Every HoursBack report includes a five-day implementation plan. Not "here are the tools we recommend" - a specific day-by-day structure that tells you exactly what to do and in what order.
After delivering enough of these, we have learned something: the order matters as much as the tasks. Most people who try to implement AI recommendations on their own do them in the wrong sequence. They set up the complicated tool first because it is the most exciting, skip the simple workflow audit that would have saved them two hours of configuration, and then wonder why things are not working.
Below is the generic version of the plan. Every client's version is personalised to their specific workflows and the tools we have recommended. But the underlying structure is the same every time - and if you use it, you will avoid the mistakes we see most often.
Day 1: Map before you build
Task: Write down, in plain English, the three to five repetitive tasks that eat the most time in your week.
Not a list of every task. Not a brainstormed inventory. The three to five that cost you the most hours on a weekly basis, described precisely enough that someone else could do them from your description.
Example of too vague: "answering emails." Example of specific enough: "I spend around 45 minutes each morning replying to new enquiries from our website contact form - usually 6-10 messages, each one asking roughly the same three questions about pricing, availability, and timescales."
This is not busywork. It is the foundation everything else sits on. If you cannot describe a task precisely, you cannot automate it reliably. Spending 30 minutes getting this description right on day one saves hours of wasted configuration later.
The trap on day one: Jumping straight to tools. Day one has no tools in it. That is deliberate.
Day 2: Quick wins only
Task: Implement one quick win from your plan - something that takes under 60 minutes and produces a visible result today.
Quick wins in a HoursBack report are specifically chosen to be high-value and low-effort. Common examples include setting up a meeting scheduling link to eliminate the email back-and-forth around booking calls, creating a set of saved reply templates for your most common email responses, or building a simple ChatGPT or Claude prompt that drafts a specific document type you produce regularly.
The reason day two is quick wins only is psychological as much as practical. Implementing one thing that works - however small - changes how you approach the rest of the week. The three hours you might have spent on configuration that does not quite come together on day two goes into something you know will deliver a return. That momentum matters.
The trap on day two: Trying to do more than one thing. Implement one quick win completely and verify it is working before moving on. Two half-finished quick wins are worth less than one working one.
Day 3: Set up your primary tool
Task: Configure the main tool your plan recommends - the one that addresses your highest-volume, highest-cost workflow.
This is the day with the most setup work, and it gets a full day because full configuration rarely happens in stolen half-hours between other tasks. Book the time. Tell your team you are unavailable for the morning.
What "primary tool" usually means in practice: an AI writing assistant configured for a specific task (not general use), an automation between two tools you already use (connecting your email to your CRM, for example), or a Claude Project or Custom GPT set up for the specific workflow you identified on day one.
The key word is "configured." Setting up a free trial is not configuration. Configuration means the tool is set up with your specific instructions, your reference documents are uploaded, you have tested it with at least five real examples from your work, and the output is good enough to use without significant editing.
The trap on day three: Declaring it "done" before testing it. The tool feels impressive immediately after setup. Test it on a realistic example from your actual workflow - something representative, not the easiest possible case - before considering it complete.
Day 4: Connect and consolidate
Task: Connect your new tools to your existing workflow so they are part of how you work, not a separate thing you have to remember to use.
This is where most AI implementations stall. The tool works. It produces good output. But it lives in a different tab, requires a different login, and does not fit naturally into the sequence of steps you already follow. Within a fortnight, you have stopped using it.
Consolidation means one of three things, depending on your setup:
- Integration: Connect the new tool to something you already open every day - your email client, your project management tool, your calendar. Zapier, Make, or the tool's native integrations usually handle this without any code.
- Shortcut: If integration is not possible, create a shortcut that reduces the friction of using it - a browser bookmark, a phone widget, a keyboard shortcut that opens the tool directly.
- Trigger: Define exactly when in your existing workflow you will use the tool. "Every time a new enquiry comes in, I will open the template and draft the reply using [tool]." A specific trigger is harder to forget than a general intention.
The trap on day four: Skipping this step because the tool "works fine as it is." The friction of switching between tools is invisible until it is the reason you have stopped using something. Spend the time on day four to reduce it.
Day 5: Measure and commit
Task: Write down what changed this week and commit to the 30-day review.
Measure in the simplest possible terms: how many hours did the tasks you identified on day one take this week compared with last week? The number does not need to be precise. An honest estimate - "Monday's enquiry replies took 20 minutes instead of 45" - is sufficient.
If the number is positive (time saved), that is your baseline for the 30-day review. If the number is neutral or negative, this is the moment to identify why: is the tool configured incorrectly, is the workflow wrong, or is this simply not the right place to start? A clear answer on day five is far more useful than three months of hoping it will sort itself out.
The 30-day review is important enough to schedule right now, before you close this plan. Block 30 minutes in your calendar for four weeks from today, labelled "AI implementation review - hours saved?" That calendar entry is the accountability mechanism. Without it, the plan becomes another set of notes you intend to revisit.
The trap on day five: Moving straight to the next tool without measuring this one. The data from your first implementation shapes every decision you make about subsequent ones. It tells you which workflow types respond best, which tools are worth the setup cost, and - most importantly - what your actual hours-back number is. This is the information your business needs.
What comes after day five
The five-day plan is not the end of AI implementation - it is the beginning of a repeatable cycle. Once one workflow is running well, you apply the same structure to the next one on your list. Over four to six implementation cycles, you build up a set of working automations that collectively save 5-10 hours a week. That is the typical result our clients reach within 60-90 days of their assessment.
The Implementation Kickstart add-on - available after an assessment - is designed for clients who want to work through days 2-4 with us on screen, setting up the specific tools in their report rather than following written instructions alone. It is not a replacement for the plan; it is a supported version of it for the tools that benefit from a second pair of eyes during setup.
Ready to get a personalised version of this plan?
The five days above are the structure. What makes the plan in your report different is that every tool, every task, and every day-by-day instruction is specific to your workflows, your business, and the hours you are currently losing.
A HoursBack Assessment identifies where your time is going, recommends the specific tools to address it, and produces a ready-to-use five-day plan you can start on Monday morning.
Book your HoursBack Assessment - £799
Or start with the free 2-minute AI readiness quiz to get a quick read on where your biggest opportunities are.
Ready to reclaim 5-10 hours a week? Book your AI workflow assessment. 60-minute diagnostic, custom report in 48 hours, agent blueprints and automation recipes built around your business.
Know someone who could use this? Get a referral link and earn £50 for every friend who books an assessment.
Get AI tips for your business
Practical advice on saving time with AI. One email per week, no spam.