A post was split to a new topic: New reviews didn’t appear after 72 hours
What’s up Helga,
Thanks for your email and the help center link, I’ll read it over. For now, it’s not a big deal to me – I don’t need faster caching for my purposes.
Good luck,
-Edward
3 Likes
I’d like to weigh in on this discussion from a technical and business operations standpoint, not as a complaint, but as a structured analysis, because this thread has been going for over a year, and I think it’s time we remove the ambiguity and address the underlying assumptions clearly.
Let’s walk through the logic and math together:
1. Google Places API is not the limiting factor
You’ve mentioned “technical limitations” several times, so let’s break down the only real third-party dependency involved: the Google Places API.
- You can retrieve up to 5 reviews per Place ID per call.
- A paid Places API account includes $200 of free usage per month, which covers 40,000 review requests.
- Beyond that, the cost is $17 per 100,000 additional requests.
So, for example:
- If Elfsight serves 10,000 clients using the reviews widget…
- And you updated each of them once per day…
- That’s 10,000 API calls per day, or ~300,000 per month.
Subtract the 40,000 free tier → that’s 260,000 paid calls, costing roughly $44.20/month total. Even double it to $100/month for safe margins.
That’s 1 cent per customer per month , which makes it clear this is not about cost.
2. Infrastructure load from daily syncing is negligible
You mentioned backend strain and complexity. But review syncing is a lightweight process:
- You’re not doing full data scraping, you’re hitting a standardized endpoint with minimal payload.
- Syncs can be queued, batched, and rate-limited. A simple cron job staggered over a 24-hour window eliminates any CPU spike issues.
- Even at scale, the resources needed for daily API polling are minimal with modern cloud architecture. Systems like Firebase, Supabase, AWS Lambda, or even Redis-based caching can handle this effortlessly.
In other words: this is a solved problem. Every review app that offers daily refreshes is running the same basic logic.
3. The user impact is significant, measurable, and ongoing
From a UX and business psychology perspective:
- A widget showing a review from “4 days ago” in a service business can create false impressions of inactivity or lack of engagement.
- Many industries rely on daily social proof (restaurants, contractors, SaaS providers, etc). 72-hour cache windows sabotage the very function of the widget.
- When clients know they received a new review, and it’s not reflected on their site, they don’t blame Google, they blame us, the developers and marketers who installed Elfsight.
The longer the cache delay, the more frequently users are put in awkward conversations with their own clients. That creates brand erosion, not just for us, but for Elfsight too.
So what’s the real blocker?
If the cost is pennies per user per month, and the technical workload is minor, we’re left with one possibility: Elfsight is intentionally gating this feature as an upsell.
And if that’s the case, it would be far more respectful to say so directly.
There’s nothing wrong with monetizing value, but calling it a “technical limitation” for over a year while competitors offer 3–24 hour refresh on standard plans undermines trust more than any review delay could.
Request for a clear response
So, respectfully and constructively, I’m asking for a direct answer:
Is the decision to delay review updates beyond 24 hours a cost-saving measure, a technical limitation, or a business strategy to encourage plan upgrades and upsells?
Clarity here would actually go a long way, both in setting expectations and restoring confidence.
Thank you to the team for your time and for continuing to engage transparently. The product overall is excellent, and this is one of the last pieces that, if handled cleanly, could elevate Elfsight from good to elite in this space.
5 Likes
Cannot agree more !
2 Likes
Hello,
In my humble opinion, the update should be at least once every 24 hours.
4 Likes
This is the best remark and most impactful message so far, well done Grant Kiley! I truly hope the Elfsight team finally understands how critical this 24-hour review update and refresh is for us service providers. And that they actually start working on it because so far, it’s been over a year of promises with no real action. Every response is just another excuse, instead of fixing the most serious and urgent issue with their top-selling widget.
2 Likes
Hi Grant,
First of all, thank you for taking the time to write such a thoughtful and structured analysis. I really appreciate your clear effort to break down the assumptions and costs, and I’ll do my best to answer your questions just as thoroughly and transparently.
