Calderstone Surveyors LLP

• The most urgent problem is not a marketing one. Fee-earners are spending significant time drafting reports by hand, hunting across old files for standard wording and precedent, and re-keying data between systems. That is billable time consumed by administration, and the cost compounds every week it continues. Until that drafting and knowledge-retrieval problem is resolved, any increase in instruction volume simply increases pressure on a process that is already slow.

• Utilisation and work-in-progress are only visible monthly, from spreadsheets. That gap between a job going over-run and someone noticing it is where margin quietly disappears. Tightening job and matter management, and moving time recording from something patchy to something consistent, is a prerequisite for understanding what the business actually earns per instruction, which in turn shapes every sensible marketing decision about which services and client types to target.

• You have identified knowledge capture and junior onboarding as a linked problem. If the firm's institutional knowledge lives in senior partners' heads and in unstructured old files, onboarding is slow and quality is inconsistent. That is also a business continuity problem, which connects directly to the key-person dependency risk: the most significant of which is that senior partners personally hold most client relationships, and referral flow could be materially disrupted if one of them left.

• Your three priority areas, as you named them, are content creation, cost and margin management, and joining up your systems and data. Those three are not independent. The reason content is hard to produce consistently is partly because no one owns the function and partly because knowledge is unstructured. The reason margin is hard to manage is partly because time data is incomplete. And the reason systems feel disconnected is because there is no single view of job status, time, and billing in one place. A roadmap that treats these as separate projects will be slower and more expensive than one that treats them as a connected set.

• Several of your stated goals, specifically AI-assisted report drafting, a searchable precedent library, and automated client status responses, are AI adoption questions rather than digital marketing ones. A few fee-earners are already using ChatGPT informally for drafting, which is a signal that the need is real. But building firm-wide AI workflows without a governed approach to what tools are used, on what data, and with what sign-off, carries professional-indemnity risk that the informal current arrangement does not address. That governance question is precisely what Anicca's separate AI Adoption Roadmap is designed to work through, and it should run in parallel with this roadmap, not after it.

• The one area where this roadmap and the operational priorities connect directly is reporting. You currently have no single view of instruction volume, billable time, job profitability, or marketing performance. Armadello would bring your channel activity, enquiry data, and commercial performance into one place, so that when marketing work starts producing results, you can see which sources and which service lines are actually delivering margin, not just volume.

• Your existing Google Search spend is small and currently running without the commercial context to judge whether it is working. You do not know, from the data you have now, whether the enquiries it generates are the right type of instruction at a profitable margin. That is not a reason to stop it, but it is a reason to hold off expanding it until you have cleaner data to optimise against.

The practical starting point is not a channel decision. It is getting the operational and reporting foundations into a state where growth work is measurable.

That means resolving the job management and time-recording gaps first, establishing what a profitable instruction actually looks like, and then building a marketing approach around attracting more of those instructions from the right client types.

At £110,000 revenue per fee-earner, closing even half that gap across 28 fee-earners is a material number. The process problems below are where that time is going.

Where time is being lost

• Report drafting and reviewing precedent is your most cited time-consumer. Fee-earners are searching SharePoint and OneDrive for old files and clauses rather than working from a maintained, version-controlled library. The fix is not a new system, it is a disciplined content architecture within what you already have, so that a standard building-survey structure, a party-wall award template or a standard defect clause is findable in under a minute rather than in ten.

• Job set-up, conflict checks and matter administration are being done by hand each time. Without a structured workflow sitting in a practice management or CRM tool, every new instruction creates a small queue of admin that pulls either fee-earner or support staff time before the billable work can begin. Your CRM currently holds contacts only, which means none of this is automated or consistent.

• Over-runs on fixed-fee jobs are found after the fact, not while the job is live. You review job margin monthly from time records against the fixed fee in spreadsheets. By the time the over-run is visible, the work is done and the fee is agreed. Moving to a simple in-job check, ideally a live view of time recorded versus the fee budget, at the halfway point of any significant instruction, would let fee-earners and the supervising partner intervene while there is still room to manage scope or flag a variation.

