AI Implementation Consultant for Small Business: What They Do, What They Cost, and How Not to Get Fleeced - Lilach Bullock | AI Implementation Consultant
The short version:An AI implementation consultant for small business should spend more time on your processes and your people than on the software itself, because that's where projects fail. Expect to pay. . .
The short version:An AI implementation consultant for small business should spend more time on your processes and your people than on the software itself, because that's where projects fail. Expect to pay anywhere from £1,500 for a focused two-week sprint to £15,000+ for a full three-month rollout, and walk away from anyone who wants to start with a "strategy deck" before they've watched how your team works. The tools are the easy part. The hard part is getting a 51-year-old bookkeeper who's used the same spreadsheet since 2009 to trust a chatbot with her invoices. What this job title means (because half the people using it don't know either)Since ChatGPT went mainstream in late 2022, thousands of people have added "AI consultant" to their LinkedIn headline. Some of them have done the work. Most have done a weekend course. I wrote a longer piece onhow to become an AI consultantbecause I got tired of watching people with zero client hours charge £5,000 for a slide deck. For a small business owner, the practical definition is this: an AI implementation consultant is someone who comes into your business, works out which of your repetitive, time-consuming tasks can be handed to a tool safely, sets that tool up, trains your team to use it, and then leaves you able to run it without them. That's it. Not "AI strategy. " Not "digital transformation. " Just: fewer hours spent on the boring stuff, done in a way that doesn't break anything or embarrass you with a client. If you want the fuller picture of the actual job rather than the LinkedIn version of it, I've laid it out in detail inwhat AI consultants do. The short version for this piece: 70% of the work is process mapping and change management. 20% is tool selection and setup. 10% is the actual "AI" bit that everyone thinks is the whole job. A real example: the accountancy firm and the six-week client onboardingLast year I worked with a nine-person accountancy firm in the Midlands. Their client onboarding process took, on average, six weeks from signed engagement letter to first meeting with an assigned bookkeeper. Six weeks. In that time, roughly one in five new clients went quiet and didn't respond to follow-ups, which the partners had just accepted as normal churn. When I sat with the office manager and mapped the actual steps, it wasn't a technology problem at all. It was 14 manual steps spread across three people, none of whom knew what the other two had already done. Identity checks sat in an inbox for days. Engagement letters were drafted from scratch each time in Word, even though 90% of the wording never changed. Client details got typed into four different systems by hand. We didn't need a fancy AI platform. We used ChatGPT's custom instructions to draft the standard sections of engagement letters in seconds instead of 40 minutes, set up a simple automation between their intake form and their practice management software so client details only got typed once, and used an AI transcription tool to turn the first client call straight into onboarding notes instead of someone writing them up later from memory. Six weeks became eleven days. Not because of some clever algorithm, but because we removed five unnecessary handoffs and stopped three people from redoing each other's work. The AI tools saved maybe four hours a week combined. The process redesign saved the other five weeks. That ratio is typical, and it's the bit most articles about AI implementation skip entirely because "we mapped your workflow on a whiteboard for two days" doesn't sell as well as "AI-powered transformation. "What it costs in 2026Pricing varies a lot depending on scope, but here's what's realistic for a UK or US small business right now:A focused audit (one to two weeks):£1,500 to £3,500. You get a clear map of where AI can and can't help, plus two or three quick wins you can implement yourself. A single-process implementation (customer service, invoicing, content, or scheduling):£2,500 to £6,000, usually over three to four weeks including staff training. A full small business rollout across multiple functions:£8,000 to £20,000 over eight to twelve weeks, often with a monthly retainer afterwards for support. Ongoing coaching rather than a one-off project:£500 to £1,500 a month, which suits founders who want to build the muscle themselves rather than hand it off. That's closer to what I'd call anAI implementation coachthan a traditional consultant, and for a lot of small business owners it's the better fit because the knowledge stays in-house. On top of the consultant's fee, budget for the tools themselves. A decent ChatGPT business setup runs from £20 to £30 per user per month for Team or Enterprise tiers, and I've broken down exactly what you get for that money inhow much ChatGPT costs for business. Add automation platform fees, which typically run £30 to £150 a month depending on volume, and most small businesses end up spending £100 to £400 a month on tools once everything's live. The honest point almost nobody makesHere it is, plainly: most AI implementation projects fail not because the technology doesn't work, but because the business owner treats the consultant's departure as the finish line. It isn't. It's the start of the maintenance phase, and maintenance is where things quietly rot. I've seen a client spend £6,000 on a beautifully built customer service automation, use it brilliantly for two months, and then let it drift for a year while the FAQ answers it was trained on went out of date, a new product line launched that it knew nothing about, and nobody owned the job of updating it. By month nine, it was giving customers wrong pricing. The client didn't tell me this until I asked. They were slightly embarrassed, which I understood, but it happens constantly and consultants rarely warn you about it because it makes the tidy after-photo look less tidy. So the honest advice is this: before you hire anyone, decide who on