Let me address your points one by one:
1. Regarding the Google Places API and technical limitations
If you look closely at the official Google Places API documentation, you’ll see that the API allows you to retrieve only the five reviews Google considers most relevant. They are not sorted by date, and you can’t paginate beyond them to get the rest.
This means that if you poll the API repeatedly - for example, every hour - you will still be getting the exact same five reviews over and over until Google changes what it considers “relevant.” In other words, no matter how often you query it, you won’t get the freshest or the latest reviews, nor more than five of them.
Because of this limitation, in order to show all current reviews and maintain freshness, we use a different, specialized API. The cost of this alternative API is significantly higher - by an order of magnitude - than the baseline cost figures you provided.
Additionally, the scale of Elfsight is much larger than you estimated. We don’t serve 10,000 clients; we serve millions.
There is also an important aspect of our pricing model that may not be immediately obvious:
When you compare to other competitors, keep in mind that many of them allow creating dozens of widgets for hundreds of dollars per month. In contrast, on our Enterprise plans, customers can create hundreds of widgets for tens of dollars.
For example, we have agencies who run 400+ widgets under a single subscription for $64 per month. If we offered a 1-day cache time for everyone, the infrastructure and API costs for those users would be in the thousands of dollars monthly, while revenue would remain $64.
So while on a small scale the marginal costs per widget might look trivial, at our scale they are not measured in cents - they are measured in hundreds of thousands of dollars.
That is why we created a custom solution that allows customers who really need faster refresh to enable it at an extra cost.
May I ask: is paying a few additional dollars per month for this capability really such a prohibitive burden? I would sincerely appreciate your thoughts on this.
2. Regarding infrastructure load and backend strain
You’re absolutely right that polling per se is a solved problem in cloud infrastructure. Batching, queuing, and distributed cron jobs are standard approaches, and we use them extensively.
However, as noted above, our main constraint is not primarily CPU load - it is the combination of:
- Significantly more expensive API usage (because we can’t rely on free quota or low-cost endpoints)
- Our unique business model, where customers can create large numbers of widgets on low-cost plans
This combination means that for us, scaling daily or hourly refresh to all users is not economically sustainable under our current pricing structure.
3. Regarding user impact and business relevance
We definitely recognize that fresher reviews can improve perceived responsiveness.
That said, our experience - and a large volume of usage data - suggests that for most customers, ultra-fast refresh is not a critical factor in long-term business impact.
In fact, even in this discussion thread, many users expressed hesitation about paying one extra dollar per month for faster updates.
Typically, when a widget is first installed, users are keen to see their latest reviews show up quickly. But after that initial setup, most businesses stop monitoring the widgets daily, and the practical difference between a 1-day and a 3-day cache becomes marginal.
It is quite hard to imagine a realistic scenario where an individual review appearing on the website 2 days earlier would truly impact conversions or customer trust in a measurable way. As you noted, aggregate rating, quantity, and sentiment are the primary drivers.
That said, if a business believes the timing is critical to avoid any perception of inactivity or “brand erosion,” the option is available to pay a small premium to avoid that risk.
So what’s the real blocker?
As we have explained previously, the real blocker to universally offering a 24-hour cache is simply that:
We offer some of the lowest pricing on the market
We include the full feature set without restrictions
And in exchange, we limit the cache time as one of the only trade-offs
For those customers who genuinely need near-real-time updates, we’ve made sure there is an affordable way to enable it.
A clear response to your final question
Unless Google makes a more accessible API available (or pricing changes), we cannot reduce cache time globally for all users without fundamentally jeopardizing our sustainability as a business.
Our plan this year is to further improve how users can select and manage their preferred cache time in a more transparent and self-service way.
Thank you again for sharing your perspective so clearly. I truly appreciate your engagement and the constructive spirit of your message.
If you have any other thoughts or suggestions, please feel free to share - I’d be glad to continue the discussion.