• Pipeline and utilisation reporting is compiled by hand. This consumes time from whoever does it, usually a partner or office manager, and produces a report that is already partly out of date by the time it is read. Armadello would bring your time-recording data, your WIP position and your pipeline into one view without that manual assembly step, and would give the partners a reliable utilisation figure rather than an estimate.

• Time recording is patchy, and that undermines everything else. Utilisation targets, job margin, billing and pipeline visibility all depend on fee-earners recording time consistently. If the time-recording habit is uneven, no reporting tool fixes it. The practical step here is agreeing a minimum standard, for example same-day entry of all chargeable time, and making compliance visible at partner level weekly rather than reviewing it after the monthly billing run.

Structure and the referral pipeline

• Your growth plan is to deepen existing referral relationships with solicitors and lenders rather than expand geography. That is a sound choice given the talent constraints you described. But referral relationships are currently held informally, by individual fee-earners and partners. If a key fee-earner leaves, that relationship may leave with them. Moving referral contacts into a properly structured CRM, with a record of which instructions came from which source and how recently, protects the business and makes the relationship visible to more than one person.

• Recruiting qualified building surveyors is genuinely difficult, and you said so plainly. Process discipline reduces the cost of that problem: if your standard templates, precedent library and job-set-up workflow are well maintained, a new hire reaches productive billing faster and the senior fee-earners spend fewer hours supervising the basics. Onboarding a new surveyor into a well-organised SharePoint environment is meaningfully quicker than onboarding them into an informal one.

• A small number of fee-earners are using ChatGPT informally for drafting. There is no firm-wide policy governing that. This report is not the place to design one, but the gap is real and carries both professional indemnity and data protection implications in a regulated practice. Closing it is what Anicca's separate AI Adoption Roadmap covers, and it is worth picking up alongside this work rather than after it.

Who you serve and how you win them

• Your revenue split, roughly half from professional referrers such as solicitors and lenders, three tenths from developers and investors, and a fifth from private clients, tells you where to focus retention effort first. Professional referrers are your engine and they operate on relationship, responsiveness and reliability, none of which currently has a formal programme behind it.

• Panel memberships are a structural vulnerability. You described the risk accurately: it is not month-to-month attrition but a re-tender event that could remove a meaningful slice of revenue at once. Knowing which panels are due for review, when, and what each contributes to turnover is basic commercial intelligence you should hold in your CRM rather than carry in someone's head.

• Your cost of winning clients is essentially relational rather than media-based, which is genuinely efficient, but it also means there is no clear line of sight into which referrer relationships or panel memberships are generating the most instructable work. Without that, it is hard to know where to invest relationship time and where you may be under-serving a productive contact.

Retention and referral

• Nothing in your current approach is formalised. There are no review requests, no tracked referral programme, and no structured way of staying in front of solicitors or lenders between instructions. Strong relationships can absorb that for a long time, and clearly have, but formalising even a light-touch contact programme would make those relationships less dependent on individual fee-earners and more resilient as a firm asset.

• A simple, structured outreach rhythm for your top referrers, quarterly updates on relevant market conditions or recent case types you have handled, a check-in call before a panel renewal period, would cost very little and would do more to protect existing revenue than any new marketing activity. This does not require sophisticated technology; it requires a CRM that is actually used for relationship management rather than just contact storage.

• You have strong grounds for positive reviews from repeat-instructing solicitors and lenders but you are not asking for them. A short, consistent process for requesting a review or a written testimonial at the close of a successful instruction would build a public record that private clients and new referrers can find. At the moment that evidence exists only in the network, not anywhere a new contact can see it.

