A Pinterest save is a plan, not a sale. Count the sales it turned into

Pinterest reaches people while they plan, often many months before they hire anyone, and its reporting stops looking at 60 days. A save has no name attached. Capture Pinterest leads with a phone or email, keep them long enough for your planning cycle to finish, match them to signed contracts, and judge Pinterest on cost per paying customer.

A kitchen remodeler runs Pinterest ads from January to June. By July the dashboard shows tens of thousands of impressions, a few thousand saves, a respectable outbound click rate and a handful of tracked form submissions. The owner, comparing it with search ads that produce booked estimates every week, decides Pinterest is a brand exercise and cuts it. In October and November, four couples sign kitchen contracts and mention, when asked, that they found the company on Pinterest in the spring.

Nothing in that story is a measurement failure by Pinterest. Its reporting does what it says. The problem is that the business sells a product people plan for most of a year, and it judged the channel on a timescale built for products people buy this week. Home renovation, kitchens and bathrooms, landscaping, real estate, weddings and furniture all share that shape, and they are exactly the categories people go to Pinterest for.

Why Pinterest is a planning channel

Pinterest describes itself as the place people go to find ideas and plan projects. For a business that closes offline months later, that makes Pinterest an early-stage channel: it reaches people before they ask anybody for a quote, which is valuable and also the hardest point in a purchase to connect to revenue.

Pinterest is candid in that filing that measurement is a commercial problem for it, not just for advertisers. That candor is useful. It means the gap described in this post is not a secret or a flaw to expose; it is a known property of a channel that sits early in long decisions. The planning horizon is the important number, and for some of these categories it is long.

Those figures are a few years old, and your own customers may move faster or slower. The point survives any reasonable update: an average kitchen spends about eight months in planning, and the Pinterest phase is usually the start of it, when the household is collecting pictures and has not yet called anybody. We made the general argument in long sales cycles; Pinterest is the cleanest example of it in paid media.

What Pinterest's metrics count

Pinterest Analytics counts impressions, saves, Pin clicks that open a Pin in close-up, and outbound clicks that leave Pinterest. All of them are events on Pinterest. None of them carries the name, phone number or email of the person who acted, so none of them can be followed to a sale on its own.

A save is the metric Pinterest advertisers love most, and for planning categories it is a better signal than a click: somebody liked the walnut cabinets enough to keep them. But read it as a sales ledger would. A save is an anonymous person adding a picture to a board. It might become a contract in nine months, or it might sit on a board called "someday" forever. You cannot tell which, because you never learn who saved it.

Pinterest can tell you how many people saved, opened and clicked your Pins. It cannot tell you who they were, so engagement is a creative signal, not a revenue figure.

Pinterest's attribution windows stop at 60 days

Pinterest credits a conversion to an ad only if it happens within a window after somebody clicked or viewed it. The default reporting view is 30-day click and 1-day view, and the longest available window is 60 days. A contract signed eight months after the click falls outside every setting.

Older guides still describe a 30-day engagement window, under which a save could earn credit for a later purchase. If you inherited a report built before 2025, check which rule it used before comparing it with a current one. Our post on attribution windows explains why a conversion count without its window is an unfinished sentence.

The arithmetic for a planning category is blunt. Sixty days is about two months. Houzz's average kitchen spends 8.3 months, roughly 250 days, in planning alone. Even with the longest window Pinterest offers, a kitchen that follows the average path is credited to nothing, because by the time the contract is signed, the click is four windows in the past. The 1-day view default adds very little, and for a long purchase that is the right call, since a sale months after a scrolled-past impression is not evidence of much.

Offline conversions do not lengthen the window

Pinterest accepts offline conversion events, such as a signed contract, and reports them against ads. The windows for offline events are combinations of 1, 7, 30 and 60 days, so uploading sales helps Pinterest optimize toward people who buy but does not let it credit a sale made long after the click.

