Home > Facebook & Instagram > Follow-along Campaigns

---

Journey to 10 insurance leads a day [4th Day Campaign] (29)

---

Hi stm,

About the campaign.

Used client's list of 300 clients. Using 1% lookalike audience.

Client is a life insurance advisor servicing the Greater Toronto Area. My job is to generate good leads that want to buy insurance.


1 Adset - 6 Ads inside it




Messenger destination




Targeting




Placement and delivery




Results

This is the past 1 week of the campaign. Generated 12 leads. 1 confirmed call. 7 bad leads. 3 messages are still expecting a response.




Question

I should wait until 25 conversions for the algorithm to catch steam, correct?

Balance

---

looks good brov, Are you using chatbots ?

---

Hi kjrocker,

Yes I setup a basic one.




Trying this approach with the last 2 leads as it is suggested by my client.




What do you think?

---

Hi stm,

Update

4 more leads have came through since yesterday. Also Facebook has provided feedback on the ads. Pic below for reference.




Sent the message below as template to these new leads




Questions

- Should I bump up daily budget from $25 to $30?
- How much more budget should I allocate the current settings for the algorithm to understand the audience better?
- Is a higher budget required for better optimization in this case?
- Do all campaigns require an initial big budget for it to optimize?
- Is there anything I can do to tell Facebook which lead is good vs bad to manually contribute to the optimization?
- Do Facebook campaigns in general improve more and more the longer they run?
- Should I be applying any split testing techniques which I am not at this point?

Thanks

---

whats your ROI ?

---

Hi,

Rates - High intent to purchase leads for $30/pop.

ROI would depend on the cost per lead. Currently, with exactly $100 ad spend, I have been able to generate one decent lead which my client has accepted.

In the hole for $70+ at the moment.

---

If you don't make any profit, I don't see a point to raise budgets. If you want to burn more money (to fill your pixel with more conversions), duplicate your ads, test different images and text which are kinda similar to the ones you are using now. Increasing this ad set might bring you closer to break-even, after more conversions coming in and FB becomes "smarter" along the way. I still suggest to launch more campaigns (not more ad sets in current campaign) next to eachother

This way you burn money & test new creatives. Those new ones might give you way better results.
Split testing multiple landers is always good, unless advertiser this one should be way better than others.

---

Are you collecting lead info in messenger or sending people to some sort of signup page .? I think you need to optimise your collection process a bit

---

Thank you for the reply @stickupkid. I raised the budget from $25 to $42.68 over the period of a few days at a 20% to 30% scale per day. Pic below of the 2 ad sets in the campaign.




Ad set 1

I continued to run it. Soo far, I have successfully generated 3 more leads that my client would pay for. Totaling 5. I believe the algorithm is learning more. Saying that, 80% of the total leads were junk. Should I start cutting some ads? Is there a way I can tell Facebook which leads were good? Does labeling work? I have faith in this campaign.




Ad set 2

Ad set 1 was generating only men leads so I setup a female targeting camp. You recommended new campaign, I forgot and created this. Should I let this run?




These are the targeting settings + I have excluded audience targeted from ad set 1.




I will run these for some more budget. I believe I can profit of this as so far my rev is $150 and cost is $217.

I am making more creatives. From your experience, is investing in Facebook spy tools worth it for insurance ads?



Hi @kjrocker,

Thank you for your reply. People click on the messenger button to chat with my client. I have setup a quick chat type set which looks like this.




What could I be doing better? The idea is to get the lead to provide a phone number after expressing their desire to know more about insurance.

Thanks

---

Why don't you be more specific , for example asking prospects what type of life insurance they are looking for and giving them options ? And also maybe once they select an option start capturing their info ?

---

That is a good idea. I will test different chats to convert better. Would you recommend me starting a new campaign or editing the same one in the event I want to tweak messenger auto text?

---

Usually I try to create new ads so i can study data proerly but IDK what your SOPs are you can follow them but its worth a test, BTW what GEOs are you targeting ?

---

Very interesting! Generating leads for clients is one of safest business models these days when it comes to avoiding account bans.

Thanks @balance for posting in such detail! Subscribing to this one!



