What happens to ROI when you move Meta budget onto first-click campaigns?
August to December 2025
Why does return on spend fall as you scale Meta on reported ROAS?
Meta’s reported ROAS swung between 1.7 and 2.5, dropping in November and recovering in December. Total business ROI, which we calculate as store revenue divided by all ad spend, told a plainer story: it sat between 3.5 and 3.9 in August and September 2025, then slid from October as the account grew towards ₹1.73 crore of total spend by January 2026. Some of that is normal, because returns fall as you reach further past your best buyers. The problem was that we could not tell which campaigns were finding new buyers and which were just collecting sales, so every scaling decision was partly a guess.
January 2026
Why can Meta report a 2.4 ROAS on a campaign that finds few new buyers?
Short answer
Because reported ROAS credits a campaign for purchases made within days of someone clicking or seeing its ad, whether or not it introduced the buyer. One sale campaign was shown to each person about ten times. From February to April 2026, Meta reported its ROAS at 2.40 to 2.63, while its first-click ROAS was 0.88 to 1.10.
We added first-click ROAS to every campaign report. It credits a sale to the campaign that first brought the customer in, rather than the ads they saw later, and the gap showed up immediately. Across Meta, reported ROAS ran between 1.7 and 2.0 from January to April, while the campaigns we checked sat near 1 or below on first click.
One sale campaign, two stories
Average frequency over the campaign’s life: 10.3 views per person.
A campaign shown to the same person ten times will be near a lot of purchases, and Meta’s default click-and-view attribution rewards that. First click asked a different question, whether the campaign started the journey, and the answer for this one was mostly no. Its reported ROAS looked steady while its first-click ROAS slipped, which told us the headline number was flattering it, not that the campaign had suddenly got worse.
April 2026
How small should the first test of a new signal be?
Total business ROI hit its lowest point of the year, 2.36, on ₹2.53 crore of spend. That month we launched the first campaigns optimised for first-click purchases, at ₹13.1 lakh, or 5.2% of Meta spend. Their ROAS was 2.01. We deliberately kept them small.
Why hold first-click campaigns at a small share before scaling them?
Short answer
To see two things before moving the budget: whether the new campaigns keep their returns as they grow, and whether the whole business improves. Held at about 14% of Meta spend from May to July 2026, first-click campaigns returned ROAS of 1.78 to 1.85, and total business ROI rose from 2.49 to 2.69.
The obvious move after a good first month is to pour money in. We held the first-click share at about 14% of Meta spend (13.2% to 14.4%) for a full quarter instead, because one month proves nothing.
Returns and total business ROI both held. First-click campaign ROAS stayed at 1.85, 1.78 and 1.83. Total spend rose 34% from April to July while store revenue rose 53%, and total business ROI climbed each month: 2.49, 2.65, 2.69. From May to July, revenue grew faster than spend every month.
What should you see before shifting most of the budget?
With a quarter of evidence, we moved first-click campaigns to 31.6% of Meta spend. Their spend that month, ₹91.6 lakh, was seven times April’s, and their ROAS was 1.86, inside the same narrow band. Total business ROI reached 2.86. Total spend also dipped 14% that month while revenue fell only 8%, which helps ROI, so we read August as confirmation rather than proof.
Which campaigns should you cut once first-click data is in?
By late September, first-click campaigns carried 72.4% of Meta spend, with ROAS still at 1.86. The other half of the decision was what to cut. Two prospecting campaigns drew a large share of their credit from views rather than clicks. From December 2025 to September 2026 they spent ₹7.28 crore at a reported ROAS of 1.85 and a first-click ROAS of 0.91. From late August to late September we cut their spend by 84% and 87%. Most of the budget that increased went to first-click campaigns and to winners from creative testing.
What is the Adbuffs Step-Up Rollout?
Short answer
A three-step way to move a Meta account onto a new signal. Test the new campaigns at about 5% of spend, hold them at about 14% for a full quarter, then move the majority once their returns and total business ROI both hold. Here, first-click campaigns went from 5.2% of Meta spend in April 2026 to 72.4% by 29 September.
The order mattered more than any single setting. We call it the Adbuffs Step-Up Rollout, and it is how we now move any account onto a new optimisation signal.
The Adbuffs Step-Up Rollout
5%
Test
Launch a small share of spend on the new signal. Learn whether it delivers at all.
About 14%
Hold for a quarter
Keep the share flat for three months. Watch whether its ROAS holds as it grows and whether total business ROI moves.
Majority
Shift
Move the weight only once both agree, and cut what the new signal shows was borrowing credit.
Total business ROI, store revenue divided by all ad spend, referees every step. No platform’s own
number gets the final say.
That check is the part most teams skip. Meta’s ROAS, first-click ROAS and total business ROI measure different things, and only the last one counts money that reached the business. It is the D2C-store version of what we call True Omnichannel MER, which also counts marketplace and quick-commerce revenue, explained in our post “The one number that tells you if you’re actually growing”. We used first-click data to decide where money went, and total ROI to check whether the business improved alongside the campaigns.