Nextdoor bills impressions. Judge it by the neighbors who paid you

Nextdoor bills ads on impressions, clicks or a monthly subscription, and reports impressions, clicks and, at most, a count of conversions inside a seven-day window. None of those is a customer. Record each Nextdoor lead with a real phone or email, keep paid ads apart from free neighbor recommendations, and judge each by cost per paying customer.

A landscaping company has run a Nextdoor ad from its Business Page all season, and the owner is sure it works. New customers keep saying "I saw you on Nextdoor". Then a slow month arrives, he looks at what the ad actually cost, and asks the question the dashboard cannot answer: of the people who said Nextdoor, how many saw the ad, and how many saw a neighbor write "we use these guys, they're great" under somebody's request for a lawn service?

That is the problem with measuring Nextdoor, and it is different from the problem with Yelp. Nextdoor is two channels under one name. One is an ad product, billed like any other. The other is a word-of-mouth network where neighbors recommend businesses to each other for free, and for most local service businesses it is the bigger of the two. The ad dashboard measures only the first; customers credit both to "Nextdoor". Unless you record them separately, the ads will be paid for with the recommendations' results.

What Nextdoor sells, and what it bills

Small businesses buy Nextdoor ads two ways: simple ads from a Business Page, billed on impressions as they run or as a monthly subscription, and Nextdoor Ads Manager, which adds lead gen forms and pixel tracking. Nextdoor's filings say its advertising is sold per thousand impressions, per click or by subscription. Nothing is billed per customer.

Product names on Nextdoor change more often than on most platforms. Local Deals launched in 2019; Neighborhood Sponsorships, sold by ZIP code with Tips & Advice and Conversation Starter ads, are still promoted on Nextdoor's business site; Opportunity Alerts are new. You do not need to know the whole catalog. For each product you pay for, ask two questions: what am I billed for, and what contact detail does it give me?

Impressions and clicks measure attention, not customers

Nextdoor defines its dashboard metrics plainly: impressions count every view, reach counts unique neighbors, clicks count taps on the ad, and click-through rate is clicks divided by impressions. They describe what you are billed for. Optimizing them can multiply clicks without telling you whether a single extra customer arrived.

The arithmetic is right: if a click costs 75% less, the same budget buys 1 ÷ 0.25 = 4 times as many. It is also entirely a statement about clicks. If the fourfold clicks contain the same number of people who want a landscaper, your cost per customer has not moved; if they contain more, it has fallen. The dashboard cannot tell those apart, and nor can any optimization that aims at clicks. Notice that Verisure, the advertiser, judged the change by cost per lead and ROI — its own numbers, not Nextdoor's. That is the right instinct, and it is the one this post is about.

Conversions: a count, inside seven days

The Nextdoor Pixel and Conversion API credit a purchase, lead or other event to an ad clicked within 7 days or viewed within 1 day, and self-serve advertisers see only an aggregate count. For a local service whose jobs close weeks after the first call, that count measures form fills and bookings, not revenue.

Put those together with how a local service business actually sells. A neighbor clicks your ad on a Saturday, calls on Tuesday for a quote, gets a site visit the following week and signs a $6,000 patio job a month after the click. The pixel can record the quote request, because it happened within seven days. The signed job happened three weeks outside the click window, so as documented it would not be attributed to the ad. The dashboard then shows one conversion, with no value and no name, which you cannot find in your invoices.

None of this makes the pixel useless: it tells Nextdoor's delivery system which neighbors tend to act, which is what it is built for. It just is not a revenue report. Uploading hashed customer emails and phone numbers to an ad platform is also a decision about your customers' data, and you do not need to make it to measure Nextdoor. Your own records already hold everything required, as our post on attribution windows explains in general.

A Nextdoor conversion is an event that followed an ad within seven days of a click or one day of a view, reported to self-serve advertisers as a count. A local job that closes a month later is invisible to it, and a count can never be matched to an invoice.

Lead forms, messages and calls: the leads you can follow

Three Nextdoor routes can produce a contact you can follow to a sale: lead gen forms, which pre-fill name, phone, email, address and ZIP code; Message Me conversations; and Call Now calls. Record each with a real phone number or email and "Nextdoor Ads" as its source, and they can be matched to closed sales.