Cross-sell and upsell

• The cross-sell opportunity you identified is real and currently being left to chance. A developer using your building surveying service is an obvious candidate for valuation or project management work, but unless someone thinks to mention it at the right moment, the opportunity passes. A basic CRM workflow that flags which services a client has used and prompts a conversation about adjacent services would capture a portion of that work that currently goes elsewhere or simply does not happen.

• Your CRM is currently holding contacts and little else. That is the practical barrier to almost everything above: without service history, instruction volume, referrer type, and a note of when someone last heard from you, you cannot run a retention programme, track panel renewal risk, or identify cross-sell opportunities systematically. Upgrading how you use the CRM you already have, not necessarily replacing it, is the foundation for all of this.

• You are considering measuring client satisfaction more formally, which is worth doing. A short, consistent satisfaction check sent after each instruction, even a simple scored question and a free-text box, gives you a feedback signal you can act on and, where scores are high, a natural prompt to ask for a review or referral. Armadello would connect that satisfaction data alongside instruction volume and revenue by client type, giving you one clear view of where your strongest relationships sit and where attention is needed, rather than piecing it together from separate sources.

Fee-earners and admin staff are fielding a constant stream of

“where is my report”

calls and emails that job data sitting in your systems could answer automatically. That is unbillable time being spent on information retrieval, and it is the first thing worth fixing.

Customer communications

• The status-chasing loop is the core problem. Clients contact you because they have no other way to find out where their instruction is. A client portal, even a basic one that shows job stage and expected completion, would eliminate most of those inbound contacts without changing how your fee-earners work. This is a systems fix more than a communications one, but it lands as a customer experience improvement the moment it goes live.

• You have described your email marketing as barely active, rare broadcasts, and no real programme. For a professional services firm where the client relationship is the asset, that is a significant missed opportunity. A light-touch email sequence tied to instruction milestones, and a separate programme for past clients at predictable intervals, would keep Calderstone in mind without requiring much resource to maintain once it is built.

• An AI chatbot is on your radar, and it is worth being clear about what that tool can and cannot do in your context. For handling first-contact enquiries and signposting service information out of hours, a well-configured chatbot is realistic and useful. For status queries, it is only useful if it can read live job data, which requires an integration with your practice management system. If that integration is not in place, the chatbot will not reduce the status-chasing calls; it will just create a different channel for them.

• Your existing Google Search campaign is a live channel and presumably driving some enquiries. The question is whether the communication that follows an enquiry, the speed of response, the tone, what is promised and by whom, is consistent across the firm. If that hand-off is not defined, the paid-search investment is working to bring enquiries in while the follow-up experience is left to whoever happens to pick it up.

• You have no client portal and your CRM holds contacts only. That combination means there is no shared record of what has been communicated to a client, what was promised or when. When a client speaks to a different person at the firm, that person is starting cold. A CRM that records communication history, not just contact details, is the minimum needed before any structured email programme or portal work makes sense.

Internal and team communications

• Partners hold sign-off on any external spend, and there is no delegated authority for marketing decisions. That is appropriate given the budget size, but it creates a practical bottleneck if you want to run even small tests quickly. A simple written threshold, even an informal one agreed between partners, defining what can proceed without a full sign-off meeting, would remove friction without loosening financial control.

• Your preferred working pattern with an external marketing partner is a named contact and infrequent check-ins. That is a sensible preference given how little time partners have for marketing conversations, and it is a model Anicca works to. The condition that makes it work is that the reporting between check-ins is clear enough that you are not waiting for a meeting to understand what is happening. Armadello would bring channel activity, enquiry volumes and spend into one view that any partner can read without needing to request a platform export or wait for an agency update.

• A small number of fee-earners are using ChatGPT informally for drafting work. That is a communication-adjacent point worth naming: informal AI use without a firm-wide approach means there is no consistency in how it is used, what it is used for, or what the boundaries are. This report is not the place to address that, and it would not do it justice if it tried. That gap is precisely what Anicca's separate AI Adoption Roadmap is designed to work through with you.