Amy

---

Hi @kjrocker,

I will do just that. Going to create a new duplicate campaign. How can I ensure audience overlap does not occur with the same exact ads? Would excluding people that message the client via messenger over past 30 days for the exact same targeting be sufficient?

This campaign is for Toronto, Ontario. Running 45 km radius around the city.


Hello Amy,

Thank you. Following your new Facebook guide.

---

Update

7 successful leads generated soo far that my client would accept. Rev is at $210. Cost is at $347. Yesterday I did 2 good leads so I thought,

Today was going to be around -20% or -10% ROI. But then..





Instead, the campaign spent $58.42 spent with 0 leads today. 3 messages that came were junk.

Few questions:

1. Is this normal for the Facebook algorithm to do this? I doubled the budget from around $30 to $60 over the last week. Could this have caused this?
2. Is it normal for a campaign to go through this phase of burning dollars while optimizing?
3. Is there anything I can do to help the situation?
4. Should I let it run one more day?

I made NO changes to the campaign over the past few days other than raising the budget.

Thank you

---

With lead gen you want to pay extra attention to your angles, copy and overall approach. I.e. how you're positioning the offer can make or break the entire thing.

Don't just throw stuff at the wall attempting to rely mostly on stats/data and spending your days like most affs approach things – randomly throwing images/vids and slapping LP changes is not a good idea. Instead, think about how you can directly and clearly approach people in the Greater Toronto Area with your ads making it 100% clear what this is for and what people are supposed to do.

Dig in and spend a few days purely doing heavy research especially if you haven't pushed stuff in life insurance. Look for key hooks/pulls etc.

It could be the budget but also your targeting and audience.

Of course - have your ad sets gone through the learning phase yet?

There's a number of things you can do.

Start by revising your angles/ads/creatives - we haven't seen those so can't help there but judging by the tiny thumbnails how is a lady sitting on a comfy couch relevant to life insurance? Though again, can't draw any conclusions here based on a tiny thumbnail and without seeing the entire campaign.

A few more tips:

- Try running broad instead (i.e. at most just select the Toronto area and nothing else)
- Avoid interests combined with LLAs
- Avoid having too many ads inside each ad set, especially during the optimization phase. If you wanna do this, definitely go for dynamic testing instead, far more efficient. Better bang for your buck!
- Spread out and straight up test lead ads vs leads going to an opt-in page (even better, have prequalifying questions and segment the leads) vs messenger vs anything else you can muster
- Tell the client to call the hot leads ASAP - i.e. within an hour of receiving them and to ask for feedback - e.g. "Why are the leads bad?" "Why are they not converting?" "What are their primary concerns?" "What do you think may be the issue here, what are the leads after?" - from here you can typically find fantastic insights on how to improve your ads and come up with killer angles.
- I personally prefer testing hyper-aggressively, high budgets, go straight for the kill -> gathers data faster and allows you to make quick decisions. Would rather get those 25-50 leads needed to feed the starving algos in a day or two than drag it out to over a week with the risk of making irrational/rushed decisions and resetting the learning phase, or wondering wtf's going on with my CPLs day-by-day.
- If you're not a fan of potentially risking more money, approach this the other way. Setup simple posts and boost them in whatever way (PPE etc.) and just "feel" your way through what may work based on feedback/comments/engagements alongside testing various angles/creatives at lower budgets, interacting with people and then swapping over to more heavy-hitting objectives.

This is all quite different to having a pure affiliate-numbers mindset – safely ducking behind a redirect URL and not caring for whatever happens beyond or solely focusing on the traffic problem even. I.e. Be proactive and to try and closely work with said business.

The marketing and sales techniques/rules/consumer psychology etc. never change - the traffic source/medium honestly doesn't matter at all – FB, Google, display, a blog post and whatnot. It's all insights/hacks/tricks you can gain just through testing things and applying common sense, it's not rocket science. What I mean is as soon you figure out how to position stuff the proper way in front of an audience based on their stages of awareness and knowing how people behave and what they expect on a given TS your campaign success rates spike up quite rapidly and you no longer care where you're advertising. Went off on a bit of a tangent there heheh but I hope the above helps a bit.

---

Hi @shishev,

Thank you for your detailed response. I appreciate your insights and suggestions.