This is the friendliest lead record of any local platform we have looked at. Yelp's quote requests arrive with a masked relay address; a Nextdoor lead gen form arrives with whatever contact details the neighbor kept or typed. The same discipline as Meta lead ads applies, though: the pre-filled email is the one on the neighbor's Nextdoor account, which is often not the one they will give your invoicing software. A phone number survives that much better.

1. Make phone number a required field on every lead gen form. Two required fields, name and phone, is a short form and a matchable one.
2. Schedule a recurring lead gen report, and schedule a new one whenever you add ads to a campaign, because Nextdoor's help center says an existing report will not include them. Export it into your lead file with "Nextdoor Ads" in the source column.
3. Ask for a number in your first reply to every Message Me conversation and record it with the same source. A quote for a fence or a tree removal needs a call or a site visit anyway.
4. For Call Now calls, ask how the caller found you. The call reaches your line like any other, so the ad cannot be told apart from somebody who dialed your number off a neighbor's post. The answer goes in the CRM against the caller's number.
5. [Normalize the phone numbers](/blog/phone-normalization) before matching. Nextdoor's report, your phone system and your invoices will each write the same number differently.

Opportunity Alerts: paying for a seat in the thread

Opportunity Alerts is a monthly subscription that notifies a service business when a neighbor within 20 miles posts asking for a professional in up to three chosen categories. You receive the post, not a contact detail. The lead becomes matchable only once the neighbor replies to your comment or message with a phone number or email.

Read the pitch closely and it tells you how to measure the product. The promise is speed — be first — which means the thread is public and other businesses can answer it too. That is the economics of lead marketplaces in a neighborhood setting: you are paying for the chance to compete. It also means a subscription has no per-lead price, so the only meaningful unit is cost per paying customer: the monthly fee divided by the customers whose first contact was a reply to an alert. Give it its own source, "Nextdoor Opportunity Alerts", so it is judged on its own fee and not blended into the ads.

The free half: recommendations, Faves and the thread you were tagged in

Much of the business Nextdoor sends a local company comes from neighbors recommending it in threads, which no ad paid for. Nextdoor says organic recommendations are nearly 30% of all content on the platform. Those customers usually call your main line and look like any other referral, so record them as a separate source, never as Nextdoor Ads.

Here is the arguable part. For most trades, the free half of Nextdoor is worth more than the paid half, and it is the half most often credited to the wrong thing. A neighbor asks for a landscaper, three people name you, and the person who asked looks at your Business Page and calls. When you ask how they heard of you, they say "Nextdoor". If that answer lands in the same column as your ad leads, the ad has just been credited with a recommendation it did not buy — and if you cancel the ad, the recommendations will keep coming, which is the surest sign they were never the ad's.

The fix is the one we set out for referral revenue: ask at first contact and offer answers that separate the routes. For Nextdoor, three options cover it — "A neighbor recommended you on Nextdoor", "I saw your ad on Nextdoor" and "I found your page on Nextdoor" — recorded as "Nextdoor organic", "Nextdoor Ads" and "Nextdoor page". People will not always remember which it was. Record what they say; a customer who is unsure belongs in the page option, not in the ads.

Nextdoor is an ad product and a word-of-mouth network sharing one name. Customers say "Nextdoor" for both. Record them separately, or the ads will be paid for with the neighbors' recommendations.

From impressions to cost per paying customer

Cost per paying customer from Nextdoor Ads is a period's ad spend divided by the customers whose earliest recorded contact was a Nextdoor ad lead and who then paid. Organic Nextdoor and Opportunity Alerts stay separate sources, each with its own cost, and sales that match no lead stay in Direct / Unknown.

Here is the landscaping company's quarter. The figures are illustrative, built to show the shape of the calculation rather than taken from any account, but every step can be repeated with real exports. The Business Page ad ran on a $20 daily budget as a subscription: $20 × 30 = $600 a month, less 10%, is $540, or $1,620 for the quarter. A lead gen campaign in Ads Manager spent $1,500.

