are sending people to which pages, and lets you track how individual pages perform over time. Standard web analytics covers on-site behaviour. Beyond that, asking every new client directly how they found you is not a tool at all, but it is the most reliable data point you can collect, because no platform tracks a referral conversation. Before you hire anyone ### Find out which problem you are actually buying. The free measured Visibility Review is an honest, written look at your website clarity, search visibility, content rhythm and the buyer problems you most want to be known for. The plan is yours to keep. Agreed implementation starts from £750. Already know you want a monthly rhythm rather than a one-off review? See [Marketing Ops for NetSuite and ERP partners](https://artifexa.co.uk/netsuite-partner-marketing-ops/), or the full [engagement menu](https://artifexa.co.uk/engagements/) with published prices. Request the free review → Ask a question first Review delivered Within four working days Written, measured, and honest about no-fit. Nothing to install, nothing to sit through. --- Artifexa runs marketing operations for NetSuite and ERP partners and referral-led specialist consultancies, from Swindon, Wiltshire. Full machine-readable summary: https://artifexa.co.uk/llms.txt