Night Operators

Where they started

Night Operators sells digital night vision and thermal goggles, mostly to airsoft players and casual users. Kits run from about $180 to about $790. By June 2024 the founder was buying his own Meta ads with a coach guiding him. He ran his own email team and sold on Amazon too. The brand had been built on organic reach rather than paid acquisition.

That organic reach had built a real business. The store crossed a million dollars in lifetime revenue three weeks after we started, and almost all of that million was earned before we arrived.

The fall was not gradual. December 2023 had done $258,638 in total sales. January 2024 did $219,076. February did $118,142, close to half the business gone in a single month. Three more months down, $68,081 in May, and then $64,943 in June, the sixth month down in a row.

Paid search was barely switched on. Google ad spend for the whole of June 2024 came to $950.80. The store converted at 0.32%, and the average order was $220.82. Flat, at the bottom of the fall.

The constraint

One ad cannot be aimed. That was the constraint here: an unaimable account. Nothing in it could be turned up or turned off with any idea of what would happen next.

In June 2024 we wrote the account up, before there was an agreement to protect. That was a pre-engagement sales document, not today's $5,000 Profit Clarity Audit, which did not exist then. The Google Ads section ran to seven short lines.

Six of those lines describe the same absence: nothing built, nothing split, nothing tested. An account like that cannot be pointed at anything new. It can only keep serving whatever Google decides to send it. The seventh line is the one that mattered most. The account could not tell a customer who already knew the brand from a stranger who did not.

Someone typing "Night Operators" was counted with someone typing "night vision goggles". The account was harvesting demand the brand had already earned. Nobody could see what paid search bought that organic was not already delivering. That is a reported return flattering itself. Pooled and unaimable, the account had no place to put another dollar.

What changed

The rebuild had one purpose: make every dollar in the account traceable to something specific a person typed. One ad became nine themed ad groups. Brand search was pulled into its own campaign and relaunched on 5 July. That separated the money spent on people who already knew the brand from the money spent finding new ones. Two non-brand search campaigns went live on 8 July. They were split by what people actually type: Night Vision, Airsoft Gear, NVG Goggles, Helmet, Fusion, PRO, MAX 2.0, Paintball, Nerf Wars. Every ad now matched the query behind it. The keyword and competitor research that produced those nine themes ran first.

Text ads can only aim at words. A $180-to-$790 considered purchase gets compared on price and picture long before anyone clicks a headline, so the feed had to carry the aim too. Products were approved 16 July and shopping went live the same week. Performance Max followed once there was creative worth putting behind it.

At day ten the account showed a 7.85 ROAS, and the first thing we did was argue with it. A lot of it came through retargeting, from people already in the funnel. The next move was colder audiences, because those are the ones that scale. So the reporting got rebuilt outside the ad platforms. A weekly tracker held spend, revenue, new customers, cost per new customer, LTV, MER, store conversion rate and AOV. In-platform ROAS sat next to the Triple Whale number, because the two rarely agree.

Aiming the account only pays if the page it points at converts. Conversion work ran in parallel through Intelligems. A homepage hero test lifted conversion rate 71% in the test group on desktop, and revenue per visitor went from $5.50 to $8.58. A product page test moved the benefits above the fold and shortened the title: conversion rate rose 12.4%, revenue per visitor from $1.88 to $2.03. Trust badges, payment icons and reviews went onto the checkout in August. Prices were tested across the FUSION line and five accessories.

In September the account dipped and we spent two days inside it. The top finding was our own. The search campaigns were still on broad match, which let Google serve searches far above our price range, like people researching military grade kits. Broad match came out completely and the targeting went granular. By the end of September, every dollar in the account had a named place to sit.

What happened

June 2024 total Shopify sales: $64,943.18. August 2024: $221,828.33, up 75% on July. Both figures are read off the client's own Shopify analytics, and our own monthly tracker reproduces them to the cent.

Here is what moved underneath it. Google ad spend went $950.80 in June, $5,679.82 in July, then $19,222.97 in August. Average order value went $220.82, then $360.19, then $470.58. New customers went 270, then 298, then 437. Twenty times the search spend sounds reckless until you see where it started. In June, Google ad spend was 1.5% of this store's revenue, which is a rounding error. In August it was 8.7%. The account had gone from switched off to carrying real weight.

Now the part most case studies leave out. These are total store sales, and other things were running in the same window. The founder was scaling his own Meta ads. His email team ran 15% off promotions. The store sold on Amazon, with buy buttons sitting on its own product pages. A new, higher priced FUSION line launched, which is part of why the average order more than doubled. He was also posting organic video he described as going viral. We ran Google Ads, the shopping feed, the conversion tests and the reporting. We did not run the rest, and nothing here separates our share of the total from theirs.

August 2024 was also not a new high. This store did $258,638.75 in December 2023 and $219,076.38 in January 2024, before we arrived. August landed just above that January figure and below the December one. So this was a recovery to a level the brand had already reached, on an engine that could now be pointed and funded. That is a smaller claim than the headline sounds. It is the true one.

By September the account was doing more than buying first orders. We pointed it at the higher priced FUSION line the founder had launched. By 30 September the 90-day customer value had reached $398.27, and the returning customer rate was up 51% over thirty days. Those are store-wide numbers too, so the same caveat applies: his email and his own ads were running alongside ours.

The conversion rate was the other bottleneck, and at the time we called it the bigger one. We named it in writing on 15 August 2024: "the conversion rate is currently the biggest bottleneck in scaling even quicker." It did not break. Store conversion rate was 0.32% in June, 0.30% in August, 0.29% in September, and 0.40% in October. The individual tests won on their own terms, and the store-wide rate did not move. Our reading is that the traffic changed underneath them. By August, Google ad spend was twenty times June's, and cold search had campaigns of its own. The founder's Meta ads and organic video were pulling in cold visitors too. Colder traffic converts worse. We do not hold a traffic breakdown for the window, so that is an explanation, not a measurement. We are not claiming the flat rate as a win.

The engagement ran from late June to late October 2024, four months, and then the client moved Google Ads elsewhere. October came in at $226,307.28. We hold nothing after that, so this page cannot tell you whether the level held. The conversion rate is where the next gain sits, and it is where we would start if we picked this account up today.

"Targeting branded & non-branded keywords together"

Verify this case study

• Two Shopify screenshots sit behind the headline: June 2024 at $64,943.18 and August 2024 at $221,828.33, up 75% on July. Both are read straight off the client's own dashboard and sit on page 32 of our client-facing case studies deck. Ask and we will show it to you on the call.
• What does not exist: a recorded interview with this client. There is none, and there is no client quote anywhere on this page. Everything here is read off the account, the client's own dashboard, and our own dated documents from the engagement, all listed below.
• Our own monthly tracker for the account, running November 2023 to January 2025. It reproduces both headline figures to the cent. It also holds the December 2023 peak of $258,638.75 that we quote against ourselves on this page. The June screenshot's own comparison window is partial, 2 to 31 May, so it prints 1.8%. Against the full month of May the drop is 4.6%.
• The pre-engagement audit document is dated 27 June 2024. That is two days before the agreement was signed. The before-state went on paper before there was any result to protect.
• Two dated client reports. Both A/B test write-ups are in the 9 September 2024 one. The 90-day customer value and the returning customer figure are in the 30 September 2024 one.
• Platform-side proof that we held the account. The Google Ads account was linked to HoloGrowth's manager account. We also held Merchant Center and Search Console access.
• Ask about any of this on your Strategy Call. We will tell you exactly what is verifiable and how. The Google Ads work ended in October 2024 and we are not in touch with this client today. So we are not offering a reference on this account.