What you have and what is missing

• You have no existing content audit, which means there is no baseline view of what has been published, whether it still reflects your current services, or whether any of it ranks in search. Before you write anything new, a quick audit of existing pages and articles would tell you which pieces are worth updating and which can simply be retired.

• Your brand foundations are thin for marketing purposes. A logo and letterhead are enough for professional correspondence, but they are not enough to brief a writer, a designer, or a fee-earner contributing a post. Without a written style guide covering tone, vocabulary, what you claim and how you phrase it, every piece of content that goes out carries a slightly different version of who Calderstone is.

• The occasional ChatGPT use by individual fee-earners is worth acknowledging directly. It is not inherently a problem, and the partner sign-off requirement shows some instinct for quality control. But informal use without agreed guidelines means the output is inconsistent and the process is unrepeatable. The EU AI Act's transparency duty under Article 50 will require AI-generated content to be clearly disclosed from August 2026; if any of your content reaches EU-based clients or counterparties, that applies to you directly. If your work is wholly UK-based, the direction of travel is the same and the expectation will reach UK practice. Starting a simple internal log now, noting which drafts were AI-assisted and which were written from scratch, makes disclosure straightforward when it is required rather than a retrospective problem. The broader question of how to govern AI tool use across the firm sits within Anicca's separate AI Adoption Roadmap rather than this report, but the content-disclosure point belongs here because it affects what you publish.

• "Somewhat comfortable" providing spokespeople and case-study content is enough to build on. Case studies are one of the highest-credibility formats available to a professional services firm, and you already produce them occasionally. The obstacle is not willingness; it is the absence of a process that makes it easy for fee-earners to contribute without it feeling like extra work.

What a practical content programme looks like for you

• The first step is agreeing the five or six topics you genuinely want to be known for, drawn from the types of instructions you most want to win. These become the spine of a content calendar that assigns one or two pieces per month across those topics, with a clear brief for whoever is drafting and a clear sign-off step before anything goes out.

• Case studies should sit at the centre of that calendar, not at the edge. A completed instruction with a good outcome is a ready-made piece of evidence; the process just needs to be triggered consistently, with a short template that a fee-earner can fill in while the job is still fresh. Two or three published case studies per quarter, written in plain language and structured around the client's problem and your resolution, will do more for new-business credibility than almost any other format available to you.

• Technical notes and commentary on market conditions are the second format worth prioritising. Written for a non-specialist reader rather than a peer, these demonstrate expertise without being inaccessible, and they give Calderstone's Google Search activity something useful to point to beyond your service pages. Anicca's content work can take briefs from your fee-earners and turn them into finished pieces at the frequency your channels need, which removes the bottleneck of finding internal writing time.

• A short written style guide, covering tone of voice, how you refer to your firm, what you claim and what you avoid, is the infrastructure that makes everything else consistent. It does not need to be long; two or three pages agreed by the partners and made available to anyone writing on the firm's behalf is sufficient to stop the current drift between whatever the individual writer happens to produce that week.

• Content compounds over time in a way that paid activity does not. A well-written case study or technical article published this quarter will continue to bring in search traffic next year and the year after. The investment is in building the programme and keeping it running at a steady pace, not in producing any single piece.

Where new business actually comes from

• The referral and repeat-instruction base is the genuine engine of the firm. Roughly seventy per cent of new instructions originate there, and that is entirely normal for a practice at your stage. The real issue is not the proportion but the concentration: if a handful of referrer relationships account for most of that seventy per cent, the loss of one or two individuals to retirement, career moves or a competitor relationship puts a meaningful slice of revenue at risk. That concentration is the most important channel risk you currently carry.

• The panel channel, at around twenty per cent of new business, sits at the opposite end of the control spectrum. You get the work, but the panel operator sets the terms, the fee and the relationship. It functions as reliable volume but not as a channel you can actively grow or protect through your own marketing decisions. Knowing exactly which panels contribute what, and at what margin, matters more than the headline share.