We have positioned Sim as an expert advisor instead of the "get a quote" path with rates, etc. Please find images below of the 2 best performing ads for further clarity.




Ad 1 - Short image




Ad 2 - Long image




I do believe I may be lacking the direct intent behind the ad itself as you suggested. Saying that, the feedback I have from my client suggests that the leads she accepted have been good.

How would you suggest I look into finding hooks/pulls? I am sure I can improve on this ad a lot more if not change the angle completely.

Should I be using spy tools or is there a way to manually look for inspiration or what others are using?

I have decided to lower the budget from $66/day to $50/day. I am not in the position to work with a high budget. Hopefully this decrease will let the algorithm go slower and find what we need more. Since the campaign objective is to drive messenger chats, how can I tell the Facebook algorithm what type of leads to bring more of? A lot of the leads we are getting are not interested in life insurance or clicked by mistake as they have said. Maybe the ads itself is the culprit.

Last 2 days, I have spent $132 and generated one lead. 8 good leads soo far. Ad sets are active so I believe not in training anymore. Never remembered seeing it in training either.





In response to your tips, I will be setting up a small campaigns reflecting your suggestions and then post here the details prior to launching. What would be an ideal budget to start with? Is $5 or $10/day a good start per campaign?

I agree with your words on learning throughout the journey. This phase has proven to exist, without question, in every worth-while aspect of my life.

If you could give me only one answer, would you say the best way to work with the Facebook algorithm is to:

1. Test relentlessly with your own angles
2. Research relentlessly, then test only what you see working in the space.

Assuming I have the look alike audience data. What would your answer be?

Once again, thank you for your time. I will be updating this thread with the new updates shortly.

---

That right here is the key - you should play around with the angles.

The current one, per se, is not bad but it lacks intent as you said.

While this may still work great for your client, it may get the FB algos confused – they're not that clever. They need a nudge in the right direction most of the time and you achieve that by being more direct with your creatives. "Do this!" "This does exactly that" etc.

In this scenario, I would attempt to redo and re-test what's working for you know while incorporating a more direct approach.

Here's what I would do.

While having an expert is all fine and dandy and trustworthy it is not instantly clear what I get from your ad unless I'm a super-aware customer "Yes I need a life insurance expert right now, lemme get in touch with this lady ASAP." – FB's gonna have a hard time finding a steady stream of such leads as you're noticing.

The largest audience typically would be "kinda-aware", hence I would readjust the angle to be more direct - as in start answering these questions in the shoes of a consumer: What do I gain? So what that you're an expert? E.g., most of the ad is about the expert and doesn't mention much in regards to a possible client. I.e. switch over to "you", roughly - "You can get cheaper rates from a life insurance expert working in the Greater Toronto area"

Or start over with a common fear that consumers have, resolve it by positioning your client as the expert and then having a strong call to action to grab those leads. E.g. "Insurance Advice for Today and Tomorrow"...doesn't really ring much of a bell, does it? Nor does it drive you to do anything, nor does it paint any pictures in your mind. Drop it.

(Avoid mentioning "you" directly) - "People pay way too much for life insurance, in fact, many may even be getting ripped off! One client of mine was told their rates were XXX, but after a quick 15min consultation over the phone, we managed to drop that down to XX" --> "My name's Sim Gakhar and I've got 342347237423 years of experience as a life insurance expert in the Greater Toronto Area" Headline - [Click here to book a free consultation with me]... and so on, you get the point.

On top of that, start testing all sorts of creatives (perhaps different photos as well). Perhaps even going as far as having an illustration of an expert and various price tags/rates/etc. slashed over and so on and so forth.

To further add to that - why not pause the more expensive CPL campaign and use that budget to test different objectives, especially Lead ads. You can send those over with a CRM like Hubspot directly to your client etc., be sure to add phone numbers and whatever else way there may be to get in touch with your client too.

Smaller PPE camps - yep, $5-$10 a day would be fine, you'll only be feeling the waters into what may work for you and if you notice any post picking up speed "How do I get in touch?" etc. immediately change objectives targeting it (whether you wanna stick to messenger or lead ads etc.)


