Home improvement retailers often run ads that look healthy on the dashboard: plenty of clicks, a low cost per click, a steady flow of form fills. Then the month ends, and orders have barely moved. The problem is rarely the ads alone. It is what the ad platforms are told to count, and therefore what they learn to find.
This guide explains how to set up Meta and Google ads so that they measure, and optimise towards, real orders and qualified quote requests, and how to read the results without fooling yourself.
Why clicks mislead in home improvement
Home improvement purchases have a few features that make click-based measurement especially misleading:
- Long decisions. A shopper may click an ad, order samples a week later, visit the showroom, and place the order a month after that.
- Several steps. Samples, quotes, measures and showroom visits often come before the order.
- Offline orders. Many orders are placed by phone, by email or in store, where the ad platform cannot see them.
- High-value baskets. A single order can be worth far more than dozens of small ones, so counting conversions equally hides what matters.
If the ad platform only sees clicks or page views, it will find more people who click. Those are not always the people who buy floors, kitchens or tiles.
Decide what counts as a result
Before touching campaign settings, agree with your team which actions matter, and roughly how much each one is worth to the business. A typical ladder for a home improvement retailer looks like this:
- Order (online or offline): the main result, with its real value.
- Quote or measure request: a strong signal, worth a share of an average order.
- Sample order: an earlier signal, useful in volume.
- Showroom booking or call: valuable, but harder to track.
- Engagement (product page views, visualizer use): useful for understanding, not for optimising spend.
Write this down. It becomes the basis for your conversion setup and your reporting.
Set up conversion tracking that reflects reality
Track the right events on the site
Make sure each of these actions fires a distinct conversion event, on both Meta and Google:
- purchase, with the order value
- quote or measure request
- sample order
- booking or contact form
Use the platforms' standard events and recommended setup, including server-side or enhanced options where available, as browser-only tracking misses a growing share of activity. Test every event yourself after setup by completing the action and checking that it appears.
Give each conversion a value
Ad platforms optimise better when they know what a result is worth. Pass the real order value for purchases. For quotes and samples, assign an estimated value based on your own history, for example the average order value multiplied by the share of quotes that usually become orders. These estimates do not need to be perfect, but they should reflect the relative importance of each action.
Bring offline orders back in
This is the step most retailers skip, and often the most important. When an order happens by phone or in store, upload it back to the ad platforms using their offline conversion or customer data import features, matched on the details the customer gave you with appropriate consent. Done regularly, this lets the platforms see which clicks eventually led to real orders.
Check your privacy notices and consent setup before doing this, and follow the platform policies and local data protection rules that apply to you.
Choose your optimisation goal carefully
Once orders and quotes are tracked with values, optimise campaigns towards them rather than towards clicks or landing page views. If you have too few orders for the platform to learn from, optimise towards quotes or samples with values, and keep orders as the measure you report on.
Structure campaigns around what you sell
- Search on product and intent terms ("oak engineered flooring", "zellige tiles", "bathroom showroom near me") usually captures the clearest demand. Keep brand and non-brand searches in separate campaigns so you can see each clearly.
- Shopping and product feeds work well for products with clear prices online. Keep titles, images and prices in the feed accurate and consistent with the site.
- Local campaigns and location settings matter for showrooms. Use location targeting that reflects where you actually deliver or fit.
Meta
- Product and room imagery does the heavy lifting. Room scenes that show the product in a believable home tend to work better than plain cut-outs, but test this for your own range.
- Catalogue ads can show the products a shopper viewed, with the right image and price.
- Lead forms can bring cheap leads, but check their quality. If lead form submissions rarely become orders, sending people to a product page with a sample or quote option may be better.
Read results the right way
Use a longer window
Home improvement decisions take weeks. Judge campaigns over a period long enough for the typical path from first click to order, not day by day. Weekly checks for problems are fine; budget decisions should rest on longer periods.
Look at cost per order and return, not cost per click
The useful numbers are:
- cost per order and cost per qualified quote
- revenue or estimated value per unit of ad spend
- the share of sample orders and quotes that become orders, by campaign
- new versus returning customers
A campaign with expensive clicks that produces orders is better than a cheap one that produces browsers.
Compare with your own records
Ad platforms each tend to credit themselves for the same order. Compare platform-reported results with your own order records. If the totals across platforms add up to much more than your actual orders, you are seeing overlap. Use your own sales data as the source of truth.
Move budget towards what sells
Once you can see orders by campaign and product, budget decisions become much simpler:
- shift spend towards campaigns and products with the best cost per order
- pause or rework campaigns that bring traffic but no orders over a reasonable period
- check stock and margins before increasing spend on a product
- keep a small share of budget for testing new audiences, products and creative
Ardealy's Ads module is built around this idea: it proposes moving budget towards what sells on Meta and Google, and nothing changes until you approve it. Whatever you use, the principle is the same. Spend follows orders, not clicks.
Common mistakes to avoid
- Optimising for landing page views or clicks because they are easy to get.
- Counting every form fill as equal, whether it is a newsletter sign-up or a kitchen quote.
- Ignoring phone and showroom orders.
- Changing campaigns every few days before the platforms have learned.
- Trusting each platform's own numbers without checking them against real sales.
Where Ardealy fits
Ardealy connects your shop, your ads and your sales so you can see which campaigns and products lead to orders, and its Ads agent suggests budget changes that you approve before anything is spent. Business analysis adds your search presence and competitors to the picture. You can book a demo to see it on your own data.
FAQ
What should a home improvement retailer optimise ads for?
Orders with their real value where possible. If there are too few orders for the platform to learn from, optimise towards quote and sample requests with estimated values, and report on orders.
How do we track orders placed by phone or in store?
Record them with the customer details you collect, then upload them regularly as offline conversions to Meta and Google, with proper consent and in line with platform policies.
How long should we wait before judging a campaign?
Long enough to cover your usual path from first click to order, which for many home improvement products is several weeks. Check for problems weekly, but make budget decisions on longer periods.
Are lead form ads worth it?
They can be, but check how many of those leads become orders. If the quality is low, sending people to a product page with a sample or quote option may work better.
Why do Meta and Google report more orders than we actually had?
Each platform credits itself for orders it influenced, so the same order can be counted more than once. Use your own sales records as the final measure.