• Inbound web enquiries account for roughly ten per cent of new instructions. The GA4 data for the last twelve months shows 22,000 sessions and 165 enquiries, which gives a site-wide conversion rate of 0.75 per cent. That is not a crisis figure for a professional-services firm where many visitors are simply checking credentials, but it is worth knowing that roughly 99 in every 100 visitors leave without making contact. Whether that matters depends on how many of those visitors were genuinely looking for a surveyor, which the current setup does not tell you.

Paid search

• You are already running Google Search activity, spending around £450 a month on local terms. The GA4 data attributes six per cent of sessions to paid search and roughly eight enquiries a month to that channel. At £450 a month against eight enquiries, that is a cost per enquiry of just under £57. Whether that is a good result depends entirely on what a new surveying instruction is worth to the firm, and right now there is no formal target against which to judge it. The spend may be working well, or it may be underperforming relative to what the same money could do if the campaign were more tightly structured. Without a target and without proper measurement connecting an enquiry to an instruction, you cannot tell.

• The campaign has never been held to any performance target, which is the core problem. The budget is not large enough to waste, but it is also not large enough to act as a meaningful growth channel if it is running without direction. The practical first step is to set a cost-per-enquiry ceiling that reflects what a typical new instruction is worth, then assess whether the current campaign meets it. If it does, there is a case for testing a modestly increased budget. If it does not, the structure of the campaign needs reviewing before any more is committed.

Organic search and AI search

• Organic search delivers forty-six per cent of sessions, the largest single channel in your traffic, and it costs nothing to maintain at its current level. That makes it the most efficient channel you run, and it is almost certainly under-invested relative to its contribution. Direct traffic at twenty-eight per cent will include a proportion of people who found you through organic search, typed your address directly on a return visit, or clicked a saved link, so the true organic contribution is probably larger than GA4 alone shows.

• You flagged a specific concern about how your firm appears in AI search results, including tools like ChatGPT, Perplexity and Google's AI Overviews, particularly for "surveyor near me" type queries. This is worth taking seriously rather than treating as a future problem. Buyers and solicitors are already using these tools to find and shortlist surveyors, and the signals those tools draw on are substantially the same as traditional organic search signals: clear, specific, well-structured content about what you do, where you do it and who you are. A firm whose website contains thin or generic content is less likely to appear reliably in either. An audit of your current organic visibility and content would tell you where you stand and what, if anything, needs to change. That falls within the SEO and content work Anicca delivers, and it would also directly address the AI search uncertainty you raised.

LinkedIn

• LinkedIn is the one social channel you actively use. Without knowing the posting frequency or what the posts are aimed at achieving, it is difficult to comment on performance. What is worth noting is that for a professional-services firm with a referral-led model, LinkedIn's most valuable function is not audience growth but relationship maintenance: keeping your name visible to the solicitors, estate agents, mortgage brokers and existing clients who might refer the next instruction. If your LinkedIn activity is not oriented around that audience and that purpose, it is probably producing less than it could.

• You run no paid social advertising, and given the current channel mix and the absence of any measurement infrastructure, that is the right position for now. Paid social is not the next step here.

Measurement

• There are no performance targets for any channel. The Google Search activity is running without a cost-per-enquiry ceiling, the organic channel has no visibility goals attached to it, and LinkedIn has no defined purpose against which to judge it. That means there is no reliable way to decide where the next pound of marketing effort should go, and no way to know if something is quietly failing until the damage is already done.

• Armadello, Anicca's reporting product, would bring your channel data, enquiry volume, spend and cost per enquiry into one view. Rather than reconciling GA4 against the Google Ads dashboard and drawing conclusions from two separate exports, you would have one place where channel performance is compared on the same basis. For a firm where a partner is handling marketing decisions alongside fee-earning work, that kind of consolidated view is worth more than adding another channel.