I would say a combination of both - research relentlessly, see what's working in the space (this way you avoid overhead costs) and then test relentlessly with your own angles based on what's working. This requires deep knowledge of your vertical/niche but it's not like you're gonna need an MBA degree to pull it off, heheh.

To put it in simple words, just outplay everybody else. Take what's working and improve upon it. With more time spent in the niche, your very own angle ideas are going to start popping up after a while. You can even mix and match angles from a different industry and apply it here, if you know your audience well. Etc.

Wrote this kind of in a rush so let me know if it makes sense.

---

Great thread with a ton of tips. Saving the answers from @shishev to re-read them later.

Good luck @balance and I hope you update the thread.

---

Hi @shishev,

Thank you for the response. I have implemented most of your suggestions in my new campaign. I am limited with my budget so I could not execute everything in this update. Also, I have implemented some ad set and ad duplication strategies suggested on this thread for hitting the best audience pools.

Before I launch, I wanted to share the campaign setup from the start.


Ad set settings (Duplicated same Ad set 3 times)


Clicks to messenger.
Budget $10/day continuous.


Audience, Location, and other settings





No additional detailed targeting. Working with just LLA as suggested. Running a bigger radius. Will split test this in the future with your suggestion of broad 'Toronto' radius.


Placements + Optimization & Delivery





I took the best performing ad from the previous campaign. Added some text, and then duplicated it 3 times to split test just CTA text.


Ad text




Post preview




CTAs used

• Book a Free Consultation
• Get Life Insurance Planning With Me
• Book a Free Consultation With Me

The client did not want to position herself aggressively. She is open to discuss which is great.


I have also duplicated each ad 3 times in each ad set as suggested or how I understood from this thread.




Messenger Setup

Start Conversations - Using same chat format currently. I would like to split test messenger related options later. As suggested, I am learning about Generate Leads optimization which will reflect in future tests.




Ready to launch this. Please let me know your thoughts. Thanks again.

Balance

---

Duplicating is no longer 100% necessary imho - it may be worth a test but still.

Is there a particular reason you've decided to stick only to messenger/conversations as an objective?

Especially regarding leads as an objective - there's not much to learn really. Just jump in and test it, even at a lower budget as you're doing now. If you're that scared, read over their documentation if it would calm you down, but then just try it would be my advice.

I think your client would be delighted to hear that "aggressive" doesn't really mean exactly that - "the word itself is scary"... it would mean far higher quality leads, maybe even cheaper while at the same time building a solid reputation.

Or some such case, if you can build it carefully haha.

Looking forward to some new stats!

---

Hi @shishev,

Here are the stats of running this campaign today. 1 Lead was junk, one is a maybe. Is it normal for Facebook Algorithm to start slow like this?




Thank you for your words of encouragement. I created another new campaign with form lead gen as the objective.





$750 total lifetime budget. I created a form. Targeting the same audience as previous campaign. [I keep using the same 1% audience again and again, is this bad?]

4 ad sets duplicated once. Minor CTA Changes in each adset. Using dynamic creatives as suggested. I have not yet started testing other wording/angles. That is next.



Here are some pictures of the new campaign for reference











Please let me know your thoughts.


Also, would it be worth investing in a spy tool for this niche? If so, any recommendations?


Balance

---

Hey,

The first screenshot - it could be due to lower budgets and not using CBO. If the expected CPL is, say, $5-$10 you would not get too many leads a day, so to speak.

1% LLAs - earlier you mentioned 80% of the leads being junk. That could very well be because of the creatives/angles but also the audiences. If there haven't been any super-drastic changes in lead costs perhaps split-test vs a broad audience.

How many leads/day does FB's estimator show with LLA vs no LLA with a decent daily budget?

---

If using dynamic creatives there's really no need to dupe the ad sets imo, especially if budgets are kept on the tight end.

---

Lead form looks decent for a first test.


It's still a bit tough to get a feel for how this may work overall - most interesting part would be when you start playing with angles and creatives (don't get emotionally attached to any creative).


Have your goal be the easiest way to test the waters without spreading yourself too thin - not too many duplicates, too many ads/ad sets or too many variations of anything at once. Focus instead on finding the right path of what generates the highest quality leads for your client - be it campaign objectives, angles/creatives (biggest impact), audiences, targeting etc.

