Glass Floor Studio
teardown · GFS-TD-260828 · 2026-08-28
The Teardown

What’s actually wrong with your site.

Issued for Woodline Grill House
Location Milton, ON
Site reviewed woodlinegrill.ca
Audit date August 28, 2026

You are paying the delivery apps about $3,900 a month to do something your own site could do.

We crawled every page of your site and ran it through Lighthouse on a throttled mobile connection — none of which needs anything from you. Your own order reports filled in the volume picture for the last 90 days.

You take about 340 orders a month through the delivery apps at a $38 average. At 30% commission that is $3,876 a month leaving the business — roughly $46,500 a year — on orders from customers who already know your name and came looking for you.

The reason they go through an app is simple: your site gives them nowhere else to go. There is no ordering on it. The menu is a scanned PDF that opens as an unreadable image on a phone. And the hours block says you are closed when you are open.

Fixing the ordering alone pays for a full rebuild in under two months. The detail starts on the next page. If you only have five minutes, skip to §05 — that is where we tell you whether you need a new site at all.

20/100
Site health Five categories, 20 points each. The score is the eight problems in §03, bucketed and totalled — not a separate opinion. Recheck it against the list yourself. Every input is measured from your public site, so the score is always complete — it never depends on you having analytics.
Ordering & conversion2/20No direct ordering at all (01), phone not tappable (08)
Mobile experience4/20Menu is a scanned PDF (02), no responsive layout
Speed3/208.1s to first paint, 12MB hero image (04)
Trust & credibility6/20Hours wrong in three places (03), stock food photos (07)
Findability5/20One page indexed (05), no menu schema (06)
Orders via delivery apps340/mo
Average order value$38
Commission rate30%
Commission paid monthly$3.9k
Mobile load time8.1s
Pages indexed by Google1

How these were measured. From the outside, needing nothing from you: load time and page weight (Lighthouse, throttled 4G mobile), mobile rendering, SSL, structured data, and how many of your pages Google has indexed. From your own order reports, if you have them: order volume, average order value, and the commission figure calculated from them. If you don’t have those numbers to hand, those three rows read “—” and we say so — we never estimate a number and present it as measured. Everything here is reproducible, and we will show you the raw run on request. Figures in this sample report are illustrative.

Six things go wrong in the first second.

This is your homepage exactly as it is built today. There is only one layout — a phone gets this same desktop page, shrunk. We’ve numbered the six things a hungry customer meets before they give up and open an app instead.

https://woodlinegrill.ca/
905-555-0164 Order online [stock photo: generic flame-grilled burger] HOURS — CLOSED Menu (PDF, 8.4MB) Woodline Grill House — Milton No address, no map, no hours 1 2 3 4 5 6
1“Order online” opens a delivery app. Your busiest button hands the order — and 30% of it — to someone else.
2The menu is an 8.4MB scanned PDF. On a phone it opens as a pinch-to-zoom image. Most people close it.
3Hours say CLOSED. Captured 6:40pm on a Friday, while the kitchen was open and taking orders.
4Hero is a 12MB stock burger. Not your food, and it is the reason nothing appears for 8.1 seconds.
5Phone number is plain text. A customer at a red light cannot tap it to call.
6No address, no map, no hours in the footer. Nothing for Google to attach to “grill house near me”.

Eight problems, worst first.

01
CriticalOrdering

There is no way for a customer to order from you directly.

The only ordering button on the site opens a third-party app. Every order placed through it costs you 30% — about $3,876 a month at your current volume. These are not new customers the app found for you; they typed your name, landed on your site, and were handed straight to a middleman.

What we’d doDirect online ordering on your own domain, with card payment and a kitchen ticket printer. Customers who already know you stop costing you a third of the bill. Ten days.
02
CriticalMobile

Your menu is an 8.4MB scanned PDF.

On a phone it opens in a PDF viewer as a flat image — no zoom that reflows, no search, no tap-to-add. It is also invisible to Google, which cannot read a scan, so none of your dishes can ever appear in a search result.