• The measurement fix comes before anything else. Once you can see what each channel is actually producing, you can make a properly informed decision about whether the paid search budget is correctly sized, whether organic deserves more attention, and whether the referral channel needs any formal support to reduce its concentration risk. At the moment, those decisions are being made without the information to make them well.

• Your practice management system is the operational core of the business, recording instructions, tracking time and holding job information, but it appears to sit entirely separate from Xero, which handles billing and finance. That means job progress and invoicing are two different conversations, and reconciling them almost certainly involves manual steps, re-keying or a spreadsheet acting as the bridge. Connecting these two, so that completed or billable time in the practice system flows through to Xero without being typed in again, is the single highest-value integration available to you right now.

• You have no project management tool. Email and spreadsheets are doing that work, which is manageable when a team is small but creates version-control problems, missed tasks and no audit trail as volume grows. A lightweight job or task tool that connects to your practice management system would remove a significant amount of co-ordination overhead without requiring a large implementation project.

• SharePoint holds your files, and your precedent library lives inside years of stored Word reports, but nothing in that library is searchable in any structured way. That means every fee-earner who needs a comparable report or a clause they have used before has to rely on memory or dig through folders manually. This is a meaningful drag on fee-earner time and a risk to consistency across reports. Making that precedent searchable, even through a simple metadata structure or a well-governed folder taxonomy, would recover hours across the firm each week.

• Your overall approach to the systems is through their interfaces rather than any automated connections between them. That is not unusual at your stage, but it does mean that as the firm grows, every new instruction creates manual work in at least two or three places. Mapping which systems can talk to each other directly, starting with the practice management system and Xero, would give you a clear picture of where integration effort would actually pay back.

• You have no formal process for tracking which software the team uses, and no regular review of the tools you pay for. That matters because firms at this stage often carry licences for tools that are underused, and at the same time are running workarounds in email and spreadsheets because a useful integration has never been configured. A straightforward annual review of what you subscribe to and what each tool is actually used for would surface both the waste and the gaps.

• The informal ChatGPT use among fee-earners is a signal worth noting here because it relates directly to your document and precedent problem. Fee-earners reaching for an AI drafting tool to speed up report writing are doing so because the precedent library is not easy enough to use. Getting the precedent library into better shape is a precondition for using any AI-assisted drafting responsibly at a firm level. The governance question around AI tools is one Anicca's separate AI Adoption Roadmap addresses directly; this plan does not try to cover that ground.

• Armadello would give you a single reporting view that pulls channel performance, website traffic and commercial outcomes together in one place, rather than requiring someone to export from the practice system, Xero and any marketing platforms separately and reconcile them by hand. That kind of consolidated view becomes significantly more useful once the underlying system connections described above are in place, so the sequencing matters: connect the core systems first, then bring reporting across them into one place.

There is no live view, no single day-to-day owner, and no agreed standard for what counts as accurate. That is a material risk in a firm where fee-earner utilisation and work-in-progress levels can shift significantly within a single week.

What you are tracking

• Your five weekly KPIs, covering utilisation, lock-up, write-offs, jobs in progress and new instructions, are the right ones for a professional services firm of this type. The problem is not the choice of metrics; it is the frequency and format. Monthly is too slow for utilisation and WIP, where the cost of a problem compounds the longer it goes unnoticed.

• No one owns data quality explicitly. When the same figure can be prepared differently by two people using different spreadsheet versions, you get quiet disagreements about which number is right. That erodes confidence in the reports and leads to decisions being made on instinct rather than evidence.

• There is no live dashboard, which means the board pack or monthly review document takes real preparation time every cycle. That time belongs to the bookkeeper, whose role is not analytical, and to the senior partner, who has to reconstruct context each month rather than spot a trend as it develops.

What the marketing data shows

• Over the last 12 months, your website generated 22,000 sessions and 165 enquiries, giving a site-wide enquiry rate of 0.75 per cent. That figure is not tracked internally in any meaningful way at present; it simply happens in GA4 and no one is watching it regularly.