Uploading closed sales is still worth doing if you have the setup; it gives Pinterest's bidding a better idea of who becomes a customer than a form fill does, and we covered the mechanics in offline conversions. Just keep the two jobs separate. Sending sales to Pinterest feeds its optimization. Deciding what Pinterest is worth to you needs your own records, which have no window at all.

One choice inside Pinterest's reporting is worth knowing for long cycles: reporting by ad event date. It answers "what did the ads I ran in March eventually produce?" rather than "which sales happened in March?" That is the right question for a planning channel, and it is also how the calculation later in this post is laid out.

Leads with a phone or email are the bridge

The only Pinterest activity you can follow to a sale is a lead that carries a phone number or email: a lead ad submission, a form on your site reached from a Pin, or a caller who says they found you on Pinterest. Each must reach your CRM with Pinterest recorded as its source.

Pinterest lead ads are the most direct route, because the contact details arrive attached to the ad. They come with two practical constraints worth knowing before you plan a campaign around them.

The 30-day deletion is the constraint that bites. A business with a nine-month cycle that downloads leads "when we review the campaign" loses most of them before the review. The rules that make Pinterest leads usable months later:

1. Ask for both phone and email on every lead form. People change one more often than the other in a year, and a contract signed in November may carry only one of them. Two identifiers give the match two chances.
2. Move every lead into your CRM within a week, by scheduled download or by the Zapier or Salesforce integration, with "Pinterest" written in the source field. A lead that exists only in Ads Manager has a 30-day life.
3. Tag the links on your Pins so that a website form reached from Pinterest records Pinterest as its source when the visitor submits it. Your form is the only place an outbound click becomes a person.
4. Ask callers how they found you, and write the answer down. Somebody who saved your bathroom Pin in April and calls in August from memory is a Pinterest lead that no tag will ever catch.
5. Never delete or overwrite the original lead when the customer calls back later from a different channel. The first record is the evidence of where they started.

Whatever arrives, normalize the phone numbers before matching. A lead form, a CRM and an invoicing system will each write the same number differently, and over a long cycle the person also has more time to give you a different email. Lead ads on Meta raise the same issues, and the same discipline fixes them.

On Pinterest, the save is the signal and the lead is the evidence. A lead with a phone or email, kept in your own CRM long enough, is the only Pinterest record that can ever meet a signed contract.

From Pinterest spend to closed revenue, worked through

Cost per paying customer from Pinterest is the spend in a period divided by the customers whose first recorded contact was a Pinterest lead from that period. For a planning category, the sales have to be counted long after the spend: the lead file covers the campaign months, and the sales file covers the year that follows.

Here is the kitchen remodeler's half-year, with illustrative figures built to show the shape of the calculation, not taken from any account. Pinterest spend was $2,000 a month from January to June, $12,000 in total. Over those six months the business recorded 180 Pinterest leads with a phone or email: lead ad submissions, website forms reached from Pins, and callers who named Pinterest. It then matched those leads against every contract signed through the following December.

The arithmetic, step by step:

• Cost per lead: $12,000 ÷ 180 = $66.67.
• Close rate: 14 ÷ 180 = 7.8%. Low next to search, which is what a channel reaching people early in the decision should look like.
• Cost per paying customer: $12,000 ÷ 14 = $857.14. This is the figure to compare with other channels, measured the same way.
• Average contract: $448,000 ÷ 14 = $32,000.
• Revenue per dollar spent: $448,000 ÷ $12,000 = 37.3.
• What a 30-day click window could ever have seen: 1 contract of 14, or $27,500 of $448,000, which is 6.1% of the revenue. Even if every sale had been uploaded to Pinterest, the default view would have shown a channel producing one kitchen.

The other way to get this wrong is to measure too soon in your own records. Had the owner matched leads to sales in July, as in the opening story, only the contracts signed by then would have counted, and the 91-to-180-day group, the largest, would have been mostly invisible. For a planning category, wait at least as long as your typical cycle after the spend ends before judging it, and look at the lag distribution itself: a hump at three to six months is what real planning-stage work looks like.

