What an ecommerce audit is, and when you actually need one
By Aarón Briceño, ecommerce consultant and Shopify specialist.
Most stores that write to me do not have a traffic problem. They have a diagnosis problem: they know something is wrong because the numbers do not add up, but they do not know what, so they start trying things. More ad budget. Another app. A redesign. And because every change touches five variables at once, by the end of the quarter nobody can say what worked.
An ecommerce audit is the step those people skip: looking at the data before touching anything.
What it is, and what it is not
An audit is a diagnosis. You review the store — the business and the site — to answer one question: where is money being lost, and in what order should it be fixed?
What it is not, and this is worth saying because plenty of things ship under the name:
- It is not an automated report. Running PageSpeed and exporting the PDF is not an audit. Those tools tell you what is broken, not what matters.
- It is not a list of 200 errors. An unprioritised list hands the hard part — deciding — straight back to you.
- It is not a sales proposal in disguise. If the document ends in a quote, it was a sales proposal.
An audit that earns its fee ends in a handful of problems ranked by impact, each with the evidence behind it and a clear sense of what fixing it costs.
If the result of the audit does not change what you spend money on next month, it was not an audit. It was a report.
What gets reviewed
This is the split I use, and the order is deliberate: outside in, business before technology.
1. The data, before anything else
This is the part most people skip and the one most often broken. Before drawing a single conclusion you have to confirm that the numbers behind it are true:
- Is analytics measuring correctly? It is surprisingly common to find the same event counted twice, or purchases that never register.
- Is the funnel complete? Product view → add to cart → checkout started → purchase. If a step is missing, you cannot know where people drop.
- Does analytics agree with the store’s own admin? If they disagree, find out which one is lying before going further.
Auditing on dirty data is worse than not auditing: it gives you confidence in a false conclusion.
2. The funnel
With clean data, you look for the step that loses the most people who had already shown intent. Losing 97 % on the homepage is not the same as losing 40 % between cart and checkout: the second is far more expensive and almost always easier to fix.
| Step | The question |
|---|---|
| Entry | Is the traffic arriving traffic that can buy? |
| Product page | Is the information needed to decide actually there? |
| Cart | Does any new cost appear here? |
| Checkout | How many steps, how many fields, how many payment methods? |
| Post-purchase | Does anyone come back? |
3. Product page and checkout
These are the two pages where the sale is decided, and where I most often find money left on the table: photos that do not answer the question the buyer actually has, sizes or measurements that force a trip to another tab, shipping costs that appear as a surprise at the last step, a checkout with more fields than it needs.
4. The technical layer, last
Speed, indexing, mobile, console errors. It goes last not because it matters little, but because it is where everyone starts, which is usually why it has the least marginal return left. A fast store that explains its products badly still does not sell.
With one exception: when the technical problem affects indexing, it jumps to the top of the list. A page Google does not crawl has no funnel to optimise. It is a silent failure — the store works perfectly for anyone who lands on it — which is exactly why it can run for months without anyone noticing.
What you should be handed
Ask for these four things in writing before hiring anyone:
- The problems ranked by impact, not by ease and not in the order they were found.
- The evidence for each one. A screenshot, a query, a number with its source. Without evidence it is an opinion.
- The rough cost of fixing each one. That is what lets you decide.
- What NOT to touch. An honest audit also says where there is no problem. If everything is broken, be suspicious.
When you do NOT need one
Almost nobody writes this part, so here it is:
- If you are still selling very little. With few orders a month there is not enough volume to tell a pattern from noise. What you need is traffic and conversations with customers, not a statistical diagnosis.
- If you already know what the problem is. If you are certain checkout is a mess, fix it. Paying to be told what you already know is expensive.
- If you have just launched. Give it a few months of real data first.
The moment an audit genuinely pays for itself is when traffic is already arriving, orders are already coming in, and the numbers still do not add up — and above all, right before you increase ad spend. Scaling a store that converts badly is the fastest way to turn a small problem into an expensive one.
How I run this diagnosis, what I review and what I hand over is on ecommerce audit. If the audit already tells you the problem is conversion, the work that follows is on CRO for Shopify; and if what you need is someone to execute it too, the formats are in the plans.