• Your Google Search campaign is spending around £450 per month and generating approximately eight enquiries per month, which puts the cost per enquiry at just over £56. That is a reasonable figure for a professional services firm, but without any reporting connection between the paid channel and what happens to those enquiries afterwards (whether they convert to instructions, what type of work they bring in, what the average fee is), you have no way of knowing whether £450 is too much, too little or well-directed.

• Organic search accounts for 46 per cent of sessions and is your single largest traffic source. Direct traffic at 28 per cent reflects your referral-heavy new business pattern. These channel splits are useful context for prioritising where to invest, but they are sitting in GA4 unread rather than being part of any regular commercial conversation.

The practical fix

• The immediate step is connecting your marketing performance data to your commercial KPIs in one place, rather than treating them as separate concerns. When you can see enquiry volume alongside new instructions and utilisation in the same view, you can ask whether a dip in new instructions was preceded by a drop in enquiries six weeks earlier, and act accordingly.

• Armadello, Anicca's reporting product, is designed to do exactly this: pull website traffic, enquiry volumes, paid search spend and channel performance into a single live view so the numbers are available without anyone spending time compiling them. For Calderstone, the practical value would be shifting the senior partner's monthly review from reconstructing what happened to deciding what to do about it.

• Separately, someone inside the firm needs to own the data. Not as a full-time role, but as a named responsibility: one person who checks that the inputs are consistent, that GA4 is tracking enquiries correctly, and that the figures in the report are the same figures everyone trusts. Without that, even a better reporting tool produces outputs that people quietly discount.

• If fee-earners are informally using AI tools such as ChatGPT for drafting, that introduces a separate question about what data leaves the business and under what terms. That is not a reporting question; it is a governance one, and it is the territory of Anicca's AI Adoption Roadmap rather than this report. It is worth naming as a gap that sits alongside your data quality problem, even if the two are addressed differently.

What you are spending and what you are getting

• Your total marketing budget sits under £2,000 a month, the majority of which covers website hosting and occasional directory listings. That is a very thin investment relative to revenue, and it means there is almost no campaign activity to track, optimise or reallocate.

• You are running Google Search advertising at around £450 a month and generating roughly eight enquiries a month from it. That puts your cost per enquiry from paid search at approximately £56. Whether that is good value depends entirely on which service lines those enquiries are for, what proportion convert to instructions, and what fee income they represent. Right now, none of that is joined up.

• Your Armadello data shows that paid search accounts for only 6% of your sessions over the last twelve months, while organic search drives 46%. The paid channel is the smallest contributor to traffic by a clear margin, yet it is consuming the majority of your active marketing budget. That may or may not be the right balance, but you cannot answer that question confidently until you know what organic search is genuinely producing in enquiries, and what those enquiries are worth.

• You have told us you are not yet tracking return on marketing spend formally. That means you have no reliable basis for knowing whether the £450 a month on Google Search is producing more value than the same money spent elsewhere, whether directory listings are generating any work at all, or which of your four service lines (surveys, valuations, party wall, and project management or other work) is cheapest to win and most profitable to deliver.

The SaaS and software picture

• You have somewhere between ten and thirty active SaaS subscriptions. At that range, the honest risk is that a proportion of those tools are either duplicating each other, used by one person, or not actively used at all. Fee-earner salaries and professional indemnity insurance are correctly identified as your biggest costs, and premises and software sit third. It is worth a short audit of those subscriptions, not to cut indiscriminately, but to establish what each one actually does in the business so you are not paying for overlap.

• Without a map of which tools connect to which, it is also impossible to know whether your practice management, your accounting software, and your CRM (currently contacts-only, as you described it) could share data rather than requiring manual re-entry between them. That kind of double-handling is a quiet drag on fee-earner time, which is your single largest cost line.

