Question

Hi All, I run a B+M store in the UK which also sells online, niche furniture that isn't impulse at all. AOV is £400-500 approx. Thankspically I struggle with purchases through fb ads on cold traffic, and reasonable results with retargeting. I'm considering changing my purchases campaign on my retargeting to a messages campaign instead to chat to the potential leads directly, has anyone tried/had success with this before? Thanks :)

Answer 1

I think it's wortht testing , bcuz it will give a personal touch to your audience , so either pause retargeting with purchase and launch a messaging campaign , or split your retargeting budget between the two

Comments

Never stop a good performing campaign to test waters. Your advice is dangerous.  

---

It's a good shout to split but not pause. I like to get the audience in a message becasuse typically our products require some questions answering.  

Answer 2

Typically Prospecting campaigns that promote high value products are not performing well on fb. But if your combined performance with remarketing ads is good then I suggest to let them run because they feed your remarketing campaign. Messaging ads are not my typical cup of tea so I can't make a statement here. The few times I tried they were the worst performing ads

Comments

I found the same generally speaking. But my products are mostly bespoke and compliment very specific niches. For cold audiences I get consistent 5. 0% CTR and clicks for £0. 7 or so. If they actually turn into orders it's hard to tell. Often i'll receive an email or call saying 'i have seen your ad on facebook. ' which could turn into an order but it's almost untraceable. My thoughts on messages being 9/10 customers of ours will have to ask some questions before purchasing just from the bespoke nature of the products. Perhaps the messaging will nudge them to do just that and I can add them into the funnel from there.  

---

Current retargeting, this is actually on a link clicks objective too!  

---

Your Ctr is strong but the cpc is high too. If the sales cycle is big then I would suggest to track with custom events all the important actions on your website (form submissions, clicks to the phone, etc) and then use these audiences to 1) remarketing them 2) lookalikes at 1%

---

I will try lookalikes for sure. It's something I wanted to try for a while in fact!  

Answer 3

Your cold traffic is whah drives your re-targeting. Especially when it comes to high AOV products because it will likely take the users few more visits to make that decision to buy. As a real life example, I’m moving house at the moment and I need to buy a new bed, sofa etc (£1000+ worth of purchases) and I’ve clicked a load of different ads but haven’t committed yet. But the ads I have been served ads that have helped me draw up a wish list. To the people who buy ads for this company, they probably think they’re ads are shit, but I haven’t bought it yet. If you cut your cold traffic, there’s a really really high chance you’ll cut off your retargeting performance as well. Combine your CPA’s across campaigns.  

Comments

Totally understand it, the cold campaigns get us clicks at a good cost. 7p or so, and quite well targeted to. We also have a lot of traffic coming from our google ads (which perform great) My thoughts are, in the retarget states, should I optimise for messages to try and get my potential customers to ask the questions they will be thinking, rather than just keep pushing them back to the site

---

@doorstep it’s worth a go, I personally hate being put into messenger but that completely depends on what journey your customer wants to take. I’d maybe look at shifting your cold traffic to a view content conversion, so you feed your campaigns will a lot more data.  

---

Sounds like a plan, mind if I shoot you a message?  

---

@doorstep I don’t reply to DM’s I’m afraid unless you want to pay for my time  Post it in here

---

Okay, thanks  

Answer 4

The tirth is that with advertising you'll always need to test. If you test messenger ads might or might not works. The only opiotm si to test