Converting Your 500-Character Ads to 1500-Character Responsive Ads

I've been managing Google Ads for years, and one of the more tedious transitions we had to deal with was moving from the old 500-character Enhanced Text Ads to the new 1500-character Responsive Search Ads. If you're sitting there with a bunch of legacy ads and you're wondering how to scale them up, here's what actually works. Google phased out the old 500-character Enhanced Text Ad format and replaced it with Responsive Search Ads that allow up to 1500 characters total. You submit multiple headlines (up to 15) and descriptions (up to 4), and Google's algorithm mixes and matches them for each auction. The shift wasn't just about character count — it was about handing control of ad composition over to machine learning instead of you. Start by pulling your account. Go into Google Ads and export your ad history or use the campaign structure view to identify every Enhanced Text Ad still sitting around. I found that a lot of people don't realize how many legacy ads they have until they look. In one account I managed, there were over 200 ETAs spread across a dozen campaigns that nobody had touched in months.

Here's the step-by-step: First, open each campaign and click on the ad group containing your 500-character ads. Delete or pause the old ETAs — don't just leave them, because Google sometimes serves both formats simultaneously and that creates reporting chaos. Then create a new Responsive Search Ad within that same ad group. Now here's where most people mess up. They think converting means copying the old ad verbatim into the new format. That doesn't work well because the new format rewards variation. Take the core messaging from your 500-character ad — the value proposition, the call to action, the primary keywords — and distribute that across multiple headlines and descriptions rather than pasting one big block of text somewhere.

For headlines, aim for at least 8 to 10 variations. Mix different angles: some focused on features, some on benefits, some with urgency, some with keywords. For descriptions, write 3 to 4 solid ones at roughly 90 characters each. Google ignores duplicates silently, so don't bother adding them just to fill slots.

Get the Full Details

The History of the Middle Ages: 500 Ad to 1500 Ad - Walmart.com
The History of the Middle Ages: 500 Ad to 1500 Ad - Walmart.com

A Problem I Ran Into

When I was converting a client's account last year, I hit a wall with their remarketing lists. The old ETAs had very specific keyword-targeted copy that performed well for narrow audiences. When I moved everything into RSA format, the machine learning started serving the same generic combinations to everyone in the audience, and CPA jumped about 40% in the first two weeks. The fix was simpler than I expected. I created separate ad groups for the remarketing audiences instead of letting the RSA rotate across all segments. I gave each ad group its own set of 10-12 headlines that were tailored to that audience's behavior, then locked the RSA to serve primarily within that context. It brought the cost back down to within 10% of the old ETA performance, which I'd consider a win given the flexibility you gain.

Common Mistakes That Wreck Performance

One thing beginners do consistently wrong is pinning headlines. You can pin up to 3 headlines to specific positions, but pinning too aggressively defeats the purpose of responsive ads. I've seen accounts where heavy pinning resulted in worse CTR than the old ETAs they replaced. Leave at least 5 headlines unpinned so Google can test combinations. Another trap is writing headlines that are too similar. If your 10 headlines all say essentially the same thing with minor word swaps, Google has nothing useful to optimize. Each headline should introduce a genuinely different angle or piece of information. Also worth noting: RSA performance data is noisy at first. The learning phase can take 2 to 4 weeks depending on your daily impression volume. Don't make changes during that window or you'll reset the optimization clock. I've lost count of how many times I've watched someone tweak an RSA after 5 days and then complain it didn't work.

How Long This Actually Takes

If you have 50 or fewer legacy ads, you can knock this out in a few hours doing it manually. More than that and you should consider scripting. I've used Google Apps Script to bulk-create RSAs from exported ad data, and it cuts the time down to roughly 20 minutes for a large account. The script reads your CSV export of old ads, maps the headline and description fields to the new format, and creates the RSAs through the Google Ads API. It's not glamorous but it works. Not every 500-character ad should become a 1500-character RSA. If an ad is already performing exceptionally well and your account has low impression volume, sometimes it's better to leave it alone. RSAs need enough data to learn properly, and a low-volume campaign might spend more time in the learning phase than it ever would have just running the ETA as-is. I'd recommend migrating only if you're getting at least a few hundred impressions per week per ad group. There's also the edge case of highly regulated industries like healthcare or finance where you need tight control over ad copy for compliance reasons. RSAs give you less direct control over exactly what combination gets shown, so if your legal team requires every ad variant to be pre-approved, you might be better off sticking with expanded text ads or using the draft-and-experiment feature to vet combinations before they go live.

Timeline of Castles: 500 AD to 1500 AD by Robert Donovan (Student) on Prezi
Timeline of Castles: 500 AD to 1500 AD by Robert Donovan (Student) on Prezi

The Bottom Line

Moving from 500 Ad To 1500 Ad is mostly about understanding that you're trading creative control for algorithmic optimization. The format change itself is straightforward — delete the old ads, create the new ones, fill in the fields. The harder part is writing enough genuinely distinct headline variations and giving the system room to actually test them. Most accounts I've seen improve after the migration, but only if people actually put effort into the variation and don't treat it like a copy-paste job.