The arithmetic, step by step:

• Dashboard prices for the Business Page ad: $1,620 ÷ 210 thousand impressions = $7.71 per thousand; $1,620 ÷ 410 clicks = $3.95 a click; 410 ÷ 210,000 = a 0.20% click-through rate. Only 31 of those clicks became a call or message with a number recorded.
• Cost per lead for the lead gen form: $1,500 ÷ 64 = $23.44 as reported. Three leads had a junk or duplicate number, leaving 61 usable.
• Cost per paying customer: Business Page ad $1,620 ÷ 7 = $231.43; lead gen $1,500 ÷ 9 = $166.67; ads combined $3,120 ÷ 16 = $195.00.
• Close rates: Business Page ad 7 ÷ 31 = 22.6%; lead gen 9 ÷ 61 = 14.8%; organic 11 ÷ 22 = 50.0%.
• Revenue per dollar of ad spend: $21,500 ÷ $3,120 = 6.89.
• The same figure with the recommendations lumped in: $39,100 ÷ $3,120 = 12.53, and cost per customer falls to $3,120 ÷ 27 = $115.56. Nothing about the ads changed; the free channel was counted as theirs.

That last line is the whole argument of this post in two numbers. Credited correctly, the ads return $6.89 a dollar, which may still be good — compare it with what you get from Google Business Profile and Local Services Ads, measured the same way. Credited carelessly, they appear to return $12.53, and the owner will raise a budget on the strength of 11 customers his neighbors sent him. Note too that the lead gen form, with the most leads and the cheapest cost per lead, does not have the highest close rate: cheap leads and good leads are not the same thing, the general case we made in cost per lead.

Two rules keep it honest. Match on the customer, not the order: a neighbor who first called from a recommendation in April and filled in your lead form in June was found by the recommendation, so pick one rule, such as the earliest recorded contact takes the credit, and apply it to every source. And leave the sales that match no lead in Direct / Unknown rather than handing a share to Nextdoor because the dashboard is busy; unattributed revenue explains why spreading the unknown misleads.

CloseRev reads exported files and has no connection to Nextdoor, so it sees Nextdoor only through the leads you record. Combine your lead gen report, your Message Me and Call Now contacts and your "how did you hear" answers into one lead file, each with the customer's real phone number or email and a source column that keeps "Nextdoor Ads" apart from "Nextdoor organic". It matches them to closed sales on normalized phone and email, credits each sale to the customer's earliest lead, and reports each source exactly as written. A sale with no matching lead stays in Direct / Unknown.

What would have happened without the ads

Cost per paying customer tells you what Nextdoor's ad leads cost. It does not tell you how many of those customers the ads created, because some would have found you through a neighbor or your free page anyway. The test that answers it is a pause, measured in closed revenue.

Nextdoor offers its own measurement for larger advertisers — its Ads Manager page lists awareness lift, purchase intent and in-store foot traffic — and those studies can be useful for a brand with stores. A local service business can run a simpler one itself. Pause the Nextdoor ads for four to six weeks in a period that is not your peak, leave the Business Page and its recommendations alone, and keep recording every source exactly as before. Then compare Nextdoor Ads and Nextdoor organic customers in the paused weeks with the same weeks running ads, allowing for season.

• Organic Nextdoor holds steady and ad customers vanish without being replaced: the ads were adding customers. Keep them and judge them on the $195.
• Organic Nextdoor rises to fill the gap: some of the ad leads were neighbors who would have reached you through a recommendation. The ads' true cost per added customer is higher than the table shows.
• Total Nextdoor revenue barely moves: the recommendations were doing the work. Spend the budget on what earns recommendations — finishing jobs well and asking happy customers to Fave you.

Our post on incrementality tests covers how to read a result like that without over-reading one quiet month. For seasonal trades such as landscaping, compare against the same weeks last year rather than the weeks just before the pause.

Which half of Nextdoor to pay for

Pay for Nextdoor ads only on the strength of cost per paying customer from ad leads alone, measured against your invoices and compared with your other channels. The recommendations and Faves are not for sale, and they are probably the most valuable thing your business has on Nextdoor.

Nextdoor's documentation is honest about what it counts: impressions, clicks, messages, leads submitted and conversions in a short window. Its own pitch leans, reasonably, on trust between neighbors. The mistake is the one businesses make on their own, when every customer who says "Nextdoor" is filed under the ad that happens to be running. Record the route, keep the free half separate, and the question of whether to renew answers itself — and if the ads lose that comparison, the money is better spent earning the next recommendation.

Nextdoor can tell you how many neighbors saw, tapped and filled in a form. Only your invoices can tell you how many paid, and only a separate source column can tell you whether the ad found them or a neighbor did.