What we’d doA real HTML menu with categories, prices, photos and an add-to-order button on every item. Readable on a phone, indexable by Google, editable by you in under a minute. Four days.
03
CriticalTrust

Your hours are wrong, in three different places.

The homepage block said CLOSED at 6:40pm on a Friday. Your Google listing said open. Your Facebook page listed a third set. A customer who trusts the site and drives elsewhere does not come back to check which one was right.

What we’d doOne set of hours, entered once, shown on the site and pushed to your Google listing. Holiday overrides you set yourself. Half a day.
04
HighSpeed

The homepage takes 8.1 seconds to show anything on a phone.

A 12-megabyte stock photograph loads before any text does. Slightly over half of mobile visitors abandon a page that takes more than three seconds. You are paying for that photo twice: once in lost customers, once in the ad clicks that land on it.

What we’d doCompress and correctly size the hero, serve modern image formats, defer everything non-critical. Target under 2 seconds on 4G. One day.
05
HighFindability

You have one page. Google has indexed one page.

There are no pages for your categories, no page for catering, no page for the neighbourhoods you deliver to. Every search that could have found you — “halal grill Milton”, “family platter takeout” — has nothing of yours to match against.

What we’d doCategory and dish pages generated from the menu, plus a catering page and a delivery-area page. They come free with a real menu, because the menu becomes the site. Included in the build.
06
MediumFindability

No restaurant or menu structured data.

Google cannot tell it is looking at a restaurant. That is why your listing shows no menu link, no price range, no hours and no order button in search results — while the delivery apps, which do publish that data, show all of it above you.

What we’d doRestaurant, Menu and OpeningHours schema generated automatically from your live menu, so it can never drift out of date. Four hours.
07
MediumTrust

The food photographs are not your food.

Five of the six images are stock. Customers compare the photo to what arrives; a generic flame-grilled burger sets an expectation your kitchen did not make and cannot control. It also reads as a business that has not been updated in years.

What we’d doPhotograph twelve real dishes on a phone under a window — we will send the shot list and do the colour work. Half a day of your time.
08
LowMobile

The phone number is not a tap-to-call link.

It is typed as plain text in the header. A customer sitting at a light has to memorise it, leave the site, and dial it manually. Most do not.

What we’d doWrap it in a tel: link and add a sticky call button on mobile. Two hours.

What we’d ship first if you only paid us to do three things.

If a full rebuild isn’t on the table this quarter, here’s the order we’d touch things — biggest payoff for least effort first. Items 1 and 2 are the ones that stop the bleeding.

FixEffortImpact
1Direct online ordering on your own domain10dHigh
2Real HTML menu, phone-readable and indexable4dHigh
3One set of hours, pushed to your Google listing4hHigh
4Compress the hero, defer non-critical assets1dHigh
5Category, catering and delivery-area pages3dHigh
6Restaurant + Menu schema generated from the menu4hMed
7Swap six stock photos for your own dishes3hMed
8Tap-to-call link + sticky mobile call button2hHigh

You don’t need a prettier site. You need The Order Kitchen.

Five of the eight problems above cannot be fixed without the site being able to take an order. The rest are workarounds. Here’s what we’d actually do.

Recommended build

The Order Kitchen — direct ordering, a real menu, and a kitchen printer, built start to finish by one person.

The Order Kitchen solves issues 1, 2, 3, 5, 6 and 8 in the base price. Issue 4 (the hero) and issue 7 (your own photos) get done as part of the build — no separate quote. You’d ship in two weeks and the $450 you paid for this teardown comes off the price. You pay half to start and the balance when the site goes live — no other charges.

The arithmetic, so you can check it: you currently hand the apps about $3,876 a month. If even half your existing customers order direct instead, that is roughly $1,940 a month back. Against a net build price of $6,450, the rebuild pays for itself in about 3.3 months — and every month after that is yours. If two-thirds switch, it pays back in 2.5.

We are not telling you to stop using the apps. They find people who have never heard of you, and that is worth paying for. We are telling you to stop paying a finder’s fee on the customers who already typed your name.

Build price$6,900
Care plan$240/mo
Teardown credit−$450
Net price$6,450
Pays for itself in3.3 months