Have you spoken to your client on what her most common type of client is? Most common objections that people have? What they look for when it comes to life insurance? - That would probably help way more than any spy tool ever will, though you never know what you're gonna find on the spy tools (I can't recommend any).

Option 2, start another STM thread asking what people in the same vertical would do if they had your client and probe for suggestions.

---

Hi @shishev,

Thanks for the response. I have applied a few things as suggested. But before that, let me start with the results from today's traffic on the first campaign I setup yesterday.








I can not comment on the quality of leads as yet. All ad sets used up 90%+ of their budget. I plan to run this 2 or 3 days more at this budget. I feel that I want to slow down with spending and see how days of the week play one at a time. What are your thoughts on this? Should I scale? If so, in what increment? Is 15% per day safe? Should I just split test this or is there a general rule?


Now the second campaign which had the new form, I generated 1 lead. This campaign budget has CBO - ON.





Because of CBO, only 1 Adset got traffic.





I have generated 1 successful lead as the info of this lead was real and I passed it on with a time booked phone appointment. I look forward to see how this runs tomorrow.


I have also made 4 additional campaigns.

3 of them are with CBO OFF variations of the CBO campaign that generated 1 lead above. 1 is duplicate of the first campaign above to test wider LLA audience of as suggested. I chose 1-2% of her list.





C1 - Lead Gen - CBO OFF is a duplicate of the campaign that generated one lead above with CBO OFF. -- $25/Day/Adset
C1 - Lead Gen - CBO OFF - Is a duplicate as well with cost control of $15 per lead with CBO OFF. -- $25/Day/Adset
C1 - Lead Gen - CBO OFF - Is a duplicate as well with cost control of $25 per lead with CBO OFF. -- $50/Day/Adset






I am looking forward to working with angles. From feedback from Sim, the leads told her that they liked her approach of telling her story and about herself in general. I have a few ideas in mind. The suggestion of 'pay too much' is working from what I see. Majority of the clients want to protect their mortgage or their kids in the event they pass away.

In regards to the path, I feel that I have not yet tested creatives, objectives, audiences, etc to have any criteria or standard in place. Messenger today seems great. But, had you not suggested different lead gen methods, I may have not explored them, robbing myself.

The new thread is a great idea. I will start it soon.

Thanks again,

Balance

---

One quick note, why do you have so many ad sets in the CBO Lifetime campaign? That's not gonna work.

Also, cost-cap you should only (imo) use when applying bigger budgets in order to test ideas (i.e. $500, $1k, $5k/day etc) so your car doesn't slide off the cliff, crash and burn – this may be constricting the algo currently.

With CBO and FB's latest updates it's generally not a good idea to be duplicating things often and too much - i.e. keep everything to a minimum and let the algo optimize. It works like a charm at least for me and friends + clients.

Give Facebook larger pools to work with, the algo is smart enough nowadays.

Instead of having many copies of the same ad, throw in 1 copy with no LLAs (broad) and see what happens over time.

In general, I'm a far bigger fan of testing everything over time - i.e. 7 day periods instead of the day by day chaos which would get you to panic and keep wondering why your camps are fluctuating so much and then pausing and changing things when you shouldn't be.

Your budgets are too low and you're not getting enough data in order to draw conclusions in regards to week parting.

Focus on nailing down great quality leads consistently, at a profit. Then worry about scaling or week parting.

Just be sure to keep your testing consistent throughout – aim to squeeze maximum insights out of your allocated budget. Otherwise, you may end up with inconclusive data.

---

shishev, I've launched a CBO slider method (Tim Burd) with a 500€ a day. I've the campaign using cost cap with 10 adsets using different bids and audiences. Do you think CBO & Cost Cap will work for new campaigns? I know the algo likes it better when there's already a ton of pixel data but do you think it's a good idea to start a fresh campaign this way?


Balance, why do you have so many ads? Have you tried Dynamic Creatives? It works pretty well but you'll need a bigger budget. I always start with 500-750€ daily budget with rules to control the losses. Might be a good idea for your campaign

---

Thank you for the response @shishev and @luis v

This is the result from today's traffic:

• From the 2% group, there has been two good leads
• C1 - Lead Gen - 1 Good lead soo far. 1 Junk.
• C1 - Lead Gen - CBO OFF - Can not comment on the lead quality as I will know when my client talks to them and provides feedback.

