The problem with Shopify's contact form for customer support

If you are running customer support for a Shopify store, you probably started with the built-in contact form. It is the obvious first move and it works — right up until support becomes a real part of someone's week.

Shopify gives you a contact form out of the box. Add the page, and customers can reach you without hunting for an email address. For a new store that is the right amount of tooling.

Here is what it does, precisely: it takes what they typed, emails it to the address on your account, and forgets it ever happened. That is the whole feature. Everything below follows from it.

What it is genuinely good at

Worth being fair before the criticism, because the form gets dismissed too quickly and that costs new merchants time they do not have.

It is free, it works immediately, and it requires no decision. It keeps your email address off a public page, which matters more than people expect once scrapers find it. It is one fewer app in an admin that already has too many, and one fewer subscription. For a store in its first few months, when support means three messages a week and you are answering them from your phone between other jobs, adding a helpdesk is overhead that solves a problem you do not have.

The argument here is not that the form is bad. It is that it does one half of a two-part job, and the second half becomes visible at a predictable point.

The moment it stops being enough

It is not a ticket count, which is why volume-based advice is unhelpful here. Some stores handle sixty a month on the built-in form without strain; others struggle at fifteen because the questions are harder or two people are answering.

Four signals, any one of which means you have crossed the line:

• You have caught yourself searching your own inbox to check whether you replied to someone
• Two of you have answered the same customer, or nearly did
• A customer has written "following up on my last message" and you had no record of the first
• You have started copying support messages somewhere else so you do not lose them

The fourth is the clearest. Building a manual system alongside your inbox is your own behaviour telling you the inbox is not doing the job.

Where it starts to hurt

Nothing has a state

An email is read or unread, and that is the only status it will ever have. There is no "waiting on the customer", no "resolved", no "someone else is handling this". You hold all of that in your head, or you do not hold it at all.

Two people answer the same person

If more than one of you watches that inbox, this happens within the first month. Nothing marks a message as claimed, so the only defence is talking to each other about it, every single time.

Read on a phone is gone

You glance at a message in a queue, think "I will do that properly at my desk", and it drops below the fold. The customer does not chase — they just do not come back. Nothing tells you this happened, which is what makes it the most expensive item on this list.

The reply does not come back to the form

Your answer goes out from your normal email, so their response lands in your normal inbox, in a thread that has nothing to do with the original submission. After two rounds you are scrolling to work out what was already said.

The order is not attached

Someone asks where their order is. The form gave you their name and a message. Finding the order means a search in another tab, then the tracking number in another, then back. Four to eight minutes, and most of it is not typing.

You answer the same thing forever

"Do you ship to Canada" arrives weekly and you compose it fresh every time, because there is nowhere to keep the good version.

No record survives

Six months on, a customer refers to what you told them. Whether you can find it depends on whether you can remember the right search term.

The workarounds, and where each one stops

A shared inbox with labels

Better than nothing. Labels are a status field you maintain by hand, which works until you are busy — and busy is exactly when it matters. No order context, no assignment, still no record of who did what.

Forwarding to a spreadsheet or Notion

Now you have a record and a second place to keep up to date. In practice one of the two goes stale within a fortnight, usually the one that is not where the work happens.

Gmail collision detection

Genuinely helps with two people answering at once, if everyone is in Gmail on desktop. Does not help with anything else on this list.

A full helpdesk

Solves all of it, and is priced for support teams. For a store doing fifty tickets a month, the cost and the setup are both out of proportion to the problem.

Making the built-in form better before replacing it

If you are not ready to change tools, four adjustments recover a real share of the loss and take about an hour in total.

Ask for the order number on the form

Not required — required fields on support forms stop people writing in, which is the opposite of what you want. Optional, clearly labelled, and maybe half of them will fill it in. That is half your lookups gone.

Add a subject or category field

Three or four options: order status, returns, product question, something else. It costs the customer two seconds and it lets you batch similar work, which is faster than switching between kinds of problem all day.

Set an expectation on the form itself

One line above the submit button — "we usually reply within a working day" — measurably reduces follow-up messages. A large share of second emails are not impatience, they are the customer checking the first one arrived at all.

Write your five most common answers down

Anywhere you can paste from. This is the highest-return hour in this list and it stays useful whatever tool you end up with. See what running support out of Gmail actually costs you for why repeated answers are more expensive than they feel.

What the alternative looks like

The concrete difference is worth seeing rather than describing. This is the same set of messages as a list of tickets rather than a list of emails — each one carrying a status, an order number where there is one, and how long the customer has been waiting on a reply.

Nothing here is clever. It is the same information your inbox already has, arranged so you can see at a glance which two of five need you and which three are waiting on someone else.

What to actually do

Under about ten a week: stay with the built-in form. Genuinely. The overhead of anything else exceeds what it saves you. Do write down your five most common answers somewhere you can paste from.

Somewhere in between: the specific things worth fixing, in order — make sure nothing can be silently dropped, get the order onto the same screen as the message, and keep replies threaded to the original.

Once support is a real part of someone's week: you want tickets with status, order context, and a shared view of what is handled. That is a helpdesk, and the question becomes which one is proportionate rather than whether to have one.

The honest version

Shopify's form is not bad. It is a form. It does the collecting part and none of the managing part, and it never claimed otherwise.

The tell that you have outgrown it is not ticket volume. It is the first time you cannot remember whether you replied to someone.

Related

What running support out of Gmail actually costs you puts a number on the gap described here. If someone other than you is answering, managing your first customer support hire covers what the built-in form makes impossible to oversee.