What good would look like

• The practical shift is connecting cost to outcome at the channel level. You need to be able to see, in one place: what you spent on each channel in a given month, how many enquiries came from it, what proportion converted to instructions, and what revenue those instructions represent. At your revenue scale, even a rough version of this is enough to make better decisions.

• Armadello would bring your Google Ads spend, your enquiry volume by channel, and your website performance into one reporting view rather than requiring you to pull platform exports separately and try to reconcile them. That is the starting point for moving from instinct to evidence on budget questions.

• Once you can see cost per enquiry by channel alongside a conversion rate and an average fee, you have the basis for a proper conversation about whether your current £450 a month on Google Search should grow, shrink, or stay where it is, and whether there is a case for putting any budget behind the organic and referral channels that are currently doing most of the work for nothing.

Regulatory standing and professional compliance

• Your RICS registration and professional indemnity insurance are current, which is the right foundation. The incoming changes to RICS home-survey standards are a real operational concern, not just an administrative one, because any adjustment to what a report must contain or how a valuation is conducted flows directly into your fee-earner workflows, your template library, and the wording of client-facing documents. Tracking that change through your existing risk register rather than separately is the practical move.

• Professional negligence is your top-named risk, and rightly so: a single disputed report in a high-value transaction can consume time, legal cost and management attention that far outweighs the original fee. The connection to marketing here is direct. If your website or proposal documents describe your service in terms that do not match what the engagement letter or survey report actually commits to, you create unnecessary exposure. Consistency between what is marketed and what is contractually delivered is worth checking now, not after a complaint.

• Your lender panel DPAs are in place, which matters. The question is whether the same discipline applies to the marketing-side tools: your website analytics platform, any form-collection tools, email or CRM tools, and any third-party pixels or tracking scripts. If those have not been reviewed as a group alongside your case-management DPA, the gap is real even if it is invisible day to day.

Marketing-data compliance: consent, cookies and tracking

• The practical first step is a light-touch audit of every tool that touches visitor or client data on your website and in your marketing workflow. That means your cookie banner, the consent it actually captures, the analytics platform(s) running in the background, and any form data that flows into your basic contacts CRM. Most professional-services firms find at least one tool collecting data with no DPA in place and a cookie banner that does not reflect what is actually running on the page. Finding that yourself is far less damaging than a client or regulator finding it for you.

• Armadello, Anicca's reporting product, would consolidate your channel and performance data, which is relevant here because one side effect of pulling everything into one governed view is that you can see exactly what data is flowing, from where, and under what consent basis, rather than relying on each platform's own reporting in isolation. That visibility makes a compliance conversation with a regulator or a concerned client much easier to have.

• The EU AI Act's transparency duty under Article 50 requires clear disclosure when a user is interacting with AI or being shown AI-generated content, applying directly to EU-facing businesses from August 2026. You have told us you operate in the UK only and that AI regulation is not a current concern, and that is a defensible position for now. But Article 50 is a clear signal of where UK practice is heading, and folding a simple disclosure review into your consent audit now, rather than retrofitting it later, costs almost nothing.

• You are planning Cyber Essentials Plus certification. That process will surface some of the same questions about data handling and access controls that a marketing-tool audit would raise. Running both exercises close together, rather than sequentially with a gap between them, avoids duplication and means findings from one inform the other.

The AI tool gap

A small number of your fee-earners are using ChatGPT informally for drafting, with no firm-wide policy in place. Your AI usage policy is in draft.

Client-data confidentiality with AI tools is your third-named risk, and it is the right one to be watching: the concern is not AI in the abstract but what happens when a fee-earner pastes client or property details into a consumer AI tool with no data-processing agreement behind it. That is a GDPR and professional confidentiality problem, not just an IT question.

This report is not the place to close that gap; it is what Anicca's separate AI Adoption Roadmap is built to address, covering policy, sanctioned tooling and governance. The honest move is to name it here and route it there, because the risk is live now and the draft policy sitting in a folder is not the same as a policy in use.