Two rules keep the figure honest. Use one crediting rule for every channel: if a couple first filled in a Pinterest lead form in March and clicked a branded search ad in October, decide once whether the earliest recorded contact takes the credit, and apply that rule to search and Pinterest alike; our post on duplicate leads explains why consistency matters more than which rule you pick. And leave sales that match no lead in Direct / Unknown, rather than handing a share of them to Pinterest because its save count is high. We set out why in unattributed revenue.

CloseRev reads two CSV exports, your closed sales and your leads, and has no connection to Pinterest. Put each Pinterest lead in the lead file with the phone number or email it came with and "Pinterest" in the source column, and it matches closed sales to those leads on normalized phone and email. A sale that matches no lead goes to Direct / Unknown rather than to any channel.

What a pause test can and cannot tell you

A pause test compares sales with Pinterest running against sales without it, which is the only way to see what Pinterest added beyond what would have happened anyway. On a long cycle the result arrives late: a pause in spring shows up in contracts in autumn, so the test has to be read on the same timescale as the cycle.

Matching leads to sales tells you which customers came through Pinterest. It does not tell you whether they would have found you anyway: a couple who saved your Pin might have searched your name later regardless. For that you need a period, or a set of service areas, without Pinterest spend, compared with a period or area with it. Our post on incrementality tests covers how to read one.

The awkward part is time. A two-month pause of a planning channel shows almost nothing in the next two months of sales, because the sales in those months were seeded before the pause began. Read the result in Pinterest leads with a phone or email first, which move within weeks, and confirm it in signed contracts once a full cycle has passed. A test you call after six weeks on a nine-month product measures the past, not the pause.

Which businesses this applies to

Every category in which a household plans for months and then signs offline has the same gap: kitchens and bathrooms, wider remodeling, landscaping, real estate, weddings and furniture. The cycle length differs, so the waiting period before judging Pinterest differs, but the method does not.

• Kitchens and bathrooms. The longest planning stage in the home, and the clearest Pinterest fit. Expect a hump of contracts many months after the leads, as in the worked example. See kitchen and bath remodeling.
• Landscaping. Strongly seasonal: people save garden ideas in winter and sign in spring, so a lead in January that closes in April is a normal path, and a short window lands right in the gap. See landscaping.
• Real estate. A buyer saving interiors or neighborhoods may be a year from a transaction, and the lead is often a newsletter or home-value request rather than a showing. Match on the email they gave you. See real estate.
• Weddings. Vendors are booked many months before the date, by couples who plan heavily on Pinterest. Both partners may contact you from different emails, so record both and keep them on one lead.
• Furniture. Shorter for a sofa, longer for a custom dining table or a whole room. Showroom sales often close offline with a phone number at the till, which is exactly what makes them matchable. See furniture retail.

In all of them, the save count is worth watching as a trend. A sharp drop in saves on new creative tells you something real about the creative, weeks before any sale could. It should never be priced in dollars, and it should never be added to a revenue report.

The number to bring to your next Pinterest budget decision

Bring cost per paying customer from Pinterest leads, counted after a full planning cycle, beside the same figure for your other channels and the distribution of days from lead to contract. Those are numbers Pinterest cannot compute, because it never sees your contracts and stops looking at 60 days.

If the answer is "Pinterest customers cost $857 each, close at 7.8%, and sign four to six months after the first lead, against $1,100 from paid social measured the same way", the conversation is about how to get more of it. If the answer is "saves are up 40%", it is about the dashboard. Pinterest may well be one of your better channels. It is only the reporting window that makes it look like one of your worst.

Judge Pinterest on the timescale your customers plan on. Collect its leads with a phone or email, wait out the cycle, match them to signed contracts, and price it by the paying customers it produced.