Warm regards,
Andrei
CEO, Elfsight
16 Likes
I’ll gladly pay the $1 extra for 24hr cache for my ‘all in one’… how do I get this going asap? Thanks
3 Likes
Hi there, @Bret_Iron 
I have to say that it’s possible to reduce cache only for our Google Reviews app, not for the All-in-One Reviews app. I am really sorry 
2 Likes
Hi Andrei,
Thank you for your reply. I just wanted to mention that I’m a loyal customer and I only use the Google Reviews service on your website. I would be okay to pay an extra $1 per month to receive updated reviews on my site every 24 hours.
Please let me know how I can enable this. Feel free to contact me by email or privately so we can activate this feature for my account.
Thank you so much!
2 Likes
I think there is no reason to increase rates for this feature.
1 Like
Hi @George7
If you’d like to adjust your cache time, please reach out to our Support Team — they’ll be happy to assist you with it 
2 Likes
Hi Andrei,
Thanks for the detailed and transparent response, it’s genuinely refreshing to see this level of engagement from a CEO. I appreciate the clarity around the API constraints and the economic reasoning behind the current cache setup.
It’s great to hear there’s already a path for users who need faster refresh, and I think making that more visible will go a long way in aligning expectations.
Really appreciate you meeting the conversation head-on, it definitely builds confidence in how Elfsight operates.
6 Likes
End of October 2024? It’s now October 2025 and I do not see your implementation of this HIGHLY requested feature. Please advise of when we can expect this feature to be implemented.
1 Like
Thank you for your comment, @user29523!
First off, I’m really sorry that the current cache time doesn’t quite fit your needs. As much as we’d love to offer reduced cache times, it does come with some extra costs on our side.
We’ve been looking for ways to make it work for everyone, and we’re happy to let you know that we’ve introduced an option to reduce the cache for a small additional fee. If you’d like more details, feel free to reach out to our Support Team, they will be happy to assist.
I totally understand that this isn’t the solution you were hoping for, but we’ve shared a detailed explanation of the reasons and challenges in this post: Update Displayed Reviews Every 24 hours (Reduce Cache) - #88 by Andrei
We’re still exploring other possibilities and will definitely update this thread with any new developments.
Thank you so much for your understanding 
3 Likes
This aspect is not, in my opinion, a sufficient argument to justify the purchase. Perhaps it should be based on more concrete or strategic elements.
Cet aspect n’est pas, à mon sens, un argument suffisant pour justifier l’achat. Il faudrait peut-être se baser sur des éléments plus concrets ou stratégiques.
2 Likes
Merci pour votre commentaire, @user19590 !
Nous comprenons parfaitement qu’une option de cache réduit intégrée serait formidable. Nous avons vraiment essayé de la rendre fonctionnelle, mais malheureusement, cela engendrerait des coûts importants de notre côté.
Mais bonne nouvelle : nous pouvons vous aider à réduire le temps de mise en cache moyennant un léger supplément, sur demande, et notre équipe d’assistance se fera un plaisir de vous aider.
Au fait, pourriez-vous m’en dire plus sur ce que vous entendez par « éléments plus concrets ou stratégiques » ? J’aimerais connaître votre avis!
Thanks for your comment, @user19590!
We totally get that a built-in reduced cache option would be awesome. We really tried to make it work, but unfortunately, it would come with some significant costs on our end.
But here’s the good news – we can help reduce the cache time for a small additional fee by request, and our Support Team would be delighted to assist you with that.
By the way, could you tell me more about what you mean by ‘more concrete or strategic elements’? I’d be happy to hear your thoughts!
2 Likes
As a client I hope you succed in increasing the Caché, but if it´s possible it will be better every 12 hours
2 Likes
Hi there, @Daniel_Jose_Paredes 
Thanks a lot for the feedback!
We totally get your point and also hope the devs will find a solution to reduce cache globally.
The good news is that we’ve now added an option to manually adjust the cache time upon request! However, to make this possible, we’ve had to introduce an additional fee, as it came with some costs on our end.
If you’d like to adjust your cache time, please reach out to our Support Team — they’ll be happy to assist you 
1 Like
Please fix this or maybe make option that we can manual refresh it!!