your team will own the tool after the consultant leaves. Not "we all will. " One named person, with two hours a month blocked in their calendar to check it's still accurate. If you can't name that person now, you're not ready for implementation. You're ready for a coach who'll build that habit with you gradually instead. Work with meWant AI doing the heavy lifting in your marketing?I build the systems that handle the boring 80 percent, so you get your week back. Done properly, with the human kept in. Where AI implementation pays off for small businessNot every part of a business benefits equally, and a decent consultant will tell you where not to bother. From what I've seen across dozens of small business engagements, the clearest wins sit in a handful of areas:Customer service:first response drafts, FAQ handling, and ticket triage. I've written specifically about this inhow AI is revolutionising customer service for small business owners, and it's usually the fastest payback of any area, often visible within the first fortnight. Content and marketing drafts:first-draft blog posts, social captions, email sequences, meeting summaries. Admin and scheduling:calendar coordination, meeting notes, standard document drafting. Repetitive data entry:moving information between systems that don't talk to each other natively. Where AI implementation rarely pays off, at least not yet, for a business under 20 people: full financial forecasting, anything requiring genuine judgement about a specific client relationship, and creative work where your brand voice is the actual product. I go into more of these specifics inAI workflow automation for small business, including which workflows waste your time despite looking impressive in a sales demo. What a proper engagement looks like, step by stepWhether you hire someone or do it yourself, this is roughly the sequence that works:Step 1, Week 1:Map the process as it happens, not as the org chart says it happens. Shadow the person who does the job daily. Step 2, Week 1 to 2:Identify the two or three biggest time drains. Ignore everything else for now. Step 3, Week 2:Choose the tool. Resist the urge to buy the most expensive option. Start with what you already have, often ChatGPT or the AI features already inside your existing software. Step 4, Week 3 to 4:Build a small pilot with one person, not the whole team. Run it alongside the old method for two weeks, not instead of it. Step 5, Week 5:Compare results honestly. Time saved, errors introduced, staff confidence. If it's not clearly better, stop here. That's a legitimate outcome. Step 6, Week 6 onward:Roll out to the full team with written instructions, not just a live demo everyone forgets by Friday. Step 7, ongoing:Name an owner, schedule a monthly 30-minute review, and revisit every quarter. Questions to ask before you hire anyoneAsk these three before signing anything, and pay close attention to how they answer, not just what they say:"Can you show me a before-and-after from a business roughly my size?" Vague case studies from enterprise clients tell you nothing about a seven-person team. "What happens if the pilot doesn't work?" A good consultant has a plain answer. A nervous one dodges it. "Who owns this once you're gone?" If they don't ask this question themselves before you do, that's a warning sign. When you don't need to hire anyone at allI'll say this plainly because it costs me potential clients and I don't care: if your business has fewer than five people and your biggest pain point is drafting emails, writing social posts, or summarising meetings, you probably don't need a consultant. You need two afternoons with ChatGPT and a decent set of custom instructions. Save the consultant's fee for when you're trying to connect three systems together, retrain a team that's resistant to change, or roll something out across more than about eight people, where the coordination alone becomes a job in itself. Frequently asked questionsHow much does an AI implementation consultant cost for a small business?Expect £1,500 to £3,500 for a focused audit or single quick win, £2,500 to £6,000 for one full process implementation with staff training, and £8,000 to £20,000 for a multi-department rollout over two to three months. Ongoing coaching runs £500 to £1,500 a month instead of a lump sum. Do I need an AI implementation consultant or can I do it myself?If your team is under five people and the problem is drafting content or summarising meetings, do it yourself with ChatGPT and a couple of afternoons. Hire someone when you're connecting multiple systems, training more than about eight staff, or dealing with resistance to change that needs an outside voice to move. What's the biggest reason AI implementation projects fail in small businesses?Nobody owns the tool after the consultant leaves. The build is usually fine. Three to nine months later, the information it relies on is out of date, nobody's checking it, and it starts giving wrong answers to customers or staff quietly stop using it. How long does a typical AI implementation project take?A single process, done with a two-week pilot before full rollout, takes about five to six weeks from first meeting to the whole team using it independently. Full multi-department rollouts typically run eight to twelve weeks. I go much deeper on this inthe AI marketing guide. About the authorLilach BullockForbes Top 20 Social Media Power Influencer. 21 years of marketing experience. Currently obsessed with AI implementation done well. Lives in London, speaks worldwide, writes a weekly newsletter that 15,000+ founders read on Sunday mornings. Sundays onlyGet theSunday newsletter. One email a week. AI experiments, marketing tactics, and the workflows Lilach is building right now in her own business. Subscribe freeLet’s get your marketing running onAI. Book a free 30-minute callWe figure out what you need, where AI fits in, and what working together would look like. Book the call →Or take the 30-second calculatorYou’ll see the hours and the money quietly leaking out of your week, and the three workflows worth building first. Take the calculator →Or grab the free AI resource libraryPrompt packs, templates, checklists, and swipe files. The exact tools I build for paying clients. Yours, free. Get the library →