Where to Start Tomorrow

If you have read this far, hopefully you have a list of what you need to do. But we've consolidated what you need.

There are about thirty items below and you are not meant to do them all this month. They are ordered so that the things which take time to collect start first, the things you can look up happen next, and the decisions come last. Work down it, and stop when you have found your weakest link.

Step one. Write the job description.

• Write one sentence saying what your website is for.
• Ask three people in your business to do the same, separately, without showing them each other's answers.
• Compare the four answers. The gap between them is the size of the problem.

Step two. Start the clock.

• Start an enquiry log today. For your next twenty enquiries, record the time it arrived and the time a person replied.
• On those same twenty, tally how many you could have sent a proposal to without going back to ask for something more.
• For one week, tally every task you repeat, with rough minutes against each.

Step three. Diagnose.

• Pull last month's counts: visits, service page views, enquiries, responses, customers. Find the biggest drop.
• Ask an AI assistant the question a customer would ask, naming your service and your city, and see whether you are mentioned.
• Check that your site states plainly what you do, who you do it for, and where.
• Read the headings of your main service page out loud, in order. Do they tell a story?
• Check your name, address, and phone number match exactly across your site, your Google listing, your socials, and any directories.
• Run your homepage and main service page through PageSpeed on the mobile setting.
• Show your homepage to three people who do not work with you, for seven seconds each, and write down their exact words.
• Read your homepage and top service page aloud. Mark every sentence a competitor could also claim.
• List every page in a spreadsheet with its purpose and its group: Value, Awareness, Credibility, or Conversion.
• Map every way a visitor can currently raise their hand to a rung on the ladder, and find the missing middle.
• Fill in your own contact form on your phone, as a stranger, and time it.
• Send yourself a test enquiry and confirm it lands somewhere a person will actually see it.
• Write out every step between a client saying yes and the work being delivered, including every email you send by hand.

Step four. Fix what costs nothing.

• Repair anything the diagnosis showed is simply broken: the form, the notification, a missing phone number.
• Rewrite the homepage headline so it says the actual thing you do.
• Put the plain sentence on the page: what, who, where.
• Resize the images, and delete scripts belonging to tools you no longer use.
• Remove one field from your contact form. Then measure. Then remove another.
• Replace the form confirmation with a specific promise about when a human will reply.
• Write an honest automatic acknowledgement, and mean it.
• Email your most recent satisfied client and ask them to approve one quotable sentence with their name on it.
• Delete the pages that fit none of the four groups.

Step five. Choose one job, not four.

Businesses that attempt everything at once end up commissioning a whole new rebuild, and that resets your evidence to zero. You will have a new website and no idea which part of it made the difference, or worse, one that still fails.

• Pick the job with the biggest drop.
• Define the outcome and the one metric you expect to move.
• Leave the other jobs alone until that metric is heading in the direction you expect.

Step six. Keep score.

• Build the five-number scorecard on one page and backfill three months so you have a baseline rather than a starting point.
• First Monday of every month, review it and write one line next to each number explaining what changed and why you think so.

Step seven. Decide what gets built.

By this point you know what your site is not doing and roughly what that is costing you. That is a brief with a number attached to it, which is a very different thing to tell an agency than "we need a new website".

• Write the brief: what is failing, what it is costing you, and what a fix has to achieve.
• Get it quoted against that outcome, not against a page count.

We want to make a closing remark about scale. This guide is written for businesses under fifty people, where you as the founder do not have a full time IT or marketing team. You might have an agency you use. Or maybe when you started out you got your nephew to build the site. That was the right call at the time, and it got you off the ground. The question is whether your website is still doing the job the business now needs it to do.

There is an assumption in this industry that everything we just described here belongs to large companies, and that smaller businesses get a simplified copy. We know that assumption is wrong, and it is the reason Qode exists and why we shared this publicly to all small medium businesses. The techniques in this guide are not expensive. If anyone says that it is, they are either overstating the effort or overbaking the requirements. Maybe they are selling at enterprise prices to enterprise buyers, or have outgrown your business compared to when you first engaged them, nothing wrong with that. But with the technology readily available today, we know that is no longer the case, and the businesses that work this out early get a real advantage over their larger competitors who are slower to move.

We truly believe that the size of an organisation should not decide the quality of what it can build.

If you want a hand with the diagnosis, we do a version of it as a conversation, no charge and no obligation. Bring whatever numbers you have.

Qode. We build websites, ecommerce, and digital products. The applied AI we build into them is what creates lasting value.