This is a quick update. I will post a more detailed response with a day or 2 more of traffic. I look forward to see how the camps perform tomorrow. Thanks again.

---

Hey @luis v at 500 EUR a day it may just work if you want to control your costs, but only if you know your average CPAs, otherwise you may not get traffic/results if the cost cap is lower than what FB can find.

---

Hi @shishev,

Thanks for your help on my first insurance lead gen campaign. I ran the campaign for a few days past my last update. The final results were:





6 leads were bad. The rest had intention which was good. She was satisfied. I got paid for all 17.


After reflecting on this thread, reading other threads on the forum, and looking at the final campaign stats, I concluded I got lucky generating these leads. There is so much I could have done differently. Considering my budget, advice given to me on this thread and others, and my understanding of the Facebook algorithm, I should have moved much slower and applied relevant algorithm strategies. Somewhere, my fear of not knowing what the algorithm would do made me a bit too cautious about any move. Luckily, I was encouraged by you and tested lead gen vs the messenger I was running.

From what I have learned with this campaign, I thought it would be a good idea to share with all. I will quote this thread in places where I have found answers, clarity, etc.

Q: How should I bump up daily budget?
A: Depends on the ROI. If ROI is great, duplicate with a new campaign with 100-300%+ daily budget higher. Or go slow and not burn the audience by 10% to 30% daily budget increase every 3 days or so. Reference


Q: How much more budget should I allocate for the algorithm to understand the audience better?
A: My understanding is that the algorithm needs time. You need conversion data for the algorithm to work for you. Do it slow or fast. With slow, ensure the budget is appropriate vs audience size. Example: One $10/day ad set is not enough for 1 million audience. To scale aggressively, we must get more or adequate amount of data first by starting with decent daily budget so when scaling, the new duplicate ad set/campaign performance can be estimated based on statistical relevance.


Q: Do all campaigns require an initial big budget for it to optimize?
A: Depends on the audience size. The algorithm and us both need to know enough to know what to do. Slower the better for longevity and overall profiting but this is all vs time, so subjective yet great perspective.


Q: Is there anything I can do to tell Facebook which lead is good vs bad to manually contribute to the optimization?
A: Raising the overall quality score can benefit me overall. In terms of telling Facebook, it optimizes based on CTR and content relevancy to the audience.


Q: Do Facebook campaigns in general improve more and more the longer they run?
A: Answer is yes. Saying that, the performance will eventually die as a natural process. By duplicating the top ad 2 or 3 times, changing headlines/CTA, or redoing the ad can help extract remaining value from the campaign.


Q: Should I be applying any split testing techniques?
A: $50 to $100/day minimum per ad set budget. 2 campaigns with 1 ad set each. Slightly different targeting with 3 ads inside duplicated 3 times or dynamic on a LLA 200k population. This setup to then be tested with different lead gen types. Let the algorithm pick our winner.


Notes

• Setting up a new campaign is a better to split test vs duplicating ad set as the new campaign may bring a whole new result plus allows the same testing opportunity. Thanks @stickupkid.
• Do not mix LLA with additional interest unless testing with higher % LLA.
• Messenger leads - Must have a clear conversation flow which leads to lead info capture.
• Creatives - Be clear and simple with your objective. Make CTA an easy decision. Increase CTR with content/images/testing etc.
• Algorithm needs 25 to 50 conversions. Budget is necessary for tough objectives. Can test waters with PPE objectives to see traction and then pursue further with other objectives.
• Testing multiple creatives is a must. The ad must be clear so the algorithm and audience can understand what to do.
• Ad set duplicate strategies, as used in this thread, are no longer necessary or effective. Algorithm works fine with larger pools. Use campaign duplication to test anything further.
• Cost-cap - Need big budgets and used as a safety measure.
• Prioritize generating leads at a profit with consistency above scaling.

I am starting a new thread as suggested. Thank you once again to the community.



Balance

---

Home > Facebook & Instagram > Follow-along Campaigns