The most expensive way to buy improvement work is to approve a project because someone confident described a problem you had not verified. An hour of structured looking usually tells you which of five possible problems you actually have, and that changes both the quote you accept and the questions you ask. Nothing here requires a paid tool or any code.
Do it in order, take notes as you go, and resist fixing anything until the hour is over. The point is a map, not a repair.
Minutes 0–10: shop your own store on a phone
Use an actual phone on mobile data, not your laptop on office wifi, and buy something. Start from a category, filter it, open a product, add to cart, go to checkout, and complete a real order you can refund afterwards. Note every moment you would have given up if it were not your own store: the wait for the first image, a filter that needed two taps to close, a shipping cost that appeared only at the end, a form that fought your autofill.
- Real phone, mobile data, real card
- Write down every hesitation, not just the failures
- Note where the price and delivery time first become clear
- Refund the order and check that the refund flow works too
Minutes 10–20: the numbers your customers generate
Open your Core Web Vitals report in Search Console, which uses field data from real Chrome users and not a simulation. Look at mobile, by page group. A failing product-page group is a different problem from a failing homepage, and the grouping tells you which template deserves the work. Then run one product page through PageSpeed Insights and read the field data section at the top, not the lab score below it — the score is a simulation, the field data is your customers.
Minutes 20–30: is Google seeing what you think
Still in Search Console, open Indexing → Pages. Compare the indexed count against roughly how many products and categories you have. Then read the not-indexed reasons: "crawled, currently not indexed" in volume points at thin or duplicated pages, and anything important sitting under noindex is an emergency. Finish in Performance: filter out queries containing your brand name, then compare the last 28 days with the previous 28. Non-brand clicks are the honest measure of your visibility.
Minutes 30–40: does your analytics tell the truth
Compare last month's revenue in analytics with last month's revenue in WooCommerce. A few per cent apart is normal; a large gap or a gap that changed suddenly means the numbers you have been making decisions from are wrong. Then check that a purchase event actually fires with a transaction id, and that you are not counting each order twice. If this check fails, stop the audit and fix tracking first — every remaining measurement depends on it, and GA4 ecommerce tracking covers the usual causes.
Minutes 40–50: look at your store like a stranger
Open your homepage and a product page in a private window and ask the questions a first-time visitor asks silently. Who are these people. Will the parcel arrive. What happens if it is wrong. Is anyone there if something goes badly. Check whether shipping cost, delivery time and the returns policy are findable before the cart, whether there is a real address and a way to reach a human, and whether the photography looks like one catalog or four suppliers. This is where trust is won or quietly lost.
Minutes 50–60: the operational blind spots
Three quick checks that consistently surface something. First, your plugin list: count them, and mark every one nobody can justify by name — that is your update risk and part of your monthly bill. Second, backups: find out when the last one ran and whether anyone has ever restored it. Third, monitoring: do you find out about an outage from a system, or from a customer email? The answers here rarely need a project; they usually need a decision. If the plugin count is past forty, or the last restore test was never, you have found the two highest-value items on the list before you have looked at a single page.
Reading your own notes
Now sort what you wrote into three piles. Things that are broken — a failing checkout step, tracking that lies, pages that should be indexed and are not — get fixed first, regardless of anything else, because they are losing money today. Things that are slow get measured against the field data before anyone proposes a rebuild. And things that are merely dated go last, because aesthetic dissatisfaction is the easiest problem to overspend on.
Then put a number on the largest item. If the phone test showed a checkout that loses people, the cart abandonment calculator turns that into a monthly figure. If the problem is conversion generally, the revenue calculator prices the improvement you are considering. A budget derived from your own numbers is the best defence against a confident quote.
What this hour cannot tell you
It will not identify which plugin is executing three hundred milliseconds of JavaScript on your product template, whether your database is carrying years of expired transients, or how much of your traffic a redesign would put at risk. Those need tooling and someone who reads waterfalls for a living. But you will know which of them to ask for — and, just as often, that you do not need to ask yet. If you want to see where that line falls in practice, our audit of an online coffee store is published in full: the measurements, the twenty-three findings and the reasoning behind what we did and did not recommend.
An hour of looking is worth more than a month of opinions, including ours.